RBFM v5n2.indb

Transcrição

RBFM v5n2.indb

Editorial
Revista Brasileira de Física Médica. 2011;5(2):111-3.
Homenagem à Marília Teixeira da Cruz
27 de outubro de 1937 a 25 de abril de 2011
E
ncontrei Marília pela primeira vez quando prestávamos exame vestibular oral para o curso de
Física da Faculdade de Filosofia, Ciências e Letras (FFCL)
da Universidade de São Paulo (USP), em janeiro de 1957.
Muito alegre e falante, contava em voz alta, pelos corredores do prédio da Rua Maria Antônia, o que o professor Abrahão de Moraes lhe havia perguntado. Dizia mais:
“estamos apavoradas se vamos passar e há dois rapazes
discutindo qual deles irá entrar em primeiro lugar”.
Conseguimos ingressar no curso de Física e ficamos
muito amigas desde então, estudando constantemente
juntas, Marília, Vittoria e eu. Marcávamos as provas orais
de cada matéria para os mesmos dias, de forma que pudéssemos prepará-las juntas. Estudávamos sempre na
casa da Marília, pois havia os livros do pai dela, que se
formara em Física, na mesma faculdade, em 1940, na
quinta turma. É interessante lembrar que o doutor Lauro
Monteiro da Cruz, pai de Marília, já era médico quando foi
estudar Física porque fora nomeado professor catedrático
de Biofísica da Escola Paulista de Medicina. Teve aulas
com professores italianos e franceses, os quais vieram
implementar o curso de Física na FFCL e adotaram seus
próprios livros, manuscritos, na língua original.
Marília foi sempre muito festeira, inigualável na arte de
congregar pessoas, principalmente em ocasiões especiais, pois, segundo ela, unir pessoas era muito importante. Já no início do segundo ano do curso de Física, para
dar boas-vindas aos calouros, com a ajuda da Vittoria, fez
um grande bolo: era uma réplica do Atomium, monumento
inaugurado, em Bruxelas, para a Expo 58. O monumento tinha 103 m de altura e representava a célula unitária
do cristal de ferro, ampliado 165 bilhões de vezes. Vittoria
se lembra do trabalho que tiveram para equilibrar o bolo
em forma de bolas sobre a base. Quando levaram o bolo
à Faculdade, Marília parou o carro tranquilamente sobre os
trilhos do bonde Vila Buarque, e o motorneiro esperou pacientemente o desembarque.
No terceiro ano do Curso de Física, em 1959, Marília
foi bolsista do professor Cesare Mansueto Giulio Lattes,
que viera à USP como Catedrático da Cadeira de Física
Superior. Responsável pela Disciplina de Física Superior, que
atualmente se chama Estrutura da Matéria, Lattes incumbiu
seus assistentes para montar o Laboratório: era necessário ampliar o número de equipamentos, alguns poucos dos
quais tinham sido trazidos do laboratório, então instalado na
Avenida Brigadeiro Luís Antônio. Marília montou, entre outros, o experimento de Lenard e o de Efeito fotoelétrico.
Em pesquisa, Marília trabalhou no projeto de colaboração Brasil-Japão, de 1959 até 1967, quando Lattes
transferiu-se para a Universidade Estadual de Campinas
(Unicamp).
Emico, Vittoria, Maríla e Suzana no auditório Abrahão de Moraes, em 1968.
Associação Brasileira de Física Médica®
111
Okuno E
Na Física, no churrasco do final de ano, 1977.
Em 1962, foi contratada como Instrutora Extranumerária,
junto à Cadeira de Física Superior, da FFCL da USP; em
1970, tornou-se assistente junto ao Departamento de Física
Nuclear, do Instituto de Física da USP, passando, em 1973,
à Assistente Doutora no mesmo departamento até sua aposentadoria em 1994.
No doutorado, foi orientada pelo professor Shigueo
Watanabe e defendeu a tese: “Propriedades termoluminescentes da fluorita brasileira de coloração violeta”, em 1973.
Desde então seu trabalho de pesquisa concentrou-se nas
áreas de dosimetria termoluminescente de raios X, gama e
nêutrons, proteção radiológica e Física Médica.
Após o doutorado, Marília recebeu do professor
Goldemberg uma proposta para realizar a monitoração individual dos funcionários e docentes do Instituto de Física,
que era feita pelo Instituto de Pesquisas Energéticas e
Nucleares (IPEN). Ele, diretor do Instituto de Física na época,
também providenciou o auxílio financeiro necessário para a
compra de detectores TLD-100 da Harshaw, forno e estufa
para tratamentos térmicos, fotomultiplicadora, multímetro e
componentes para a montagem de um sistema de leitura
dos dosímetros. Conseguimos um minúsculo espaço na
oficina mecânica do prédio Van de Graaff, no qual, no segundo semestre de 1977, os trabalhos para a montagem do
primeiro leitor termoluminescente se iniciaram. Após vários
testes que demonstraram o bom desempenho do sistema,
foi implantado, a partir de 1979, o Serviço de Monitoração
Individual no Laboratório de Dosimetria do Instituto de Física
da USP (IFUSP), que obteve credenciamento da Comissão
Nacional de Energia Nuclear (CNEN).
Para chegarmos a esse ponto, muito trabalho foi necessário para o aperfeiçoamento do monitor, do badge com filtro,
seleção e aferição, bem como estudo da dependência energética da resposta dos detectores. Marília encarregava-se da
irradiação dos monitores com raios X e gama de Cobalto60 e Césio-137, indo diversas vezes aos hospitais, pois não
havia no laboratório local para usar fontes de radiação. Esse
trabalho era feito à noite ou nos finais de semana, nos horários em que não havia atendimento a pacientes.
112
Marília batalhou com garra junto à Diretoria do IFUSP
para conseguir a construção de um prédio para instalar as
fontes radioativas de que o Laboratório de Dosimetria necessitava. O Prédio das Fontes, assim denominado, foi inaugurado em 1986 e nele foram instaladas, nos meses e anos seguintes, fontes de Cobalto-60, Césio-137, nêutrons e tubos
de raios X, que são até hoje utilizados pelos pesquisadores
e estudantes do grupo. Os cálculos de blindagem das salas
que compõem o prédio, prevendo a instalação dessas fontes, foram todos efetuados pela própria.
Marília publicou trabalhos de pesquisas em revistas internacionais e orientou estudantes de Mestrado e Doutorado,
além de ter ajudado vários estudantes a finalizarem suas dissertações e teses.
Por meio da Disciplina de Física das Radiações, que
Marília começou a ministrar no Instituto de Física da USP
em 1973, formou-se uma geração de Físicos Médicos que
trabalham em hospitais e clínicas, principalmente na área de
Radioterapia. Ela dedicava-se com paciência aos alunos:
fazia provas orais individuais ou em grupo e, àqueles que tinham mais dificuldade, dava sempre a chance de voltar para
nova entrevista, tantas vezes quantas fossem necessárias
para que aprendessem o básico de tudo. Esse método era
comum na época em que éramos alunas. Alguns estudantes chegaram a fazer exame na casa dela, aos sábados ou
domingos, e houve o caso de um deles que foi ao Rio de
Janeiro para fazer a prova, durante as férias de Marília.
“Marília era ‘professora’, no sentido maiúsculo da palavra: dedicava-se ao ofício, dominava o assunto que ensinava, gostava de alunos, empolgava-se com discussões,
exigia compromisso e desempenho e estava sempre disponível para esclarecer dúvidas”, segundo Martha Aldred
– uma de suas alunas.
Trabalhou com afinco para o fortalecimento da Associação
Brasileira de Física Médica (ABFM), tendo sido tesoureira por
quatro gestões, vice-presidente de 1989 a 1991 e presidente
de 1991 a 1993. Durante sua presidência, viajou pelo Brasil,
de Norte a Sul, dando palestras, incentivando físicos a ingressarem na ABFM e oferecendo ajuda a quem necessitasse.
Como mencionado, Marília adorava reunir pessoas: no
laboratório, ela instituiu que cada um levasse um bolo no
Emico, Beth e Marília em Embu, em 1981.
Homenagem à Marília Teixeira da Cruz
dia de seu aniversário. Para ela, tudo era motivo de comemoração: até hoje, quando alguém defende dissertação de
Mestrado ou tese de Doutorado no grupo é uma festança.
Na casa centenária em que Marília morava, na Rua
Bela Cintra, no quintal enorme com bananeira, goiabeira,
laranjeira, pitangueira, jabuticabeira, reuniam-se diferentes
comunidades para comemorar datas especiais. Nesses
encontros eram saboreadas iguarias sempre apreciadas,
como a feijoada preparada pela Geralda, que ainda trabalha na casa. Tenho a certeza de que todos que participaram desses eventos, inclusive colegas e amigos do exterior,
guardam lembranças saudosas.
A propósito, na época em que amadureciam as goiabas, imediatamente após colhê-las, Marília me levava especialmente. Ela sabia que eu havia crescido em goiabeiras
e um dia havia dito que o sabor da goiaba mudava para
perfume se não comesse logo. Ela até colocava saquinho
nas goiabinhas para evitar vermes.
Marília lutou contra uma doença implacável e durante
mais de dois anos recorreu a tudo que pudesse ajudá-la.
Infelizmente, veio a falecer, tendo sido enterrada no cemitério
da Consolação, em São Paulo, na tarde de 26 de abril de
2011, onde descansa em paz.
Pessoa muito séria, religiosa e generosa, de otimismo
invejável, Marília, com sua fala alta e alegre, jamais será esquecida. Deixa uma lacuna imensa e uma saudade infinita,
eterna, apesar da passagem do tempo.
Festa junina no laboratório de dosimetria, em 2003.
Agradecimentos
À Martha Aldred pelas sugestões.
Emico Okuno
Departamento de Física Nuclear do Instituto de Física da
Universidade de São Paulo. [email protected]
Na casa de Marília, em 2010.
113
Artigo Original
Optical and energy dependent response of
the alanine gel solution produced at IPEN to
clinical photons and electrons beams
Reações óptica e dependente de energia da solução
gel de Alanina produzida no IPEN para fótons clínicos
e feixes de elétrons
Cléber F. Silva1,2 and Letícia L. Campos2
1
Radiologia da Faculdade Método de São Paulo (FAMESP) – São Paulo (SP), Brazil.
Instituto de Pesquisas Energéticas e Nucleares da Comissão Nacional de Energia Nuclear (IPEN-CNEN) – São
Paulo (SP), Brazil.
2
Abstract
The DL-Alanine (C3H7NO2) is an amino acid tissue equivalent traditionally used as standard dosimetric material in EPR dosimetry. Recently, it has
been studied to be applied in gel dosimetry, considering that the addition of Alanine in the Fricke gel solution improves the production of ferric ions
radiation induced. The spectrophotometric evaluation technique can be used comparing the two spectrum wavelengths bands: 457 nm band that
corresponds to ferrous ions and 588 nm band that corresponds to ferric ions concentration to evaluate the dosimetric properties of this material.
The performance of the Alanine gel solution developed at IPEN has been firstly studied using the spectrophotometric technique aiming to apply this
material to 3D clinical doses evaluations using MRI technique. In this work, the optical and the energy dependent response of this solution submitted
to clinical photons and electrons beams were studied. Different batches of gel solutions were prepared and maintained at low temperature during
12 h to solidification. Before irradiation, the samples were maintained during 1 h at room temperature. The photons and electrons irradiations were
carried out using a Varian 2100C Medical Linear Accelerator of the Radiotherapy Department of the Hospital das Clínicas of the University of São
Paulo with absorbed doses between 1 and 40 Gy; radiation field of 10 x 10 cm2; photon energies of 6 MeV and 15 MeV; and electron with energies
between 6 and 15 MeV. The obtained results indicate that signal response dependence for clinical photons and electrons beams, to the same doses,
for Alanine gel dosimeter is better than 3.6 % (1σ), and the energy dependence response, to the same doses, is better 3% (1σ) for both beams.
These results indicate that the same calibration factor can be used and the optical response is energy independent in the studied dose range and
clinical photons and electrons beams energies.
Keywords: Alanine, dosimetry, high-energy radiotherapy, instrumentation.
Resumo
A DL-Alanina (C3H7NO2) é um tecido de aminoácido equivalente, tradicionalmente utilizado como material dosimétrico padrão em dosimetria por
EPR (ressonância paramagnética eletrônica). Recentemente, estuda-se aplicar tal material em dosimetria por gel, considerando que a adição de
Alanina na solução Frickle gel melhora a radiação induzida pela produção de íons férricos. A técnica de avaliação espectrofotométrica pode ser
usada comparando as duas bandas de comprimentos de onda do espectro: banda de 457 nm que corresponde aos íons férricos e a de 588 nm que
corresponde à concentração de íons férricos, para avaliar as propriedades dosimétricas desse material. O desempenho da solução gel de Alanina
desenvolvida no IPEN foi primeiramente estudado usando a técnica de espectrofotometria, com o objetivo de aplicar esse material em avaliações de
doses clínicas 3D usando a técnica da ressonância magnética. Neste trabalho, as reações óptica e dependente de energia de tal solução, submetida
a fótons clínicos e feixes de elétrons, foram estudadas. Diferentes lotes de soluções por gel foram preparados e mantidos em baixa temperatura
durante 12 horas para solidificação. Antes da irradiação, as amostras foram mantidas durante 1 hora em temperatura ambiente. As irradiações de
fótons e elétrons foram realizadas usando um acelerador linear médico Varian 2100 C do Departamento de Radioterapia do Hospital das Clínicas da
Universidade de São Paulo, com doses absorvidas entre 1 e 40 Gy; campo de radiação de 10 x 10 cm2; energias de fóton de 6 e 15 MeV; e elétron
com energias entre 6 e 15 MeV. Os resultados obtidos indicam que a dependência da reação do sinal por fótons clínicos e feixes de elétrons, às
mesmas doses, para o dosímetro gel de Alanina é maior do que 3,6% (1σ), e a reação de dependência de energia, às mesmas doses, é maior que
3% (1σ) para ambos os feixes. Tais resultados indicam que o mesmo fator de calibração pode ser utilizado, e a reação óptica é independe de energia
na variação da dose estudada e dos fótons clínicos e energias dos feixes de elétrons.
Palavras-chave: Alanina, dosimetria, radioterapia de alta energia, instrumentação, reação óptica.
Corresponding author: Cléber Feijó Silva – FAMESP – Av. Jabaquara, 1314 – Mirandópolis – CEP: 04046-200 – São Paulo (SP), Brasil – E-mail: [email protected]
115
Silva CF, Campos LL
Introduction
Nowadays, the three-dimensional mapping of the absorbed dose distribution in the volume of interest has become
a very important tool to check if the radiation treatment
was properly applied, considering the absorbed dose delivered to the tumor, since with a lower dose the treatment
has no effect, and a larger dose puts at risk healthy tissues
around the tumor. It is, therefore, extremely important to
create techniques that can be used to check the distribution of absorbed dose to the tumor and tissue around it.
Among these radiation dosimetry techniques, the gel dosimetry has been largely studied.
The first publication in Gel Dosimetry area was in 1984
by Gore et al1, when the Fricke solution was incorporated
into a gel matrix and this system was combined with magnetic resonance imaging (MRI) to make possible three-dimensional radiation dosimetry. Therewith, it was born the
modern gel dosimetry2. Gel dosimeters have been studied
using different compositions of the dosimetric solution and
gel materials such as organic gels or polymer gels3,4. The
High Dose Laboratory of IPEN developed a alanine gel
dosimeter based on the alanine dosimetric solution proposed by Costa5, using spectrophotometry and electronic paramagnetic resonance (EPR) evaluation techniques,
and improved by Mizuno6 with the addition of gelatin at the
dosimetric solution and using spectrophotometry as evaluation technique aiming to obtain a gel dosimeter enable
to evaluate 3D dose distribution using MRI technique. The
DL-Alanine (C3H7NO2) is an amino acid tissue equivalent
that improves the production of ferric ions radiation induced, which can be estimated through spectrophotometric
technique to measure the ferric ions concentration, aiming
to evaluate the dosimetric properties of this material.
Table 1. Chemical composition of Alanine gel solution.
Compound
Ferrous Ammonium Sulfate
Xylenol
Sulfuric Acid
DL-Alanine
Tri-distilled water
Gelatin (300 Bloom)
C (mol/L)
0.0010
0.0002
0.2375
0.6735
5.5500
10 % of the tri- distilled water volume
Acrylic support
curvettes
Figure 1. Irradiation set up to photons and electrons irradiations.
116
Materials and methods
Alanine gel solution
The dosimetric solution was prepared following the method described by Mizuno6 using 300 Bloom gelatin. The
solution was conditioned in cuvettes 1 cm x 1 cm x 4.5 cm
with optical path of 10-2 m and maintained at low temperature during 12 h to solidification. Before irradiation, the
samples were maintained during 1 h at room temperature. The chemical composition of the dosimetric system
is shown in Table 1.
Samples irradiation
The samples were always positioned on a specially designed acrylic support in a solid water RW3 phantom
that consists of 30x30x30 cm3 plates positioned on
and under the acrylic support for guaranteeing the desired depth and backscattering conditions, presented
in Figure 1.
Photon and electron irradiations
The photons and electrons irradiations were performed
using a Varian 2100 C Medical Linear Accelerator of the
Radiotherapy Department of the Hospital das Clínicas of
the University of São Paulo with doses between 1 and
40 Gy, radiation field of 10x10 cm2, photon energies of
6 and 15 MeV, electron energies of 6, 9 and 15 MeV, and
dose rate of 320 cGy/min.
Each batch was composed of 35 cuvettes filled with
gel solution, shared in 7 groups; each group was irradiated
with one different dose, except one that was not irradiated,
considered as background.
Spectrophotometric evaluation
The optical response (absorbance) was measured using a
Shimadzu UV-2101 PC spectrophotometer using the following setup parameters. See the Table 2.
Table 2. Spectrophotometer setup parameters.
Clinical Beams
Solid water plate
Acrylic support
Solid water plate
In this work, the optical and energy dependent response of this solution submitted to clinical photons and electrons beams were studied, considering that these dosimetric properties are of crucial importance for characterizing
and standardizing a dosimetric system7.
Parameters
Wavelength range (nm)
Light source
Slit width (nm)
Absorbance (%)
Transmittance (%)
Scan speed (nm/min)
Precision (nm)
400 ~ 700
Tungsten and Deuterium
2
-9.999 ~ +9.999
-999.9 ~ +999.9
1600 (fast and 2 nm interval)
0.1
Optical and energy dependent response of the alanine gel solution produced at IPEN to clinical photons and electrons beams
Each presented value is the average of 5 measures,
and the error bars are the standard deviation of the mean.
Results
Absorbed dose response
The Alanine gel dose response curves for clinical photon
(6 MeV) and electron (6 MeV) beams are showed in Figures
2 and 3 respectively.
Energy response
The Alanine gel energy response curves for clinical photons and electrons beams are showed in Figures 4 and 5
respectively.
Discussion
Dose response
In the dose range studied, between 1 and 40 Gy, the optical
response presents a linear behavior for both clinical beams.
The optical response to the same doses of the Alanine gel
solution for photons and electrons radiation is better than
3.6%, indicating that the sensitivity can be considered independent of the radiation type for the studied energies.
Energy response
The energy response of the Alanine gel solution to the
same doses is better than 3% (1σ), indicating that the optical response can be considered independent of beam
energy in the studied energy range.
1,04
Absorvance (a.u.)
2
1
Equation
y = a + b*x
R^2
0,99367
Error
Parameters Value
a
-0,04
0,02
b
0,002
0,06
0
5
10
15
20 25
Dose (Gy)
30
35
40
Relative Absorvance (a.u.)
Energy = 6 MeV
Dose Rate = 320 cGy
Equation
y = a + b*x
0,99915
R^2
Parameters Value Error
a
-0,04
0,01
0,057 0,002
b
15
8
20 25
Dose (Gy)
30
35
40
45
Figure 3. Electron dose response curve of Alanine gel solution.
10
12
Energy (MeV)
14
16
Figure 4. Photon energy response curve of Alanine gel solution.
Relative Absorvance (a.u.)
Absorvance (a.u.)
1
10
6
1,08
2
5
0,98
0,96
Energy = 6 MeV
Dose Rate = 320 cGy
0
1,00
45
Figure 2. Photon dose response curve of Alanine gel solution.
0
1,02
Mean
Dose Rate = 320 cGy
Dose = 30Gy
Mean
Dose Rate = 320 cGy
Dose = 30Gy
1,04
1,00
0,96
6
8
10
12
Energy (MeV)
14
16
Figure 5. Electron energy response curve of Alanine gel solution.
117
Silva CF, Campos LL
Conclusions
References
The obtained results indicate that it is possible to evaluate
the absorbed doses for both clinical photons and electrons
radiation beams using the same calibration curve for different energies.
The obtained results also indicate that the Alanine gel
dosimeter presents good performance and can be useful
as dosimeter in the radiotherapy area using MRI technique
for 3D dose distribution evaluation.
1.
Acknowledgment
The authors are thankful to Comissão Nacional de Energia
Nuclear (CNEN), Conselho Nacional de Desenvolvimento
Científico e Tecnológico (CNPq), Coordenação de
Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
and Faculdade Método de São Paulo (FAMESP) for the
financial support, and to the Radiotherapy Department of
the Hospital das Clínicas of the University of São Paulo for
assistance during the samples irradiation.
118
2.
3.
4.
5.
6.
7.
Gore JC, Kang YS, Schulz RJ. Measurement of irradiation dose
distributions by Nuclear Magnetic Resonance (NMR) imaging. Phys Med
Biol. 1984;29:1189-97.
Baldock C. Historical overview of the development of gel dosimetry: a
personal perspective. J Phys. 2006;56:14-22
Uusi-Simola J, Heikkinen S, Kotiluoto P, Serén T, Seppälä T, Auterinen I et
al. MAGIC polymer gel for dosimetric verification in boron neutron capture
therapy. J Appl Clin Med Phys. 2007;8:114-23.
Guillerminet C, Gschwind R, Makovicka L, Spevacek V, Soukoup
M, Novotny J. Comparative study for polyacrylamid gels and
thermoluminescent dosimeters used in external radiotherapy. Rad
Meas. 2005;39:39-42.
Costa ZM. Desenvolvimento de sistemas de DL-Alanina para dosimetria
da radiação gama e de elétrons. [Dissertação de Mestrado]. São Paulo:
Instituto de Pesquisas Energéticas e Nucleares; 1994.
Mizuno EY. Desenvolvimento e caracterização de um gel Alanina para
aplicação na medida da distribuição da dose de radiação usando a
técnica de espectrofotometria. [master’s thesis]. São Paulo: Instituto de
Pesquisas Energéticas e Nucleares; 2007.
Attix FH. Introduction to radiological physics and radiation dosimetry. New
York: John Wiley & Sons; 1986.
Artigo Original
Optimization of pediatric chest radiographic
images using optical densities ratio
Otimização de imagens radiográficas de tórax pediátrico
utilizando a razão de densidades ópticas
Rafael T. F. Souza1, Diana R. Pina2, Sérgio B. Duarte3 and José R. A. Miranda1
Instituto de Biociências de Botucatu, Universidade Estadual Paulista
“Júlio de Mesquita Filho” (UNESP) – Botucatu (SP), Brazil.
2
Departamento de Doenças Tropicais e Diagnóstico por Imagem,
Hospital das Clínicas da Faculdade de Medicina de Botucatu (UNESP) – Botucatu (SP), Brazil.
3
Centro Brasileiro de Pesquisas Físicas (CBPF/MCT) – Rio de Janeiro (RJ), Brazil.
1
Abstract
The aim of this study is the optimization of radiographic images for the pediatric patients in the age range between 0 and 1 years old, through
Optical Density Ratio (ODR), considering that pediatric patients are overexposed to radiation in the repeated attempts to obtain radiographic images
considered of good quality. The optimization of radiographic techniques was carried out with the RAP-PEPP (Realistic Analytical Phantom coupled to
homogeneous Phantom Equivalent to Pediatric Patient) phantom in two incubators and one cradle. The data show that the clinical routine radiographic
techniques generate low-quality images at up to 18.8% when evaluated by the ODRs, and increases in doses up to 60% when compared to the
optimized techniques doses.
Keywords: optimization, phantom, dosimetry, image quality, pediatrics.
Resumo
O objetivo deste estudo é a otimização das imagens radiográficas para pacientes pediátricos na faixa etária de 0 a 1 anos de idade, através da
razão de densidade óptica, considerando que pacientes pediátricos são superexpostos à radiação nas repetidas tentativas de se obter imagens
radiográficas consideradas de boa qualidade. A otimização das técnicas radiográficas foi realizada com o fantoma RAP-PEPP (fantoma analítico
realístico acoplado a um fantoma homogêneo equivalente ao paciente pediátrico), em duas incubadoras e um berço. Os dados mostram que as
técnicas radiográficas de rotina clínica criam imagens de qualidade inferior em até 18,8%, quando avaliadas por razões de densidade óptica, e
aumentam em doses de até 60% ao serem comparadas às doses das técnicas otimizadas.
Palavras-chave: otimização, fantoma, dosimetria, qualidade da imagem, pediatria.
Introduction
The process of optimization of pediatric radiographic images is part of an effective quality control program, which
must be implemented in every institution that makes use
of ionizing radiation1,2.
This process have a great importance in the radiological protection context, keeping in view that pediatric
patients are subjected to a lot of X-rays, depending on
the conditions under which they are, as weight (between
750 g and 2000 g), gestational age (between 6 and 9
months), respiratory problems and other intrinsic conditions of the newborn3.
Moreover, these patients are overexposed to radiation
in the repeated attempts to obtain radiographic images
considered of good quality3. It is important to point out that
radiation exposure during the first 10 years of life presents
a risk attributed to the lifetime (for radiation-induced biological effects, deterministic and stochastic) 3 to 4 times higher when compared with exposures between the ages of
30 and 40 years and up to 7 times higher when compared
with adults aged over 50 years.
Thus, it is extremely necessary the standardization of
radiographic techniques to conduct pediatric tests in order
to obtain good quality images with doses as low as reasonably achievable4,5.
In this context, the objective of this study is the optimization of radiographic images inserted in pediatric patients aged between 0 and 1 year old by Optical Density
Ratio (ODR), following the ALARA (As low as reasonable
Corresponding author: Rafael Toledo Fernandes de Souza – Departamento de Física e Biofísica, Instituto de Biociências de Botucatu, Universidade Estadual
Paulista “Júlio de Mesquita Filho” – Distrito de Rubião Júnior, s/n – CEP: 18608-970 – Botucatu (SP), Brasil – E-mail: [email protected]
119
Souza RTF, Pina DR, Velo AF, Alvarez M, Duarte SB, Miranda JRA
achievable), which prioritizes the acquisition of radiographic images considered of good quality with doses as low
as reasonably achievable.
Methodology
The process of validation of radiographic techniques (combinations of kVp and mAs) seeks to find the combination to
be optimum, which is able to generate radiographic images of good quality. This process was accomplished with
the RAP (Realistic Analytical Phantom) aid, that is aimed at
assessing the quality of radiographic images6.
The RAP, as illustrated in Figure 1, is a phantom consisting of realistic and analytical structures, which follows the
recommendations of ICRU (International Commissions on
Radiation Units and Measurements) Reports 44 and 487,8.
The analytical structures found in RAP are test objects that
allow the quantification of the image quality produced by
different kVp and mAs combinations.
This phantom consists of a PMMA (polymethylmethacrylate) plate with dimensions of 30cmx30cmx5 cm, in
which the test objects are inserted: (A) five acrylic steps
plus air gaps, simulating cavities; (B) five acrylic steps plus
PVC, simulating bone structures; (C) four nylon spherical
segments simulating tumours; (D) three nylon cylinders simulating fat tissue; (E) six aluminium spheres to determine
the degree of visualization of the bone tissue boundaries;
(F) four groups of human organic micro-calcification, simulating cortical bone grain with different sizes; (G) one resolution grid (0.1 mm Pb Nr 1000943 LP/mm; Nuc. Assoc.
Carle Plate, N.Y-07-538); (H) 1/2 human thoracic vertebra;
(I) two steel spheres separated by 0.80 cm for magnification analysis; and (J) one tin wire to determine the coincidence of light and radiation fields6.
The process of optimizing radiographic techniques was
carried out with the RAP-PEPP phantom, which is formed
by coupling the RAP to the PEPP (homogeneous Phantom
Equivalent to Pediatric Patient), as shown in Figure 2.
Figure 2 (A) illustrates the adaptation of the RAP to the
PEPP. Table 1 shows the PEPP structure in the AnteriorPosterior (AP) projection. In this adaptation, the lower pair
of PEPP was replaced by the RAP, which has the same
equivalence in attenuation of the part that was replaced.
The configuration of the posterior pair, referring to PEPP,
was maintained original (thicknesses of PMMA and aluminum) for the simulation of the pediatric patient chest.
Figure 2 (B) illustrates the RAP-PEPP coupled structure.
Quantitative analysis is performed using the parameter called Optical Density Ratio (ODR), which is calculated through the ratio between the optical densities of two
distinct regions and interest in the images obtained with
Figure 2. (A and B) Illustration of the RAP-PEPP structure used
to obtain the radiographic techniques based on the quality of the
acquired images.
Table 1. Thicknesses of PEPP material simulators, PMMA and
aluminum (Al) in the Anterior-Posterior projection.
Figure 1. Schematic view of structures in the RAP.
120
Top pair
PMMA (mm)
Al (mm)
35.00
0.42
Lower pair
PMMA (mm)
Al (mm)
35.00
0.83
Optimization of pediatric chest radiographic images using optical densities ratio
the RAP-PEPP. For this study, we selected values of optical density (OD) for test objects designated by (A) and (B),
represented schematically in Figure 1. The test object (A)
represents the overlap of soft tissue and bone, which consists of steps with different amounts of PMMA and PVC.
The test object (B) represents the overlap of soft tissue
and air, simulating the interior of the chest, which consists
of steps with different amounts of PMMA and air9. For this
study, the steps with the greatest amount of PVC (test
4,4
Used equipments
Incubator FANEM Vision
Incubator Isolette
Cradle FANEM AQ 50
ODR - Soft tissue and bone
4,2
4,0
3,8
3,6
3,4
3,2
3,0
2,8
2,6
2,4
2,2
2,0
Chi^2
4,20882E-4
4,69281E-4
4,63976E-4
50
R^2
0,9994
0,9992
0,9989
55
Results
60
65
70
75
80
kVp
Figure 3. Optical Density Ratio measurements between soft
tissue and bone as a function of the voltage applied to X-ray
tube (kVp) were evaluated, respectively, for the incubators FANEM Vision (black), Isolette (dark gray) and Cradle FANEM AQ
50 (light gray).
2,5
2,3
Figures 3 and 4 present the results of quantification of the
radiographic images quality by optical density ratio measurements in radiographic images with the RAP-PEPP
phantom. The ODRs were evaluated between soft tissues
and bone, and between soft tissue and air, for the incubators FANEM Vision and Isolette, and to the Cradle FANEM
AQ 50, presented, respectively, in black, dark gray and
light gray.
Table 2. Combinations between the voltage (kVp) and load
(mAs) applied to X-ray tube, absorbed entrance surface doses
of the RAP-PEPP (ESD) given in μGy, and the optical density
ratio (ODR) obtained between bone and soft tissue (Bone-soft),
and between the soft tissue and air (Air-soft), evaluated for optimized radiographic techniques in this study (Ot.) and techniques
commonly used in the HCFMB-UNESP clinical routine (Rot.).
Used equipments
Incubator FANEM Vision
Incubator Isolette
Cradle FANEM AQ 50
2,4
ODR - Soft tissue and air
object A), air (test object B) and soft tissue (between test
objects A and B) were chosen to simulate, respectively, the
bone tissue, air inside the lungs and soft tissue that exist in
the chest of a pediatric patient.
The ODRs were measured in two incubators, one manufactured by FANEM Vision and the other model manufactured by Isolette, and a cradle manufactured by FANEM
(AQ 50 model), using a x-ray equipment manufactured by
GE VMX Plus model, which has constant potential10. These
devices belong to the Hospital das Clínicas da Faculdade
de Medicina de Botucatu (UNESP-HCFMB).
It was further quantified the dose absorbed in the entrance surface. This was estimated from of the amount of
exposure measured in air, on the phantom (RAP-PEPP) in
the center of the useful beam of X-rays, using an electrometer (model 9015) and an ionization chamber (model 10X5-6), both manufactured by Radical Corporation.
Measurements were made taking into account the backscattered radiation.
The association of the radiographic image with better
quality and the absorbed dose in the entrance surface was
considered as a figure of merit (FOM).
2,2
2,1
2,0
1,9
1,8
1,7
1,6
Chi^2
2,45094E-5
3,41642E-4
2,13143E-5
50
55
Incubator
FANEMVision
R^2
0,99948
0,99979
0,99948
60
65
70
75
Ot.
Rot.
80
kVp
Figure 4. Optical Density Ratio measurements between soft
tissue and air as a function of the voltage applied to X-ray tube
(kVp) were evaluated, respectively, for the incubators FANEM Vision (black), Isolette (dark gray) and Cradle FANEM AQ 50 (light
gray).
Incubator
Isolette
Cradle
FANEMAQ 50
Ot.
Rot.
Ot.
Rot.
kVp
mAs
ESD(μGy)
55
60
66
71
50
60
57
65
50
55
60
2.50
1.25
2.50
1.00
3.20
1.00
3.20
2.00
2.00
2.50
2.00
52.71
82.57
131.6
66.83
74.01
42.17
112.2
84.32
65.25
81.56
84.35
ODR
Bone-soft Air-soft
3.05
1.95
2.70
1.84
2.41
1.73
2.19
1.69
4.00
2.30
3.20
2.05
3.22
1.93
3.05
1.84
4.28
2.34
3.95
2.17
3.78
2.02
121
Souza RTF, Pina DR, Velo AF, Alvarez M, Duarte SB, Miranda JRA
The recommended optimal radiographic techniques
were then compared with those normally applied in clinical
routine of HCFMB-UNESP. These techniques were applied
in the RAP-PEPP in order to quantify the image quality
parameter using ODR. Table 2 presents the radiographic
techniques obtained in this study and the techniques normally applied in clinical practice, which are obtained empirically and may vary by a factor of 10 between different
diagnostic services specialist in pediatric patients11.
Discussion and conclusions
The data presented in Table 2 show that radiographic techniques (combinations of kVp and mAs) used in HCFMBUNESP routine clinical generate radiographic images with
quality reduced by up to 18.8% and 13.7% when evaluated by the ODRs, respectively, between the soft tissues
and bone, and between soft tissue and air. These reductions are due to higher effective energy of x-ray beam
used, compared to the energies used in radiographic techniques optimized. It is worth emphasizing that the voltages applied to the optimized radiographic techniques have
values below those recommended by the Commission of
European Communities (between 60 and 65 kV)12.
The doses used in clinical routine show also increased by 60% when compared to the doses applied by the
optimized techniques. These last are still, mostly, below
80μGy, a value recommended by the Commission of
European Communities on the exams for children12.
The radiographic optimum techniques obtained using
the RAP-PEPP has provided the acquisition of the figures
of merit. The results of the image quality quantification
showed that the figures of merit do not have the highest
ODRs, but are obtained with voltages applied to the X-ray
tube between 50 and 60 kVp and load in a range between 1-3 mAs.
Absorbed entrance surface doses show linear behavior with the load applied to the X-ray tube and quadratic
with the voltage applied to the tube10,13. Thus, the radiographic techniques used in routine provide this increase
in absorbed dose due to the higher values of voltage and
load applied to the tube.
The optimizing of the chest radiographic images quality
performed for pediatric patients in this study is a contribution to better quality images when compared with those
obtained by clinical routine and have lower radiation doses, thus following the 3D principle (diagnostic, dose and
dollar), which prioritizes the acquisition of quality images,
122
providing a safe medical diagnosis, with radiation doses as
low as reasonable achievable, consequently encouraging
the lowest cost to the institution for the execution of radiographic examinations2.
Acknowledgment
We would like to express our gratitude to FAPESP for financial support.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Brasil. Ministério da Saúde. Secretaria de Vigilância Sanitária. Portaria
Federal n° 453, de 1º de junho de 1998. Brasília: Diário Oficial da União,
Poder Executivo; 1º de junho de 1998.
Mohamadain KE, da Rosa LA, Azevedo AC, Guebel MR, Boechat MC,
Habani F. Dose evaluation for paediatric chest x-ray examinations in
Brazil and Sudan: low doses and reliable examinations can be achieved in
developing countries. Phys Med Biol. 2004;49(6):1017-31.
Armpilia CI, Fife IA, Croasdale PL. Radiation dose quantities and risk in
neonates in a special care baby unit. Br J Radiol. 2002;75(895):590–5.
Punwani S, Zhang J, Davies W, Greenhalgh R, Humphries P. Paediatric
CT: the effects of increasing image noise on pulmonary nodule detection.
Pediatr Radiol. 2008;38(2):192-201.
Vano E, Ubeda C, Leyton F, Miranda P. Radiation dose and image quality for
paediatric interventional cardiology. Phys Med Biol. 2008;53(15):4049–62.
Pina DR, Duarte SB, Ghilardi Netto T, Trad CS, Brochi MA, de Oliveira
SC. Optimization of standard patient radiographic images for chest,
skull and pelvis exams in conventional X-ray equipment. Phys Med Biol.
2004;49(14):N215-26.
International Commission on Radiation Units and Measurements. Tissue
Substitutes in Radiation Dosimetry and Measurement. ICRU Report
44. Bethesda, MD: International Commission on Radiation Units and
Measurements; 1989.
International Commission on Radiation Units and Measurements. Phantoms
and Computational Models in Therapy, Diagnosis and Protection. ICRU
Report 48. Bethesda, MD: International Commission on Radiation Units
and Measurements; 1992.
Pina DR, Duarte SB, Ghilardi Netto T, Morceli J, Carbi EDO, Souza RTF, et
al. Controle de qualidade e dosimetria em equipamentos de tomografia
computadorizada. Radiol Bras. 2009;42(3):171-7.
Curry TS, Dowdey JE, Murry RC. Christensen’s physics of diagnostic
radiology. Philadelphia: Lee & Febiger; 1990.
Lima AA, Carvalho ACP, Azevedo ACP. Avaliação dos padrões de dose em
radiologia pediátrica. Radiol Bras. 2004;37(4):279-82.
Commission of the European Communities. European guidelines and
quality criteria for diagnostic radiographic images. Report EUR 16260EN.
Bruxelas: European Communities/Union, 1996.
Johns HE, Cunningham JR. The Physics of Radiology. Illinois: Charles C
Thomas Publisher; 1983.
Artigo Original
Using a tandem ionization chamber for
quality control of X-ray beams
Utilizando uma câmara de ionização tandem para
controle de qualidade dos feixes de raios X
Maíra T. Yoshizumi and Linda V. E. Caldas
Instituto de Pesquisas Energéticas e Nucleares, Comissão Nacional de Energia Nuclear (IPEN-CNEN/SP) –
São Paulo (SP), Brazil.
Abstract
X-ray beam qualities are defined by both the mean energies and by the half-value layers (HVL). Many international protocols use the half-value layer
and the beam voltage to characterize the X-ray beam quality. A quality control program for X-ray equipment includes the constancy check of beam
qualities, i.e., the periodical verification of the half-value layer, which can be a time consumable procedure. A tandem ionization chamber, developed
at Instituto de Pesquisas Energéticas e Nucleares, was used to determine the HVL and its constancy for five radiotherapy standard beam qualities.
This ionization chamber is composed by two sensitive volumes with inner electrodes made of different materials: aluminum and graphite. The beam
quality constancy check test was performed during two months and the maximum variation obtained was 1.24% for the radiation beam quality T-10.
This result is very satisfactory according to national recommendations.
Keywords: tandem system, X-ray beams, quality control.
Resumo
As qualidades do feixe dos raios X são definidas tanto pelas energias médias quanto pelas camadas semirredutoras. Muitos protocolos internacionais
utilizam a camada semirredutora e a tensão do feixe para caracterizar a qualidade do feixe dos raios X. Um programa de controle de qualidade para
equipamentos de raios X inclui a verificação constante das qualidades do feixe, ou seja, a verificação periódica da camada semirredutora, que pode
ser um procedimento que consome tempo. Uma câmara de ionização tandem, desenvolvida no Instituto de Pesquisas Energéticas e Nucleares, foi
utilizada para determinar a camada semirredutora e sua constância para cinco qualidades do feixe-padrão de radioterapia. Essa câmara de ionização
é composta por dois volumes sensíveis com eletrodos internos feitos de materiais diferentes: alumínio e grafite. O teste de verificação da constância
da qualidade do feixe foi realizado durante dois meses, e a variação máxima obtida foi de 1,24% para a qualidade T-10 do feixe de radiação. Esse
resultado é muito satisfatório, de acordo com as recomendações nacionais.
Palavras-chave: sistema tandem, feixes de raios X, controle de qualidade.
Introduction
When using ionizing radiation, mainly for medical purposes, the delivered dose must be very accurately known.
One way to achieve the recommended limits of dose accuracy is maintaining a quality control program of radiation beams.
One of the recommended procedures in a quality
control program for X-radiation equipment is the verification of the beam qualities constancy in terms of the
half-value layer (HVL)1. The HVL of each beam quality is,
conventionally, obtained by adding high purity absorbers
of known thickness at midway between the radiation tube
and the detector. The use of a tandem system is an alternative for the HVL verification. This method provides
an accurate response and it is easier and faster than the
conventional one2.
A tandem system is composed by two dosimeters
with different characteristics. Thermoluminescent dosimeters in tandem system were studied to determine the
effective energy of unknown radiation fields3-5.
At Instituto de Pesquisas Energéticas e Nucleares, a
double-faced ionization chamber was developed to be
used as a tandem system6. The main objective of this tandem chamber is to verify the HVL value constancy in a
quality control program.
In this work, the tandem ionization chamber was
used to establish a methodology for checking the
beam quality constancy for radiotherapy standard
X-ray beams.
Corresponding author: Maíra Tiemi Yoshizumi – Instituto de Pesquisas Energéticas e Nucleares – Av. Prof. Lineu Prestes, 2242 – CEP: 05508-000 – São
Paulo (SP), Brasil – E-mail: [email protected]
123
Yoshizumi MT, LVEC
Face A and 0.30% for Face G. Thus, both cases are within
the recommended value of ±0.5%8.
Energy dependence for X-rays and tandem curve
Both faces of the tandem ionization chamber were calibrated against the secondary standard ionization chamber
using the radiotherapy beam qualities described in Table 1.
The reference system for these radiation beam qualities is
a parallel-plate ionization chamber, PTW, model M23344.
1.008
Face A
1.006
1.004
Relative value
A special double-faced ionization chamber, developed at
IPEN, was utilized. This ionization chamber is disc-shaped
and contains two sensitive volumes of 0.6 cm3 each one.
Because the inner electrodes are made of different materials - one is made of aluminum (Face A) and the other
of graphite (Face G), this ionization chamber constitutes a
tandem system. This tandem chamber was developed to
be used in low-energy X-ray beams.
A 90Sr+90Y check source, Physikalisch-Technische
Werkstätten (PTW), model 8921, with nominal activity of
33.3 MBq (1998), was used to perform the response stability tests of the tandem ionization chamber.
An X-ray unit, Pantak/Seifert, model ISOVOLT 160HS,
with standardized beam qualities was used in this work.
The radiation qualities (suggested by BIPM7) characteristics are presented in Table 1. A secondary standard planeparallel ionization chamber, PTW, model M23344, was
used as a reference system for these beam qualities.
1.002
1.000
0.998
0.996
0.994
0.992
Results
0
5
Table1. Radiotherapy standard beam qualities of the Pantak/
Seifert X-ray equipment.
Radiation
quality
T-10
T-25
T-30
T-50 (a)
T-50 (b)
124
Tube
potential
(kV)
10
25
30
50
50
Additional
filtration
(mmAl)
--0.4
0.2
4.0
1.0
Half-value
layer
(mmAl)
0.043
0.279
0.185
2.411
1.079
Air kerma
rate
(mGy.s-1)
3.229±0.003
2.753±0.002
9.492±0.005
0.833±0.002
3.878±0.003
15
20
Measurement
The tandem ionization chamber was submitted to response
stability tests and energy dependence for X-radiation.
1.008
Face G
1.006
1.004
Relative value
Response stability tests
The response stability of the tandem ionization chamber
was tested for both sides of the ionization chamber (Face
A and Face G).
The leakage current was measured during 20 minutes,
before and after each irradiation, and it was always negligible during the whole test period.
For the short-term stability test, ten consecutive measurements were taken using the 90Sr+90Y check source.
The standard deviation of the set of measurements presented a maximum value of 0.14% for Face A and 0.05%
for Face G. For both faces, the standard deviation is within
the recommended of 0.3%8.
Comparing the mean values of each set of ten measurements, the medium-term stability test was evaluated.
The short-term stability test was performed 10 times and,
as can be seen in Figure 1, the maximum variation obtained in the medium-term stability test was 0.27% for
10
1.002
1.000
0.998
0.996
0.994
0.992
0
5
10
15
20
Measurement
Figure 1. Medium-term stability test for Face A and Face G of
the tandem ionization chamber.
Table 2. Calibration coefficients and correction factors of the
tandem ionization chamber using X-radiation, radiotherapy
beam qualities.
Radiation
quality
T-10
T-25
T-30
T-50 (a)
T-50 (b)
Calibration coefficients
(mGy/nC)
Face A
30.88±0.16
22.26±0.13
23.92±0.11
16.83±0.19
17.77±0.11
Face G
35.75±0.17
38.98±0.17
37.89±0.15
43.95±0.32
41.47±0.17
Correction factors
Face A
1.291±0.008
0.930±0.007
1.000±0.006
0.703±0.008
0.743±0.006
Face G
0.944±0.006
1.029±0.006
1.000±0.006
1.159±0.009
1.094±0.006
Using a tandem ionization chamber for quality control of X-ray beams
The calibration coefficients obtained for each radiation
beam quality were normalized to the T-30 beam quality,
obtaining correction factors. In Table 2, the calibration coefficients and correction factors are shown. The energy dependence of the tandem ionization chamber can be seen
in Figure 2.
As the energy dependence of Faces A and G are different, a tandem curve can be obtained by the ratio of the
responses. For each beam quality, the chamber response
obtained for Face G was divided by the response obtained
for Face A. The tandem curve presented a good curvature
and it is shown in Figure 3.
Constancy check of X-radiation beam qualities
The ratio of the responses was used to determine the
constancy of the beam qualities. According to national
recommendations, the maximum variation of the HVL
is ±3%9. Nine measurements were performed using the
five beam qualities studied, and the maximum variation
1.4
obtained was 1.24%, thus within the recommended value. Figure 4 shows the constancy of beam quality T-30
using the ratio of the responses. The same behavior was
obtained for the other beam qualities.
Conclusions
Both sides of the tandem ionization chamber presented
good stability response, within internationally recommended limits. The ratio of responses gave rise to a good
tandem curve, enabling the ionization chamber to be
used in radiotherapy beam quality constancy checks. The
results of nine measurements showed a stable response
of the tandem chamber for X-ray beams. The simplicity
in the use of this tandem ionization chamber in the verification of HVL values and its satisfactory results showed
that this ionization chamber is a good alternative for the
conventional method.
4
Face A
Face G
1.3
2
Variation (%)
Correction factor
1.2
1.1
1.0
0.9
1
0
-1
0.8
-2
0.7
-3
0.6
T-30
3
0.0
0.5
1.0
1.5
2.0
-4
2.5
0
4
6
8
10
Measurement
Half-value layer (mmAI)
Figure 2. Energy dependence of the tandem ionization chamber
for both Faces A and G.
Figure 4. Constancy check of the beam quality T-30.
Acknowledgments
2.8
2.6
Ratio of the responses
2
The authors would like to thank Conselho Nacional
de Desenvolvimento Científico e Tecnológico (CNPq),
Coordenação de Aperfeiçoamento de Pessoal de Nível
Superior (CAPES), Fundação de Amparo à Pesquisa
do Estado de São Paulo (FAPESP) and MCT (INCT for
Radiation Metrology in Medicine), Brazil, for partial financial
support.
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
References
0.0
0.5
1.0
1.5
2.0
2.5
1.
Half-value layer (mmAI)
Figure 3. Tandem curve of the ionization chamber.
2.
International Atomic Energy Agency - IAEA. Calibration of dosimeters used
in radiotherapy. Technical Reports Series nº 374. Vienna; 1994.
Maia AF, Caldas LVE. A simple method for evaluation of half-value layer
variation in CT equipment. Phys Med Biol. 2006;51(6):1595-601.
125
Yoshizumi MT, LVEC
3.
4.
5.
6.
126
Gorbics SG, Attix FH. LiF and CaF2:Mn thermoluminescent dosimeters in
tandem. Int J Appl Radiat Isot. 1968;19(2):81-9.
Da Rosa LAR, Nette HP. Thermoluminescent dosemeters for exposure
assessment in gamma or X-radiation field with unknown spectral
distribution. Int J Appl Radiat Isot. 1988;39(3):191-7.
Miljanic S, Vekic B, Martincic R. Determination of X-ray effective energy
and absorbed dose using CaF2:Mn and LiF:Mg,Ti thermoluminescent
dosemeters. Radiat Prot Dosim. 1999;85:381-4.
Costa AM, Caldas LVE. Response characteristics of a tandem ionization
7.
8.
9.
chamber in standard X-ray beams. Appl Radiat Isot. 2003;58(4):495-500.
Bureau International des Poids et Mesures BIPM. Measuring condition
used for the calibration of ionization chambers at the BIPM. Rapport
BIPM-04/17. Sèvres; 2004.
International Electrotechnical Commission - IEC. Medical electrical
equipment - Dosimeters with ionization chambers as used in radiotherapy.
IEC 60731, Genève; 1997.
Instituto Nacional do Câncer - INCA. TEC DOC-1151: Aspectos físicos da
garantia da qualidade em radioterapia. Rio de Janeiro: INCA; 2000.
Artigo Original
Influence of elemental weight of human
tissues estimated by ICCT software in
absorbed dose calculation
Influência do peso elementar dos tecidos humanos
estimados pelo software ICCT no cálculo da dose absorvida
Felipe Massicano1, Rafael G. Possani1, Felipe B. Cintra, Adriana V. F. Massicano2
and Hélio Yoriyaz1
Centro de Engenharia Nuclear (CEN) do Instituto de Pesquisas Energéticas e Nucleares (IPEN-CNEN/SP) – São Paulo (SP), Brazil.
2
Diretoria de Radiofarmácia (DIRF) do IPEN-CNEN/SP – São Paulo (SP), Brazil.
1
Abstract
Therapeutic use of radiopharmaceuticals in Nuclear Medicine has been well established and presented good success rates against many forms of
cancer. The biologic effects of radionuclide therapy are measured via a physical quantity, the absorbed dose, which is defined as per unit mass of
tissue. Therefore, it is of great important an accurate dosimetry to assess the potential effects of treatment and to confirm or contradict the treatment
predictions. The most common method used to estimate the absorbed dose at organ level was developed by Medical Internal Radiation Dose (MIRD)
Committee, called MIRD system. However, this method does not have adequate patient data to obtain a dose estimate accurate in therapy. In recent
years, internal radionuclide radiation dosimetry system evaluated spatial dose distribution. This system is based in Monte Carlo radiation transport
codes with anatomical and functional information of the patient. The high accuracy is, at least in part, due to the Monte Carlo method allows human
tissues to be characterized by elemental composition and mass density. Thus, a reliable estimation of human tissues (elemental composition and
mass density) must be obtained. According to Schneider, Bortfield and Schlegel, the tissue parameters (mass densities (ρ) and elemental weights
(ωi)) can be obtained using Hounsfield units provided from Computed Tomography (CT) images. Based on this, the Nuclear Engineer Center of IPEN
developed the ICCT software (Image Converter Computed Tomography). It converts CT images in tissue parameters (mass densities (ρ) and elemental
weights (ωi)). This work intended to verify if the estimate values by software ICCT of the tissue parameter and elemental weights (ωi) are plausible to
estimate the absorbed dose with reasonable accuracy.
Keywords: nuclear medicine, tissue, dosimetry, radionuclide.
Resumo
O uso de radiofármacos na Medicina Nuclear vem se estabelecendo como terapia contra diversos tipos de câncer, apresentando boas taxas de sucesso.
Os efeitos biológicos da terapia radionuclídica são medidos por intermédio de uma quantidade física, a dose absorvida, que é a dose absorvida pelo
tecido divida pela massa desse tecido. Portanto, é de grande importância uma dosimetria precisa para avaliar os potenciais efeitos do tratamento e
confirmar ou contradizer os prognósticos do tratamento. O método mais comumente utilizado para estimar a dose absorvida num órgão foi desenvolvido
pelo Comitê Médico de Dose da Radiação Interna (MIRD), chamado de sistema MIRD. Entretanto, esse método não leva em consideração dados
importantes do paciente para assim obter uma boa estimativa da dose para a terapia. Atualmente, há sistemas de dosimetria radionuclídica que avaliam
a distribuição espacial da dose no interior do paciente. Tais sistemas são baseados em códigos de Monte Carlo para transporte de radiação juntamente
com informações anatômicas e funcionais do paciente. A alta exatidão deve-se, ao menos em parte, ao código de Monte Carlo permitir que os tecidos
humanos sejam melhormente caracterizados pela composição elementar e pela densidade em massa. Desse modo, uma precisa estimativa dos tecidos
humanos (composição elementar e densidade em massa) deve ser obtida. De acordo com Schneider, Bortfield e Schlegel, os parâmetros do tecido
(densidades em massa (ρ) e pesos elementares (ωi)) podem ser obtidos usando as unidades de Hounsfield fornecidas nas imagens de tomografia
computadorizada. Com base nisso, o Centro de Engenharia Nuclear do IPEN desenvolveu o software ICCT (Image Converter Computed Tomography –
Conversor de imagem de tomografia computadorizada). Ele converte as imagens de tomografia computadorizada em parâmetros do tecido (densidades
em massa (ρ) e pesos elementares (ωi)). Este trabalho teve o propósito de verificar se os valores estimados dos pesos elementares (ωi) mediante o
software ICCT são plausíveis para estimar a dose absorvida com uma razoável exatidão.
Palavras-chave: medicina nuclear, tecidos, dosimetria, radionuclídeo.
Corresponding author: Adriana Felipe Massicano – Instituto de Pesquisas Energéticas e Nucleares (IPEN-CNEN/SP) – Av. Prof. Lineu Prestes, 2242 –
Cidade Universitária – Butantã – CEP: 05508-000 – São Paulo (SP), Brasil – E-mail: [email protected]
127
Massicano F, Possani RG, Cintra FB, Massicano AVF, Yoriyaz H
128
Introduction
Therapeutic use of radiopharmaceuticals in Nuclear
Medicine has been well established and presented good
success rates against many forms of cancer. The goal of
this treatment is to deliver a lethal radiation dose to the
tumor while avoiding or limiting the dose to critical organs1.
Therefore, an accurate dosimetry to assess the potential
effects of treatment and to confirm or contradict the treatment predictions is of great importancy2.
To determine the average absorbed dose at organ
level, the formalism developed by the Medical Internal
Radiation Dose (MIRD) Committee is widely considered as the reference method3. To support the calculation of nonuniform absorbed doses and to account for
nonuniform activity distributions at the level of imaging
instrumentation voxels, the MIRD Committee has also
published S value tabulations for different voxel sizes
and source-target voxel distances. For the reason that
the use of previously tabulated S values requires a fixed
anatomic model, this approach is not easily amenable to
geometries that differ substantially from the fixed anatomic models4.
In recent years, internal radionuclide radiation dosimetry system evaluated spatial dose distribution. This system
is based in Monte Carlo radiation transport codes together
with anatomical and functional information of the patient5,6.
Anatomical information can be obtained from medical images, e.g. with MRI or CT, expressed in 3 dimensions (3D)
in voxel format. Similarly, SPECT and PET imaging systems can provide 3D representation of activity distributions
within patients (functional information). It is called patientspecific dosimetry system. This system provides most accurate dose calculations on the patient compared with the
MIRD method4.
The high accuracy is, at least in part, due to the
Monte Carlo method allows human tissues to be characterized by elemental composition and mass density1.
Thus, a reliable estimation of human tissues (elemental
composition and mass density) must be obtained. It can
be obtained using Hounsfield units provided from CT
images7.
Based on this, the Nuclear Engineer Center of IPEN
developed the ICCT software (Image Converter Computed
Tomography)8. It converts CT images in tissue parameters
(mass densities (ρ) and elemental weights (ωi)). The method implemented in ICCT was described by Schneider,
Bortfield and Schlegel7.
The aim of this work was to compare the variation in
absorbed dose caused only by differences in tissue parameters ωi (elemental weight) estimated through two forms:
ICCT software and data acquired from the literature. For simulation, the source energies and type were chosen due to
the radionuclide characteristics used in Nuclear Medicine.
For this work, it was used the MCNP5/MCPLIB04 code
to perform the transport of radiation and to estimate the
absorbed dose.
Software ICCT (Image Converter Computed Tomography)
ICCT software uses the method developed by Schneider,
Bortfield and Schlegel7, which is based on a stoichiometric
calibration of Hounsfield units (H) with mass density and
elemental weights.
Using experimentally determined parameters, the ICCT
calculates Hounsfield units for 71 human tissues, whose
compositions were taken from literature9,10. Mass density
and elemental weights of any Hounsfield unit are obtained
through linear interpolation.
The ICCT binned into 24 groups the Hounsfield scale
for human tissues: one group for air, with range of -1000
to -950; one group for lung tissue, with range of -949 to
-120; seven groups for soft tissues, with range of -119 to
+120; and 15 groups for skeletal tissues, with range of
+120 to +1600.
Within each group, the elemental composition and
weights are constant. The mass density continuously
increases with the Hounsfield units, except in the small
range of 14–23, in which is assigned a constant density
of 1.03 g.cm-3.
The MCNP5 code
The MCNP code was developed at Los Alamos Laboratory
(Los Alamos, NM) and is used worldwide to solve neutron,
photon and electron couple transport problems. A main
feature is to provide several options for developing spatial and energetic distributions using complex geometric
shapes.
Therefore, the MCNP code offers several possibilities
for the users to model their problem11.
Assess of influence of elemental weight (ωi) in the
absorbed dose calculation
In order to evaluate the influence of elemental weight in
the absorbed dose, calculations have been performed
through simulations in which the absorbed dose for two
situations was estimated: (1) using the elemental weight
(ωi) of one of the 24 groups stipulated by ICCT, which
corresponds to a given range in Hounsfield scale; (2)
using the elemental weight (ωi) of human tissues obtained
by literature9,10, which corresponds to the same range in
Hounsfield scale.
To evaluate elemental weight alone, the mass density
(ρ) stipulated was the same either by group of ICCT or data
of literature. The mass density of group stipulated by ICCT
was adopted as reference value.
To simulate geometry, it was considered a cube with 2
mm dimension which will contain both tissues and radiation sources. This cube was immersed in a 2 cm diameter
water sphere.
Isotropic source was distributed in the whole cube, and
were considered photon sources of 0.02 to 2.75 MeV and
electron source of 0.1 to 4.0 MeV. The source energies
Influence of elemental weight of human tissues estimated by ICCT software in absorbed dose calculation
and type were chosen due to the radionuclide characteristics used in Nuclear Medicine.
In order to absorbed dose measurement within cube, it
was used the *F8 tally, which obtains deposit energy (MeV)
inside of the cube. The absorbed dose (Gy) was calculated
by expression:
Ex1.60217646E - 13
D=
[Gy]
(1)
ρxV
In this expression, E is the deposited energy (MeV) measured by *F8 tally; the constant 1.602E-13 is used for
convert MeV in Joule; ρ (g.cm-3) is the tissue mass density
in the cube; and V (cm3) the cube volume.
The comparison between absorbed doses of two situations was calculated as percentage relative difference:
D0’ - D0
Dose Difference(%) =
D0
x100
(2)
D0 is the absorbed dose using tissues acquired by
ICCT; and D0’ corresponds to the absorbed dose using tissue acquired by literature.
and O=-13.9% between the tissues contribute with 9.81%
in dose difference, 8.90% of which is due to C and 0.91%
to O. The dose difference high due to C is justified because the urine composition contains only 0.5% of C, while the group 7 contains 13.4%. However, from 0.1 up to
2.75 MeV, the dose differences were insignificant with a
maximum value of 1.37% at 2.75 MeV.
The differences for electrons were insignificant with a
maximum value of order of 0.88% at 0.7 MeV. Above this
energy, it has been observed a tendency to decrease as
energy increases.
Results of influence of elemental weight (ωi) in the absorbed dose calculation for the range skeletal tissue
The tissues defined for this case were the group 12 composition, obtained by ICCT software, and D6, L3 including cartilage (male) composition10. Table 2 shows the
elemental weight of each element that makes up the two
tissues.
The chart presenting the differences in dose for photons and electrons calculated with MCNP5/MCPLIB04 is
shown in Figure 2. They were obtained assuming that the
Results and discussion
In order to appraise the influence of elemental weight in the
absorbed dose, the experiment mentioned in section II-C
were performed.
Results of influence of elemental weight (ωi) in the
absorbed dose calculation for the soft tissue
The tissues defined for this case were the group 7 composition, obtained by ICCT software, and urine composition9.
Table 1 shows the elemental weight of each element that
makes up the two tissues.
Figure 1 shows the differences in dose for photons and
electrons calculated with MCNP5/MCPLIB04. They were
obtained assuming that the results acquired from group 7
composition (ICCT) are the reference values.
In terms of absolute values, the differences are considerable for photons reaching a maximum of 11.27% at
0.03 MeV. The analysis of this energy has shown that the
deviations in the elemental weight of H=-0.7%, N=2.0%,
Na=-0.2%, P=0.1%, S=0.2%, Cl=-0.4% and K=0% between the tissues contribute with 2.59% in dose difference. The differences in the elemental weight of C=12.9%
(a)
(b)
Table 1. Elemental weight of tissue composition obtained by
ICCT and literature for the soft tissue range.
Tissues
Group 7
ICCT
Urine
H
C
N
Elemental weight (%)
O
Na
P
S
Cl
K
10.3 13.4 3.0 72.3 0.2
0.2
0.2
0.2
0.2
11.0 0.5
0.1
-
0.6
0.2
1.0 86.2 0.4
Figure 1. Differences in absorbed dose for the soft tissue range
in two situation: (a) for photons and (b) electrons source. The
dose was calculated with MCNP5/MCPLIB04 code.
129
Massicano F, Possani RG, Cintra FB, Massicano AVF, Yoriyaz H
results acquired from group 12 composition (ICCT) are the
reference values.
As shown in the chart (a) of Figure 2, in the range from
0.02 up to 0.04 MeV, it was observed a small increase of
0.42% in dose differences with a maximum value of -1.08%
at 0.04 MeV. In the range from 0.04 up to 1.46 MeV, it was
observed a small decrease of 0.91% in dose differences.
After this range, it was observed a small increase of 0.26% in
dose differences up to 2.75 MeV reaching -0.43%. However,
almost all the differences were smaller than -1.0%.
Table 2. Elemental weight of tissue composition obtained by
ICCT and literature for the skeletal tissue range.
Tissues
Elemental weight (%)
O Na Mg P S
H
C
N
Cl K Ca
Group
12
7.5 35.8 3.1 38.1 0.1 0.1 4.8 0.2 0.1 0.1 10.3
ICCT
D6,L3
incl.
7.3 26.5 3.6 47.3 0.1 0.1 4.8 0.3 0.1 0.1 9.8
cartilage
(male)
(a)
(b)
The chart (b) of Figure 2 shows the dose differences
for electron in which a trend to increase the dose differences as energy increases was observed. The differences are
considerably smaller, reaching a maximum value of -0.25%
at 4.0 MeV.
The negatives signals mean that the calculated doses
with the group 12 (ICCT) composition were greater than
the calculated with the D6, L3 including cartilage (male)
composition10.
The small dose differences found were due to the high
precision of H, P and Ca, that are very important to dose
calculation in skeletal tissue12.
Conclusions
Accuracy in estimating human tissues composition is essential for the patient-specific dosimetry system. With this
concern, it was developed the ICCT software, which converts CT images in tissue parameters (mass densities (ρ)
and elemental weights (ωi)). In this work, it was proposed
the comparison between the absorbed dose caused by
differences in tissue parameters ωi estimated by ICCT and
the data acquired from the literature.
Regarding to soft tissue, considerable differences
in absorbed dose for photons reaching a maximum of
11.27% at 0.03 MeV were found. The analysis of this energy has shown that the differences in the elemental weight of C=12.9% and O=-13.9% were mainly responsible
for this high difference. Nevertheless, the other energies
analysis obtained insignificant values of dose differences,
with a maximum value of 1.37% at 2.75 MeV. For electrons, the differences in absorbed dose were insignificant,
with a maximum value of order of 0.88% at 0.7 MeV. It
becomes clear that considering photons source with low
energies must be made very carefully in estimation of the
soft tissue.
Concerning to skeletal tissue with photon sources of
0.02 to 2.75 MeV and electron source of 0.1 to 4 MeV,
most differences in absorbed dose were smaller than
-1.0%. In this case, the small dose differences found were
due to the high precision of the elements H, P and Ca estimated by ICCT. These elements are very important to dose
calculation in skeletal tissue.
In conclusion, the ICCT software reaches a reasonable
approximation for determine the elemental weights (ωi) of
tissues, obtaining low variances in the absorbed dose, except in relation to soft tissues, in which was found a high
variation of C and O, leading to significant differences in the
absorbed dose for energy smaller than 0.1 MeV.
Acknowledgment
Figure 2. Differences in absorbed dose for the skeletal tissue
range in two situations: (a) for photons and (b) electrons source.
The dose was calculated with MCNP5/MCPLIB04 code.
130
The authors thank the Conselho Nacional de
Desenvolvimento Científico e Tecnológico (CNPq) for partial economic support.
Influence of elemental weight of human tissues estimated by ICCT software in absorbed dose calculation
References
1.
2.
3.
4.
5.
6.
Divoli A, Chiavassa S, Ferrer L, Barbet J, Flux GD, Bardiès M. Effect of
patient morphology on dosimetric calculations for internal irradiation
as assessed by comparisons of Monte Carlo versus conventional
methodologies. J Nucl Med. 2009;50(2):316-23.
Stabin MG. The case for patient-specific dosimetry in radionuclide therapy.
Cancer Biother Radiopharm. 2008;23(3):273-84.
Zanzonico PB. Internal radionuclide radiation dosimetry: a review of basic
concepts and recent developments. J Nucl Med. 2000; 41(2):297-308.
Sgouros G. Dosimetry of internal emitters. J Nucl Med. 2005; 46(1):1827.
Grudzinski JJ, Yoriyaz H, Deluca PM Jr, Weichert JP. Patient specific
treatment planning for systemically administered radiopharmaceutical
using PET/CT and Monte Carlo simulation. Appl. Radiat. Isot.
2010;68(1):59-65.
Yoriyaz H, Stabin MG, dos Santos A. Monte Carlo MCNP-4B-based
absorbed dose distribution estimates for patient-specific dosimetry. J
Nucl Med. 2000;42(4):662-9.
Schneider W, Bortfeld T, Schlegel W. Correlation between CT number and
tissue parameters needed for Monte Carlo simulations of clinical dose
distributions. Phys Med Biol. 2000;45(2):459-78.
8. Massicano F. Quantificação de imagens tomográficas para cálculo de
dose em diagnose e terapia em medicina nuclear. [master’s thesis] São
Paulo – Instituto de Pesquisas Energéticas e Nucleares (IPEN); 2010.
9. Woodard HQ, White DR. The composition of body tissues. Br J Radiol.
1986;59(708):1209-18.
10. White DR, Woodard HQ, Hammond SM. Average soft-tissue and bone
models for use in radiation dosimetry. Br J Radiol. 1987;60(717):907-13.
11. X-5 Monte Carlo Team. MCNP: A general Monte Carlo N-particle transport
code, version 5. Report LA-CP-03-0245. New México: Los Alamos
National Laboratory; 2003.
12. Yoriyaz H, Moralles M, Siqueira PT, Guimarães CC, Cintra FB, Santos A.
Physical models, cross sections, and numerical approximations used in
MCNP and GEANT4 Monte Carlo codes for photon and electron absorbed
fraction calculation. Med Phys. 2009;36(11):5198-213.
7.
131
Artigo Original
Preliminary study of the 270 Bloom Fricke
xylenol gel phantom performance for 3D
conformal radiotherapy using multiple
radiation fields
Estudo preliminar do desempenho do simulador Fricke
xilenol gel 270 Bloom para radioterapia conformal
tridimensional usando campos de radiação múltiplos
Christianne C. Cavinato1, Benedito H. Souza2, Henrique Carrete Jr.2, Kellen A. C. Daros2,
Regina B. Medeiros2, Adelmo J. Giordani3 and Letícia L. Campos1
Gerência de Metrologia das Radiações do Instituto de Pesquisas Energéticas e Nucleares (IPEN-CNEN/SP) – São Paulo (SP), Brazil.
2
Departamento de Diagnóstico por Imagens da Universidade Federal de São Paulo (UNIFESP) – São Paulo (SP), Brazil.
3
Serviço de Radioterapia da Universidade Federal de São Paulo (UNIFESP) – São Paulo (SP), Brazil.
1
Abstract
The complex cancer treatment techniques require rigorous quality control (QC). The Fricke xylenol gel (FXG) dosimeter has been studied
to be applied as a three-dimensional (3D) dosimeter since it is possible to produce 3D FXG phantoms of various shapes and sizes. In this
preliminary study, the performance of the FXG spherical phantom developed at IPEN, prepared using 270 Bloom gelatin from porcine
skin made in Brazil, was evaluated using magnetic resonance imaging (MRI) technique, aiming to use this phantom to 3D conformal
radiotherapy (3DCRT) with multiple radiation fields and clinical photon beams. The obtained results indicate that for all magnetic resonance
(MR) images of the FXG phantom irradiated with 6 MV clinical photon beam can be observed clearly the target volume and, in the case of
coronal image, can also be observed the radiation beam projection and the overlap of different radiation fields used. The Fricke xylenol gel
phantom presented satisfactory results for 3DCRT and clinical photon beams in this preliminary study. These results encourage the additional
tests using complex treatment techniques and indicate the viability of applying the phantom studied to routine quality control measurements
and in 3DCRT and intensity modulated radiotherapy (IMRT) treatment planning.
Keywords: dosimeter, conformal radiotherapy, quality control, radiation oncology.
Resumo
As complexas técnicas para tratamento do câncer exigem um rigoroso controle de qualidade. O dosímetro Fricke xilenol gel (FXG) tem
sido estudado para ser aplicado como um dosímetro tridimensional (3D) já que é possível produzir simulador FXG 3D de diversas formas e
tamanhos. Neste estudo preliminar, o desempenho do simulador esférico FXG, que foi desenvolvido no Instituto de Pesquisas Energéticas
e Nucleares (IPEN), preparado utilizando a gelatina suína 270 Bloom, feita no Brasil, foi avaliado usando a técnica de imageamento por
ressonância magnética, com o objetivo de usar esse simulador para radioterapia conformal 3D com múltiplos campos de radiação e feixes
de fóton clínico. Os resultados obtidos indicam que, para todas as imagens por ressonância magnética do simulador FXG irradiado com
um feixe de fóton clínico de 6 MV, pode-se observar claramente o volume-alvo e, no caso da imagem na orientação coronal, a projeção
do feixe de radiação e a sobreposição dos diferentes campos de radiação utilizados. O simulador Fricke xilenol gel apresentou resultados
satisfatórios para a radioterapia conformal 3D e para os feixes de fótons clínicos neste estudo preliminar. Tais resultados apoiam testes
adicionais, utilizando técnicas complexas de tratamento, e indicam a viabilidade da aplicação do simulador estudado para medidas
periódicas do controle de qualidade e para o planejamento do tratamento com radioterapia com intensidade modulada e radioterapia
conformal 3D.
Palavras-chave: dosímetro, radioterapia conformal, controle de qualidade, radioterapia.
Corresponding author: Christianne Cobello Cavinato – Instituto de Pesquisas Energéticas e Nucleares (IPEN-CNEN/SP) – Av. Prof. Lineu Prestes, 2.242 –
Cidade Universitária – CEP: 05508-000 – São Paulo (SP), Brasil – E-mail: [email protected]
133
Cavinato CC, Souza BH, Carrete Jr. H , Daros KAC, Medeiros RB, Giordani AJ, Campos LL
Introduction
The sophisticated tumor treatments, such as threedimensional (3D) conformal radiotherapy (3DCRT) and
intensity modulated radiotherapy (IMRT), have grown in
use during the last few years because they have advantages over conventional radiation treatment techniques
since allow the delivery of a higher tumor dose while
maintaining an acceptable level of normal tissue complications. Before performing these radiation treatments,
is achieved a 3D target localization using computed tomography (CT) scans, for example, and 3D treatment
planning to differentiate accurately between tumor and
healthy tissue1,2.
Because of these complex treatment techniques,
the quality control (QC) must be strict. One dosimetric
system that has been studied3-5 for QC application in
these cases is the Fricke xylenol gel (FXG) dosimeter
since it is possible to produce 3D FXG phantoms of
various shapes and sizes. The dosimetric principle of
the FXG solution is the oxidation of ferrous ions (Fe2+),
originally present in a non-irradiated solution, to ferric
ions (Fe3+), which results from the action of ionizing radiation on aqueous solutions6.
In this preliminary study, it was verified the performance
of the FXG phantom developed at IPEN, prepared using
270 Bloom gelatin from porcine skin made in Brazil, for
3DCRT using multiple static radiation fields and clinical
photon beams. In a future work, this dosimeter will be evaluated using the IMRT technique, and 3DCRT additional
tests will be performed.
The FXG phantom and FXG samples for dose-response curves obtaining were prepared at High Doses
Laboratory (LDA) of IPEN. The gamma and photon irradiations were performed in the Radiotherapy Service, and
the dose evaluations were performed in the Resonance
Magnetic Service at Diagnostic Image Department of
the São Paulo Hospital (HSP), Federal University of São
Paulo (UNIFESP).
FXG phantom preparation
A spherical glass flask of 2000 mL, with 158.0 mm diameter, with one short neck, was completely filled with FXG
solution prepared according to Olsson7, using 5% by weight 270 Bloom gelatin from porcine skin, ultra-pure water
and the following analytical grade reagents: 50 mM sulphuric acid (H2SO4), 1 mM sodium chloride (NaCl), 1 mM
ferrous ammonium sulphate hexahydrate or Morh’s salt
[Fe(NH4)2(SO4)2.6H2O] and 0.1 mM ferric ions indicator xylenol orange (C31H28N2Na4O13S).
The FXG phantom was maintained under low temperature and light protected during 12 hours and 30 minutes
after preparation.
134
FXG samples preparation
Polymethylmethacrylate (PMMA) cuvettes with parallel optical faces measuring 10x10x45 mm3 and path length of
10 mm were filled with FXG solution in order to obtain the
magnetic resonance (MR) signal intensity in function of absorbed dose (dose-response curve).
Treatment planning
Computed tomography scans were obtained from an identical spherical glass flask used to FXG phantom preparation,
in this case, completely filled with tri-distilled water, using a
Philips® Brilliance CT 64-channel scanner (HSP/UNIFESP).
The 3D treatment planning was performed using the
Eclipse® External Beam Planning system version 7.3.10.
Multiple static radiation fields were used and the irradiation
parameters are presented in Table 1.
Irradiation
FXG samples irradiation: three FXG sample sets were prepared in order to obtain the arithmetic mean of three measurements. Each sample set was packed with polyvinyl
chloride (PVC) film in order to avoid contact of the FXG
solution with tri-distilled water of the water phantom used
to samples irradiation.
The FXG sample sets were maintained for approximately 30 minutes at room temperature and light protected before irradiation. The FXG samples were irradiated
with absorbed doses between 2 and 20 Gy, dose rate of
74.98 cGy.min-1, 40x40 cm2 field size, source-surface distance (SSD) of 80 cm and 0.5 cm PMMA build-up thickness, using a General Electric Company® Alcyon II 60Co
gamma radiation (HSP/UNIFESP) and a water phantom
(PMMA 22x22x10 cm3 filled with tri-distilled water).
The experimental set up for FXG samples irradiation with
60
Co gamma radiation is presented in Figure 1. All FXG sample
sets were positioned together in the water phantom (Figure 1b)
and each set was removed when the radiation exposure time
needed to obtain the desired absorbed dose was completed.
FXG phantom irradiation: the FXG phantom was positioned in a Styrofoam box containing ice cubes in order to
maintain the phantom under low temperature and light protected to be transported to the irradiation site. The phantom
was removed of the Styrofoam box and maintained during
30 minutes at room temperature and light protected before
irradiation. The FXG phantom was housed in a foam backer
to be irradiated with 6 MV clinical photon beams with absorbed dose of 20 Gy and dose rate of 300 cGy.min-1 using
a Varian® Clinac 600C linear accelerator (HSP/UNIFESP). A
Cerrobend® shielding block with shape simulating a bladder
tumor inside the phantom was used. Other irradiation parameters are presented in Table 1.
The experimental set up for FXG spherical phantom irradiation with clinical photon beams is presented in Figure 2.
Evaluation
The evaluation technique employed was the magnetic
resonance imaging (MRI) using a Siemens® Magnetom®
Preliminary study of the 270 Bloom Fricke xylenol gel phantom performance for 3D conformal radiotherapy using multiple radiation fields
Sonata Maestro Class 1.5 T MRI scanner (HSP/UNIFESP).
The MRI scans of the FXG solution were obtained on cranium protocol-T1 approximately 30 minutes after irradiation. The MR images acquisition parameters are presented in Table 2.
The softwares syngo fastView® version VX57F24 and
ImageJ® version 1.42q were used to process the MRI
scans obtained.
Results
Dose-response curve
The MR images (coronal orientation) of the PMMA cuvettes
filled with FXG solution irradiated with 60Co gamma radiation (dose range from 2 to 20 Gy) is presented in Figure 3.
The dose-response curve of MR signal intensity in function of absorbed dose obtained from the image presented
in Figure 3 is presented in Figure 4. The background values
corresponding to the MRI measurements of non-irradiated
FXG samples were subtracted from all MR signal intensity
values presented.
FXG phantom MR imaging
The MR images of the FXG phantom non-irradiated and
irradiated with 6 MV photons are presented in Figure 5. MR
slices in different orientations of the irradiated FXG phantom is presented in Figure 6.
Isodose curves and 3D reconstruction showing
the multiple static radiation fields of the FXG spherical
phantom irradiated with 6 MV photons are presented in
Figures 7 and 8, respectively.
Table 1. Irradiation parameters for the FXG phantom.
Radiation field Gantry position (°) Treatment couch position (°)
1
2
3
210.0
30.0
120.0
(a)
0.0
0.0
350.0
X1
+3.45
+3.50
+4.00
(b)
Figure 1. Experimental set up for FXG samples irradiation with
60
Co gamma radiation (a) and FXG sample sets positioned in
water phantom (b).
Field size (cm)
X2
Y1
+3.45
+2.5
+3.50
+2.5
+4.50
+2.5
SSD (cm) Monitor Unit (MU)
Y2
+3.0
+3.0
+3.5
91.9
91.9
91.9
1731
1882
1661
Figure 2. Experimental set up for FXG spherical phantom irradiation with 6 MV clinical photon beams using a Clinac 600C
linear accelerator.
Table 2. MR images acquisition parameters.
Figure 3. Coronal MR images of the FXG solution conditioned in
PMMA cuvettes and irradiated with 60Co gamma radiation.
Image orientation
Field of view (FOV) (mm)
Slice thickness (THK) (mm)
Voxel (mm)
Gap (mm)
Time of repetition (TR) (ms)
Time of echo (TE) (ms)
Flip angle (º)
Matrix size (MS) (pixels)
Number of signals averaged (NSA)
Slices number
Coil
Channels
Coronal, sagittal and axial
256
1.0
1.0x1.0x1.0
0.5
2000
3.42
15
256x256
1
176
Head
8
135
Cavinato CC, Souza BH, Carrete Jr. H , Daros KAC, Medeiros RB, Giordani AJ, Campos LL
(a)
(b)
Figure 5. Sagittal MR images of the FXG phantom non-irradiated (a) and irradiated (b) with 6 MV photons.
Figure 4. MR signal intensity curve in function of absorbed dose
of the FXG samples image.
Figure 7. Isodose curves of FXG phantom irradiated with 6 MV
photons (sagittal Eclipse® image).
Figure 6. Coronal, sagittal and axial MR slices of the FXG phantom irradiated with 6 MV photons.
Discussion
136
The MR signal intensity in function of absorbed dose in the
radiotherapy dose range interest presents linear behavior.
The obtained results indicate that for all MR images of
the FXG spherical phantom irradiated with 6 MV clinical
photon beams can be observed clearly the target volume,
and in coronal image (Figure 6) can also be observed the
radiation beams projection and the overlap of different radiation fields used.
techniques and indicate the viability of applying the phantom studied for routine quality control measurements and
in 3DCRT and IMRT treatment planning.
Conclusions
Acknowledgments
The Fricke xylenol gel phantom prepared with 270 Bloom
gelatin from porcine skin made in Brazil presented satisfactory results for 3D conformal radiotherapy and clinical
photon beams in this preliminary study. These results
encourage the additional tests using complex treatment
The authors are grateful to the staffs of the Radiotherapy
Service and Resonance Magnetic Service of the Diagnostic
Image Department of the HSP/UNIFESP to allow the FXG
irradiations and MR evaluations, respectively, and CAPES,
CNPq, IPEN and CNEN by the financial support.
Figure 8. Three-dimensional reconstruction of the target volume irradiated showing multiple fields used (Eclipse®).
Preliminary study of the 270 Bloom Fricke xylenol gel phantom performance for 3D conformal radiotherapy using multiple radiation fields
References
1. Podgorsak EB. Radiation Oncology Physics: a handbook for teachers and
students. Vienna: International Atomic Energy Agency; 2005.
2. Stanton R, Stinson D. Applied Physics for Radiation Oncology. Madison:
Medical Physics; 1996.
3. Olding T, Salomons G, Darko J, Schreiner LJ. A practical use for FXG gel
dosimetry. J Phys Conf Ser. 2010;250:012003.
4. Bero MA, Zahili M. Radiochromic gel dosimeter (FXG) chemical yield determination
for dose measurements standardization. J Phys Conf Ser. 2009;164:012011.
5. Olding T, Darko J, Schreiner LJ. Effective management of FXG gel dosimetry.
J Phys Conf Ser. 2010;250:012028.
6. Gore JC, Kang YS, Schulz RJ. Measurement of radiation dose distributions
by nuclear magnetic resonance (NMR) imaging. Phys Med Biol.
1984;29(10):1189-97.
7. Olsson LE, Pertersson S, Ahlgren L, Mattsson S. Ferrous sulphate gels for
determination of absorbed dose distributions using MRI technique: basic
studies. Phys Med Biol. 1989;34(1):43-52.
137
Artigo Original
The contribution of optically stimulated
luminescence dosimetry in quality control
in radiotherapy
A contribuição da dosimetria por luminescência opticamente
estimulada no controle de qualidade na radioterapia
Renato S. Fernandes1 and Yvone M. Mascarenhas2
Universidade de Mogi das Cruzes, Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo – São Paulo (SP), Brazil.
2
Universidade de Mogi das Cruzes/Sapra Landauer – São Carlos (SP), Brazil.
1
Abstract
In 2011, Brazil will have more than 489,000 new cases of cancer. Of these patients, a considerable contingent will be submitted to
radiotherapy procedures. Thus, efficient systems that guarantee the quality of the beams used in radiotherapy procedures are extremely
important because collaborate with the overall success of treatment. This paper presents the use of OSL (optically stimulated luminescence)
dosimetry procedures in quality control in radiotherapy, making verification of the symmetry of treatment of radioactive fields emitted by
linear accelerators. The use of OSL dosimetry was compared to procedures performed daily, using ionization chambers. Dosimeters of
aluminum oxide doped with carbon (Al2O3: C) were distributed on a map, with the delimitation of a field 20x20 cm arranged as follows:
one in the center of the field; four equidistant distributed, forming a square 10x10 cm; and 4 remaining distributed equidistant, forming
a square of 20x20 cm. This arrangement is similar to equipment used for checking the symmetry of radioactive fields using ionization
chambers. The analysis of data obtained in the symmetry axis X made a variation of 0.2% in OSL dosimetry while using the equipment
provided with an ionization chamber, 0.1%. Noting the Y axis, the variation of data for OSL dosimetry was 0.04% and ionization chambers,
0.1%. The use of OSL dosimetry proved to be simple, with accessible instrumentation and making a possible reinterpretation of the data
obtained. However, the data suggest that the broader performance of procedures using OSL dosimetry, seeking greater familiarity with the
system, can reduce the variation in results.
Keywords: dosimetry, radiation therapy, quality control, cancer, ionization chambers.
Resumo
Em 2011, o Brasil terá mais de 489.000 novos casos de câncer. Desses pacientes, um notável grupo será submetido a procedimentos
de radioterapia. Desse modo, sistemas eficazes que garantam a qualidade dos feixes utilizados nos procedimentos de radioterapia são
extremamente importantes, pois colaboram com o sucesso total do tratamento. Este trabalho apresenta o uso dos procedimentos de
dosimetria de luminescência opticamente estimulada (OSL) no controle de qualidade na radioterapia, verificando a simetria dos campos
radioativos emitidos por aceleradores lineares. O uso de dosimetria OSL foi comparado aos procedimentos realizados diariamente, utilizando
câmaras de ionização. Dosímetros de óxido de alumínio dopados com carbono (Al2O3: C) foram distribuídos em um mapa, com a delimitação
de um campo de 20x20 cm organizado da seguinte maneira: um no centro do campo; quatro distribuídos equidistantes, formando um
quadrado de 10x10 cm; e quatro restantes, sendo distribuídos equidistantes, formando um quadrado de 20x20 cm. Essa organização é
similar ao equipamento utilizado para verificar a simetria dos campos radioativos que utilizam câmaras de ionização. A análise dos dados
obtidos no eixo X da simetria fez uma variação de 0,2% na dosimetria de OSL, enquanto ao utilizar o equipamento fornecido com uma câmara
de ionização fez 0,1%. Ao notar o eixo Y, a variação dos dados para a dosimetria de OSL foi de 0,04% e as câmaras de ionização, 0,1%. O uso
da dosimetria de OSL provou ser simples, com instrumentação acessível e tornou possível a reinterpretação dos dados obtidos. Entretanto, os
dados sugerem que a realização de procedimentos mais amplos utilizando dosimetria OSL, visando uma maior familiaridade com o sistema,
podem reduzir a variação dos resultados obtidos.
Palavras-chave: dosimetria, terapia de radiação, controle de qualidade, câncer, câmaras de ionização.
Corresponding author: Renato da Silva Fernandes – Hospital das Clínicas da FMUSP – Av. Dr. Enéas Carvalho de Aguiar, 255 – CEP: 05403-000 – São
Paulo (SP), Brazil – E-mail: [email protected]
139
Fernandes RS
Introduction
The Instituto Nacional do Cancer (INCA) showed, in
2009, an estimated new cancer cases in Brazil for the
year 2010. These data reported that more than 489,000
new patients were affected by some cancer in 2010. This
information will also be valid for the year 20111.
The use of radiotherapy, either alone or combined
with other methods (chemotherapy and surgery), has
emerged as an effective curative modality, especially in
malignant neoplasm in the initial stages2.
Radiotherapy is the use of ionizing radiation for therapeutic purposes. Radiation doses used are high and any
error during the treatment procedure can cause serious
patient injury, including death3.
The need to systematize actions for the quality control of radiotherapy treatments has been evident in recent
years4.
According to World Health Organization, the quality
control in radiotherapy is established based on the actions that ensure consistency between the clinical prescription and their administration to patients in relation to
the target volume, the lowest level in healthy tissue, minimal exposure of personal checks on the patient and
adequate patient monitoring for determining the outcome
of treatment5.
An efficient system of quality assurance in radiotherapy
minimizes errors in treatment planning and administration
of the dose to the patient, allows the intercomparison of results between different treatment centers and provides the
possibilities and functionalities of a processing equipment
to be used with a high level of consistency and accuracy4.
Several types of dosimeters can be used for quality control in radiotherapy beams: ionization chambers,
diodes, films, metal oxide semiconductor (MOSFET) and
TLD (thermoluminescent dosimeters)6-8.
New dosimeters and dosimetry procedures have
been developed. We can highlight the optically stimulated
luminescence dosimetry (OSL)9.
The arrival of use of technology in radiotherapy increases its effectiveness, as well as the errors incidence10.
It is critical that the verification procedures of the
beam are simple and reliable4.
Thus, the use of OSL dosimetry shows promise. It has
advantages such as high luminescence efficiency, stability, sensitivity, precision and accuracy, control of luminescence emitted, speed reading, rereading and possibility
of low power consumption, making possible the use of
handheld devices11.
The symmetry of the fields was the parameter chosen
for this analysis.
These dosimeters were arranged on a map, with the delimitation of a field 20x20 cm arranged as follows: one
in the center of the field; four equidistant distributed,
forming a square 10x10 cm; and the remaining four
equidistant distributed, forming a square of 20x20 cm
(Figure 1). The dosimeters were arranged in this way to
ensure intercomparison with data obtained during the
commissioning of the linear accelerator.
Dosimeters were placed on 1.5 cm of solid water
(acrylic) to ensure set build up (Figure 2).
The model was irradiated with 100 cGy at a distance
of 100 cm (isocenter), using a 6 MeV beam produced in
a linear accelerator from Varian Medical Clinac 600c sn
515. To read these dosimeters, it was used MicroStar
produced by Landauer Co. (Figure 3).
The data obtained were compared with readings taken every day from Double CheckTM equipment
(Figure 4), Victoreen®. This equipment consists of a
reader, equipped with nine ionization chambers which,
when irradiated by a radiation field of 20x20 cm, capture
signals in the center of the field, 5 and 10 cm. The readings were taken between 1 and 7 October 2010.
Figure 1. Placed in fields dosimeters 10x10 cm and 20x20 cm.
OSL dosimeters were irradiated 9 (InLightTM with detectors of aluminum oxide doped carbon in the atmosphere).
140
Figure 2. Dosimeters arranged on the solid water. On the whole,
1.5 cm in the same field in order to ensure build up.
The contribution of optically stimulated luminescence dosimetry in quality control in radiotherapy
Discussion
The data obtained with the OSL dosimetry allow us to
observe a change in the symmetry of the field in the X
direction of approximately 0.2% (Figure 5A). With the
equipment use in the reading routine, this variation was
around 0.1%. Note that the fluctuations in the readings
are accompanied by two similar systems. In Y, we see
a variation around 0.04% when using the OSL dosimetry (Figure 5B). With the routinely use of the equipment,
we note a variation of 0.10%. Fluctuations in the readings are noted in a similar way in the two systems.
Conclusions
The OSL dosimetry was presented as an important tool
in the daily check of the symmetry of the radiation fields
emitted by linear accelerators.
Features, such as possibility of re-readings, stability,
sensitivity and ease of handling equipment, ensure wide
use OSL dosimetry in radiotherapy.
Expansion in the frequency and quantity measures can
provide more familiar to all, with steps leading to a smaller
range of variation.
More reading using the OSL dosimetry should be conducted to obtain more data for comparison with the system used routinely.
Figure 3. MicroStar Reader.
Acknowledgment
The authors thank the Department of Radiation
Oncology and Department of Medical Physics of Istituto
de Radiologia do Hospital das Clínicas da Faculdade de
Medicina da USP and Sapra Landauer for their support
in this work.
Figure 4. Double CheckTM model 7200, Victoreen®.
Results
-0,00%
-0,10% 0
-0,20%
-0,30%
-0,40%
-0,50%
-0,60%
-0,70%
-0,80%
-0,90%
1
2
3
4
5
0,10%
6
0,08%
Ionization
Chamber
OSL
Dosimetry
Readings
Variation of Simmetry Axis X - Ionization Chamber X OSL
Dosimetry
Variation
Variation
Variation of Simmetry Axis X - Ionization Chamber X OSL
Dosimetry
0,06%
Ionization
Chamber
0,04%
0,02%
-0,00%
-0,02%
OSL
Dosimetry
0
1
2
3
4
5
6
Readings
Figure 5. Comparison between data obtained with both ionization chamber and OSL dosimetry for variation of symmetry of the field
in the X (a) and Y (b) directions.
141
Fernandes RS
References
1. Brasil. Ministério da Saúde. Instituto Nacional de Câncer. Estimativa 2010:
incidência de câncer no Brasil/Instituto Nacional de Câncer. Rio de Janeiro:
INCA; 2009.
2. Perez CA, Brady LW. Principles and pratice of Radiation Oncology.
Philadelphia: JB Lippincott; 2004.
3. Berdaky MF, Caldas LVE. Implantação de um programa de controle de
qualidade de um acelerador linear de 6 MeV de fótons. Radiol Bras.
2001;34(5):281-4.
4. Brasil. Ministério da Saúde. Instituto Nacional de Câncer. TEC DOC - 1151:
aspectos físicos da garantia da qualidade em radioterapia. Rio de Janeiro:
INCA, 2000.
142
5. World Health Organization. Quality assurance in radiotherapy. Geneva:
WHO; 1988.
6. Bentel GC. Radiation therapy planning. New York: McGraw-Hill; 1996.
7. Hendee WR, Ibbott GS. Radiation therapy physics. St. Louis: Mosby; 1996.
8. Scaff LAM. Física da Radioterapia. São Paulo: Sarvier; 1997.
9. Yukihara, EG, Mckeever SWS. Optically stimulated luminescence (OSL)
dosimetry in medicine. Phys Med Biol. 2008;53(20):351-79.
10. Alvarez L. Tecnologia aumenta eficácia e incidência de erros da radioterapia.
O Estado de São Paulo [Internet] [cited 2011 Jan 16]. Available from: http://
www.estadao.com.br/estadaodehoje/20100208/not_imp507824,0.php.
11. McKeever SW, Moscovitch M. On the advantages and disadvantages of
optically stimulated luminescence dosimetry and thermoluminescence
dosimetry. Radiat Prot Dosimetry. 2003;104(3):263–70.
Artigo Original
Evaluation of the heterogeneity corrections
impact in lung stereotatic body
radiation therapy
Avaliação do impacto das correções de heterogeneidade
na radioterapia estereotáxica corpórea pulmonar
José Eduardo V. Nascimento, Ana Claudia M. de Chiara, Thais M. Casagrande, Tatiana M.
M. T. Alves, Wellington F. P. Neves-Junior, Anselmo Mancini, Eliana Capella, Edilson Pelosi
and Cecilia K. Haddad
Hospital Sírio Libanês – São Paulo (SP), Brazil.
Abstract
Stereotactic body radiation therapy (SBRT) refers to an emerging radiotherapy that is highly effective in controlling early primary and
oligometastic cancers at locations throughout the abdominopelvic and thoracic cavities, and at spinal and paraspinal sites. Some protocols
have been developed for this procedure. In the special cases of lung, there are protocols in use that consider heterogeneity corrections and
others that do not make use of heterogeneity correction. In this work, we recalculated, considering the different tissue densities plans initially
optimized without heterogeneity corrections to evaluate the dosimetric changes that occurs, for example, PTV (planning target volume)
coverage, dose to isocenter and dose to critical structures, and we calculated gamma function between the dose plans originated in the two
conditions. We also performed the superposition between the calculated gamma function with the respective CT slice in order to evaluate in
what conditions occur the major differences between the conditions of calculus considered. The results showed that relevant variations occur
between the two situations of calculus. With the superposition of the image relative to γ index and its respective CT slice, we could visualize
where the greatest discrepancies occur. These data allow us to evaluate with more accuracy the doses delivered to the target and organs at
risk and compare different protocols, independently of the use or non-use of heterogeneity corrections.
Keywords: lung, stereotactic body radiation therapy, conformal radiotherapy, computer-assisted image processing.
Resumo
A Radioterapia Estereotática Extra-Cranial (SBRT) refere-se à técnica de radioterapia emergente que é altamente efetiva no controle de tumores
primários em estágio inicial e oligometastático localizados nas cavidades abdominais e torácicas, e em sítios espinhais e paraespinhais.
Alguns protocolos têm sido desenvolvidos para esse procedimento. Nos casos especiais de pulmão, há protocolos que consideram as
correções de heterogeneidade e outros que não o fazem. Neste trabalho, nós recalculamos, considerando as diferentes densidades do
tecido, planos inicialmente otimizados sem correção de heterogeneidade, a fim de avaliar as mudanças dosimétricas que ocorrem como, por
exemplo, cobertura do PTV (volume alvo), dose no isocentro e dose em órgãos de risco; foi ainda calculada a função gamma entre as duas
condições de cálculo. Ainda, executamos a sobreposição do cálculo da função gamma com o respectivo corte tomográfico, para avaliar em
quais condições ocorrem as maiores diferenças entre as duas situações de cálculo apresentadas. Os resultados mostraram que ocorrem
variações relevantes entre as duas situações de cálculo. Com a sobreposição da imagem relativa à função gamma com seu respectivo corte
tomográfico, conseguimos vusualizar as regiões em que ocorrem as maiores discrepâcias. Tais dados permitem avaliar com mais precisão a
distribuição de dose no alvo e nos órgãos em risco, e comparar diferentes protocolos, independentemente do uso ou não uso das correções
da heterogeneidade.
Palavras-chave: pulmão, radioterapia estereotáxica corpórea, radioterapia conformal, processamento de imagem assistida por computador.
Corresponding author: José Eduardo Vaz Nascimento – Departamento de Radioterapia do Hospital Sírio Libanês – Adma Jafet, 91 – CEP: 01308-050 –
São Paulo (SP), Brasil – E-mail: [email protected]
143
Nascimento JEV, Chiara ACM, Casagrande TM, Alves TMMT, Neves-Junior WFP, Mancini A, Capella E, EP, Haddad CK
Introduction
Lung cancer remains the most frequent cause of cancer
death in both men and women in developed countries1.
Of the patients with bronchogenic carcinoma, 75% will
be diagnosed with non-small cell lung cancer (NSCLC).
Approximately 15-20% of NSCLC patients present early
or localized disease. Although surgical resection of Stage I
(Stage T1-T2N0) NSCLC is the classical treatment, some
patients with early-stage NSCLC are unable to tolerate the
rigors of surgery or the postoperative recovery period because of the lack of an adequate respiratory reserve, cardiac
dysfunction, diabetes mellitus, vascular disease, general frailty or other morbidities2. In addition, promising clinical results
of SBRT for early-stage lung cancer have been reported by
several groups3. Nowadays, there are protocols that make
special recommendations in all steps of the clinical procedure, immobilization, image acquisition, treatment planning
target localization, delivery of the treatment, etc. However,
the protocols do not make a unique recommendation about
the use or not of the heterogeneity correction in the treatment planning, and it is known that dose calculation with
and without tissue heterogeneity corrections have dramatic
deviations for the treatment planning. The differences exist
for the dose to the isocentric point and for the dose distributions, including target coverage and normal structure sparing. This is because the air-tissue interfaces present in the
thorax, where the effects of transient electronic disequilibrium
and increased lateral electron range in air will result in an important reduction in the central axis dose beyond the cavity
and potentially an underdosage of the tumor. So, for an accurate calculus of dose distribution in a treatment planning,
heterogeneity correction becomes extremely important, and
dose-calculation algorithms which do not account for lateral
electron scattering can yield incorrect results.
In order to evaluate accurately dose distributions, PTV
coverage, critical structures sparing and conformity index in
SBRT treatments plans originally optimized without heterogeneity corrections, and to be able to compare different SBRT
protocols, we recalculated these plans with tissue density
correction. Later, we calculated gamma function between
two situations of calculus and we performed the superposition of the calculated gamma function with the respective CT
slice, where we evaluated in what conditions occur the major
differences between the plans.
Unlike conventional radiotherapy, which is based on the delivery of a uniform prescription dose to the target volume, a
paradigm of prescribing dose for SBRT is based on the following set of conditions: a limited volume of tissue, containing
the gross tumor and its close vicinity, is targeted for treatment
through exposure to a very high per fraction, and hotspots
within the target are often deemed to be acceptable; the volume of normal tissue receiving high doses outside the target
144
shoud be minimized to limit the risk of treatment toxicity. Thus,
the gradient describing the dose fall-off outside the target
should be sharp, which is accomplished by prescribing SBRT
plans at low isodoses (e.g., 80% isodose) and with small or
no margins for penumbra at target edge. Hence, treatment
plans become complex and there are general metrics that
must be analyzed, such as target coverage (D95, dose that
covers 95% of the PTV; D99, dose that covers 99% of the
PTV), prescription ICRU reference point, plan conformity (ratio
of prescription isodose volume to PTV volume), dose fall-off
outside the target (e.g., ratio of the volume of the 50% of prescription isodose curve to PTV volume) and doses to organs
at risk. Ten plans from different protocols have been evaluated in the Oncentra Treatmet Planning (OTP). The original
plan was calculated and optimized with pencil beam without
heterogeneity corrections, 6 MV photon beam. Therefore,
with the same monitor units (MU), the plan was recalculated
with heterogeneity correction. The algorithm used here was
Collapsed Cone. The Collapsed Cone algorithm is a volumeoriented algorithm that accounts also for lateral energy transport. It will, therefore, give a reasonable accurate description
of the dose distributions in situations with marked inhomogeneities. It is based on precalculated point kernels that describe the deposition of energy from a photon interaction site
as a function of direction and distance. The dose concept is
to calculate the dose to the actual medium itself rather than to
the Bragg-Gray water cavity4. In order to evaluate if the use of
distinct algorithms would induce, we performed simulations
in homogenous medium with the two algorithms, being the
agreement between them about 0.5%.
To evaluate the dosimetric changes between the two
situations of calculus, we analyzed dose to isocenter, D95,
D99, conformity index (CI) and dose to critical structures.
Nevertheless, the results of these parameters do not show
the specific regions where these differences occur. To a better comprehension of the clinical implications that may arise
due to the use or nonuse of heterogeneity correction, we
calculated the γ index5 between both dose distributions and
we performed superposition of the image resulting from the
gamma function and the respective CT slice.
Results
Figure 1 shows the isodode distribution with and without tissue density correction in a slice relative to the central PTV
volume. With heterogeneity correction, occurred loss of PTV
coverage by the prescription dose, while the 50% prescription isodose extended over the volume. The percentage of
PTV volume receiving the prescribed dose decreased and the
50% prescription isodose volume, increased. Figure 2, PTV
dose volume histogram, illustrates the loss of PTV coverage
and more hotspots incidence.
It is illustrated the isocenter dose ratio, D95 ratio, D99
ratio from plans with and without tissue density corrections
on Figures 3, 4 and 5, respectively. In general, the isocenter
dose increases in plans with tissue density corrections, since
Evaluation of the heterogeneity corrections impact in lung stereotatic body radiation therapy
esophagus varied from -7 to 43%. The maximum dose to
heart varied from -7 to 8%. The maximum spinal cord dose
varied from -7 to 22%. The maximum dose to brachial plexus
ranged from -11 to -5%, and the maximum dose to bronchus, from 10 to 12%. The average dose to these organs had
small variations, within about 10%.
The metrics values analyzed give us useful information
of the changes that happen between both situations of
calculus, they do not refer to the specific sites where the
changes occur. Figure 6 shows a schematic representation
of the workflow applied to evaluate the specific locations
Isocenter Dose Ratio - Heterogeneity
there is less attenuation in lung volumes. The isocenter dose
difference ranged from 2.1 to 7.0% (mean, 5.5%; standard
deviation, 1.4%). The D95 e D99 ratios show a tendency to
PTV loss of coverage. The differences between D95 values
for plans with and without corrections of tissue density corrections ranged from -15.9 to 3.0% (mean, -6.5 %; standard
deviation, 11.7 %). The differences between D95 values for
plans with and without heterogeneity corrections ranged from
-17.8 to 5.8% (mean, -6.1%; standard deviation, 19.7%). The
conformity index (CI), the ratio of the volume of the prescription isodose to that of the PTV, it’s an important parameter to
evaluate the quality of SBRT treatment planning. In the cases
analyzed in this work, there were no significant changes in the
CI values for the two situations of calculus. However, 1 of the
10 cases had a CI for density unit calculus of 1.26 and, with
heterogeneity corrections, 0.4. With and without heterogeneity corrections, the CI values were within the tolerance of the
respective requirements protocol.
For the critical structures, the percentage of the lung volume receiving ≥20 Gy was limited to be not >10%, and other
constraints are listened in Table 1. Although the dose distributions change significantly with heterogeneity corrections,
the constraints were still respected. The maximum dose to
1,08
1,07
1,06
1,05
1,04
1,03
1,02
1,01
0
2
6
4
Case Number
8
10
Volume
[%]
95.00
90.00
85.00
80.00
75.00
70.00
65.00
60.00
55.00
50.00
45.00
40.00
35.00
30.00
25.00
20.00
15.00
10.00
5.00
0.00
0.00
PB OFF
CC ON
1000.00 2000.00 3000.00 4000.00 5000.00 6000.00 7000.00 Dose
[cGy]
Figure 2. Dose-volume histograms for PTV coverage with unit
density (dashed-line) and with density corrections (solid line).
1,05
1
0,95
0,9
0,85
0,8
0,75
0,7
0
2
4
6
Case Number
8
10
Figure 4. The D95 ratio between plans calculated with and without density corrections.
D99 Ratio - Heterogeneity on/off
Figure 1. Isodose distributions with unit density (left) and with
density corrections (right) in axial (upper) and sagital (lower) projections. Prescription and 50% prescription isodoses curves.
D95 Ratio - Heterogeneity on/off
Figure 3. The isocenter dose ratio between plans calculated
with and without density corrections.
1,1
1,05
1
0,95
0,9
0,85
0,8
0,75
0,7
0
2
4
6
Case Number
8
10
Figure 5. The D99 ratio between plans calculated with and without density corrections.
145
Nascimento JEV, Chiara ACM, Casagrande TM, Alves TMMT, Neves-Junior WFP, Mancini A, Capella E, EP, Haddad CK
Table 1. Dose constraints for critical structures.
Organ Spinal cord
Esophagus
Brachial plexus
Heart/pericardium
Trachea and large bronchus
Max critical volume above
threshold
<0.35 cc
<5 cc
<3 cc
<15 cc
<4 cc
Threshold dose
(Gy)
18 (6 Gy/fx)
17.7 (5.9 Gy/fx)
20.4 (6.8 Gy/fx)
24 (8 Gy/fx)
15 (5 Gy/fx)
Max point dosea
(Gy)
23.1 (7.7 Gy/fx)
25.2 (8.4 Gy/fx)
24 (8 Gy/fx)
30 (10 Gy/fx)
30 (10 Gy/fx)
End point
(≥Grade3)
Myelitis
Stenosis/fistula
Neuropathy
Pericarditis
Stenosis/fistula
“Point” defined as 0.035 cc or less.
a
a)
b)
c)
[cm]Z
0.0
-2.0
-4.0
-6.0
-8.0
-10.0
-12.0
-14.0
-16.0
-18.0
-20.0
-22.0
-24.0
-26.0
-28.0
100% = 7197.74 cGy
-20.0-16.0-12.0 -8.0 -4.0 0.0 4.0 8.0 12.0 16.0 20.0
[cm]X
[cm]Z
0.0
-2.0
-4.0
-6.0
-8.0
-10.0
-12.0
-14.0
-16.0
-18.0
-20.0
-22.0
-24.0
-26.0
-28.0
100% = 7586.56 cGy
-20.0-16.0-12.0 -8.0 -4.0 0.0 4.0 8.0 12.0 16.0 20.0
[cm]X
0.0
-1.0
-2.0
-3.0
-4.0
-5.0
-6.0
-7.0
-8.0
-9.0
-10.0
-11.0
-12.0
-13.0
-14.0
-15.0
-16.0
-17.0
-18.0
-19.0
-20.0
-21.0
-22.0
-23.0
-24.0
-25.0
-26.0
-27.0
-28.0
-29.0
-20.0-16.0-12.0 -8.0 -4.0 0.0 4.0 8.0 12.0 16.0 20.0
[cm]X
Figure 6. Schematic representation of the process to perform the image superposition between calculated γ index image and its respective slice CT. a) Tomographic slice and the isodose distribution in OTP. Left, isodoses curves originated from the non-corrected tissue density
value calculus; right, isodoses curves originated from the density corrections calculus. b) RT DOSE exported to OmniPRO IMRT® workspace. c) Images to be superimposed: left, the calculated gamma function between the two RTDOSE in b); right, the respective CT slice.
146
Evaluation of the heterogeneity corrections impact in lung stereotatic body radiation therapy
where happen the prominent differences. From the calculated plans, calculated with and without heterogeneity corrections, the RTDOSE file were exported to the OmniPRO
IMRT® workspace, in which the γ index was calculated.
Hence, we ran the superposition of the resulting image
from the calculated γ index with the respective CT slice.
The resulting images are shown in Figure 7. With the
overlay of the resulting gamma function image with the
corresponding CT slice, it’s clear that the greatest differences occur in regions of tissue-air interface, due to
non-electronic equilibrium, and in regions of low electron
density (lung), due to the transient electronic disequilibrium
and increased lateral electron range.
Discussion
The results showed that important variations occur in the isodose curves generated with and without heterogeneity correction. There are significant loss of PTV volume coverage (in
some cases, close to 20% of PTV loss of coverage) and, in
general, percentual increase in dose received by the critical
structures. It’s important to keep in mind when look to these
results that the plans were not reoptimized after calculated
with heterogeneity corrections. However, with this analysis,
we can evaluate dose delivered to the tumor site and critical structures with more accuracy. Still, current and newer
protocols are being designed with the recommendations to
use heterogeneity corrections in treatment planning calculus.
A long as the use of heterogeneity corrections changes dramatically, the analysis carried out in this work enable dosimetric comparison between different protocols.
The superposition of the image resulted from the gamma
function with the respective CT slice habilitate us to evaluate the specific location where occur the major differences
between both calculus configurations. Hence, we may infer
about the clinical implications of these differences.
Conclusion
The heterogeneity corrections affect significantly dose distributions. The evaluation ran in this work enable us to infer
with more accuracy in dose delivered to critical structures
and tumor volume, and evaluate the specific sites where
the major discrepancies occur.
References
Figure 7. Images generated from the co-registration of the
gamma function image with the corresponding CT slice.
1. Benedict SH, Yenice KM, Followill D, Galvin JM, Hinson W, Kavanagh B et al.
Stereotactic body radiation therapy: The report of AAPM Task Group 101.
Med Phys. 2010;37(8):4078-101.
2. Xiao Y, Papiez L, Paulus R, Timmerman R, Straube WL, Bosch WR et
al. Dosimetric evaluation of heterogeneity corrections for RTOG 0236:
Stereotactic body radiation therapy of inoperable stage I-II non-small-cell
lung cancer. Int J Radiat Oncol Biol Phys. 2009;73(4):1235-42.
3. Armstrong JG, Minsky BD. Radiation therapy for medically inoperable stage
I and II non-small cell lung cancer. Cancer Treat. Rev.1989;16(4):247–55.
4. Nucletron. User manual: Oncentra MasterPlan v. 3.2. Physics and
Algorithms. 4-25.
5. Low DA, Harms WB, Mutic S, Purdy JA. A technique for the quantitative
evaluation of dose distributions. Med. Phys. 1998;25(5):656-61.
147
Artigo Original
Validation of a cylindrical phantom for
verification of radiotherapy treatments in
head and neck with special techniques
Validação de um fantoma cilíndrico para verificação dos
tratamentos radioterápicos na cabeça e no pescoço com
técnicas especiais
Nicolás M. Vargas1, Gustavo Píriz2, Marcia García1 and Niurka Pérez3
Departamento de Ciencias Físicas de la Universidad de La Frontera – Temuco, Chile.
2
Física Médica/Instituto Nacional del Cáncer – Santiago, Chile.
3
QA Radioterapia/Instituto de Salud Pública – Santiago, Chile.
1
Abstract
Verification of radiotherapy treatments in head and neck requires, among other things, small volume chambers and a phantom to reproduce the
geometry and density of the anatomical structure. New documents from the ICRU (International Comission on Radiation Units & Measurements),
Report 83, established the need for quality control in radiotherapy with special techniques such as IMRT (intensity-modulated radiation therapy). In
this study, we built a cylindrical acrylic phantom with standing water, containing seven measuring points in the transverse plane and free location
(0-20 cm) in the longitudinal plane. These points of measurement are constituted by cavities for the accommodation of the ionization chamber of 7
mm of mayor diameter (semiflex, pinpoint with build cup). The results of the phantom validation yielded percentage differences less than 1% in fixed
beams and less than 2.5% in arc therapy for TPS Eclipse calculation. The preparation of this phantom, particularly made to verify the head and neck
treatments, was simple and reliable for checking the dose in radiotherapy with fixed beams and/or special techniques such as arc therapy or IMRT,
so that will be sent to various radiotherapy centers in the country for dosimetric verification in such treatments.
Keywords: quality control, radiotherapy, intensity-modulated radiotherapy.
Resumo
A verificação dos tratamentos de radioterapia na cabeça e no pescoço exige, dentre outras coisas, câmaras de pequeno volume e um fantoma para reproduzir a
geometria e a densidade da estrutura anatômica. Em novos documentos da ICRU (International Comission on Radiation Units & Measurements), Relatório nº 83,
estabelece-se a necessidade de haver controle de qualidade na radioterapia com técnicas especiais, como IMRT (radioterapia de intensidade modulada). Neste
trabalho, foi construído um fantoma cilíndrico de acrílico com água estável, contendo sete pontos de medida no plano transverso e localização livre (0 a 20 cm) no
plano longitudinal. Esses pontos de medida são constituídos por cavidades para acomodar a câmara de ionização de 7 mm de diâmetro maior (Semiflex, Pinpoint
with build cup). Os resultados da validação do fantoma produzem diferenças percentuais menores que 1% em feixes fixos e menores que 2,5% na terapia com
arcos para o cálculo TPS (sistema de planejamento de tratamentos) Eclipse. A preparação desse fantoma, feita particularmente para verificar os tratamentos para
a cabeça e o pescoço, foi simples e confiável na verificação da dose na radioterapia com feixes fixos e/ou técnicas especiais, como terapia com arcos ou IMRT;
portanto, será enviada a diversos centros de radioterapia no país para verificação dosimétrica em tais tratamentos.
Palavras-chave: controle de qualidade, radioterapia, radioterapia de intensidade modulada.
Introduction
Special treatments of radiation therapy should be dosimetric checked to ensure they receive a given dose volume.
To make a dosimetric verification, it is necessary to have a
phantom to reproduce the configuration of the treatment
plan and simulate the structure of the patient. The absolute
or relative dosimetric data used for dose calculations derived from measurements made in water, since it is the
main component of the human body1-5.
In areas with peculiar anatomical geometry, such as
head and neck, standard phantoms (water cube) does not
successfully reproduce this geometry, so the dosimetric
control for a specific patient would not take place under
Corresponding author: Nicolás Morales Vargas – Departamento de Ciencias Físicas de la Universidad de La Frontera – Av. Fco. Salazar, 01145 – Temuco,
Chile – E-mail: [email protected]
149
Vargas NM, Píriz G, García M, Pérez N
similar conditions. For that reason, we recommend using a
phantom that resembles the geometry and density of the
anatomical structure, carrying out measurements closer to
reality1-5.
In this paper, we present the design and validation of a
phantom for verification of radiotherapy treatment in head
and neck. This phantom, with the support of the Institute
of Public Health of Chile (ISP), will move to different radiation therapy centers in the country in order to evaluate and
verify dosimetric treatment of head and neck with different
special techniques.
Given the physical and chemical characteristics, it was
decided to use a water phantom in a container of PMMA
(acrylic), since it has a similar electron density (ρe,agua=5.85
and ρe,acril,=6.6), besides being a user-friendly material, durable and affordable6. Due to the physical characteristics
of PMMA, it has minor differences with the measured dose
in water. These are due to scattering in excess and energy
Ren
absorption mass coefficient
, characteristic of this
ȡ
material. These differences can be corrected by a factor
that reduces excessive scatter (ESC), and another, dp, to
adjust the depth of measurement, making it equivalent to
that of water7,8.
PMMA thickness equivalent to water is determined by
the following equation:
( )
( ).d
()
Z
ȡP A
.
dW =
ȡW
Z
A
P
(1)
P
W
where:
dp and dw are the thickness of the acrylic and water;
ȡP and ȡW are the density of the acrylic and water;
(Z A )
( )
and Z A W are the ratio of average atomic number
and mass of acrylic and water, respectively7.
It is necessary to correct the greater dispersion of
photons produced in this material because of the electron
density of PMMA. This point was raised by Casson9 and
discussed in the Protocol of the AAPM10. The correction
factors for excess dispersion are listed in Table 17,8.
P
Material and methods
The materials needed for construction and validation of
this phantom were:
Construction of phantoms:
• Acrylic cylinders.
• Acrylic tray.
• Rubber stoppers.
• Chloroform.
• Bi-distilled water.
To validate the phantom:
• Helical Tomography Phillips.
• Eclipse Software version 8.1. PBC algorithm.
• PTW ionization chamber Semiflex 0,125 cm3.
• Unidos E electrometer, PTW.
• Varian Linear Accelerator “Clinac 21 iX”.
Phantom design
To determine the diameter of the cylinder, it were reviewed
30 CT scans of head and neck of adult patients, obtaining
values between 12 and 16 cm in diameter neck. Given
this, it was made a PMMA cylinder of 20 cm long by 14 cm
in outer diameter.
For the choice of measurement points, two mutually
perpendicular planes X and Y are considered, a point was
located in the center of the circle in X plane at a distance
1/3 r (r = radius) and at a distance 1/2 r in Y plane; it was
made in order to take points at different depths depending
on the rotation of the phantom (Figure 1).
Validation process
We performed a CT scan of phantom with axial slices 5
mm thick. Phantom axially focused, locating at the center
of it (center of the tube n. 3) in the isocenter, marking the
central reference points denoted by lasers, scans the room
(Figure 2b).
To validate the phantom, numerous measurements were
made based on the Protocol 398 of the IAEA for a fixed
standard field and SSD=100 cm, with the ionization chamber located at different depths allowing phantom, ensuring
that the central beam impinges directly on the camera with
Table 1. Correction factors for excessive scatter for thickness
and field sizes
Energy
5 MeV
Thickness
(cm)
0.4
0.8
1.0
1.5
2.0
3.6
5
5
1.001
1.000
1.000
0.999
0.998
0.996
0.994
Field size
10
0.999
0.999
0.999
0.998
0.997
0.996
0.994
20
0.999
0.998
0.998
0.998
0.998
0.997
0.996
The reference values were taken from Attix7 and interpolated for the thicknesses of acrylic
of the phantom.
150
Figure 1. Front (left) and oblique (right) views of cylindrical phantom NM-1420 Beta, housing Semiflex ionization chamber of 0,125
cc. Measurement sites are horizontally located at a distance of
1/3r and vertically to 1/2r. The holes not used by the ionization
chamber are occupied by acrylic rods 6 mm diameter.
Validation of a cylindrical phantom for verification of radiotherapy treatments in head and neck with special techniques
minimal disturbance of PMMA (3 mm + 1 mm)11,12. These
measurements were made with and without PMMA bars located in different cavities of phantom (Figure 2a).
We used bars of 6 mm diameter to fill the tubes for the
accommodation of the camera and, thus, reduce the shock
generated by a cavity with air when it does not fill the cavity.
Another set of measures was made with the central beam
shining directly into the column and row of cavities for the accommodation of the ionization chamber, placing the latter at
different depths with and without filler bars (Figure 2b).
It was planned a simple, direct and fixed field in the
TPS, based on the configuration of Table 2. The ionization
chamber was located in the tube of interest (1 to 7) with
their effective point in the center of the beam, irradiating
the phantom according to the plan set.
Using the equation 1, it was determined the correction
factor for distance, Cpl. So, the maximum depth at dm and
each measuring point were calculated using the correction
factor Cpl=1,147.
With the measurements obtained at different depths
as cylindrical phantom allows (Figure 1), we determined
empirically the PDD (percentage depth dose) and compared it with the PDD measured in water phantom used
for the TPS (Table 3).
The doses measures were corrected for excessive
scatter and depth. These correction factors were determined for the different thicknesses of acrylic from the
phantom, according to the incidence angle of the central
beam (Table 3).
P7
P1
P2
P3 P4 P5
P6
a
b
Figure 2. a) Front view of the cylindrical phantom with the beam
passes at 315º the angle of the beam varies depending on the
measuring point, so it falls directly on it (P4). b) The beam shining directly into a row and a column of camera holder cavities.
Table 2. Configuration for the cylindrical phantom irradiation
Collimator
0º
Stretcher
SSD
100 cm
270º
300º
Gantry angle
0º
Field size
10x10
MU
50
Energy
6 MeV
0º
341º
321º
308
279.5º
295º
Gantry’s angles were calculated in such a way that the central beam falls directly on the
place where one wants to measure the absorbed dose.
PDD phantom v/s TPS
95
PDD phant
PDD
85
Results
The correction of depth (displacement) varies from 0.5 mm
to 4.9 mm, depending on the location of the measuring
point and the incidence of the beam.
Despite the attenuation caused by PMMA rods, they
strongly enhance the results of the measurements if they
are performed without them. (Figures 3 and 4).
PDD TPS
75
65
55
0,0
100,0
50,0
Depth (mm)
150,0
Figure 3. Measurements at different depths without filling
acrylic rods; the beam incident at 0º and 270º passes through
the various cavities in their way to the point of measurement.
Table 3. Comparison between measured PDDs cylindrical phantom with and without depth correction (Hm/w) and that used by the
TPS measured in water
50 MU
Total thickness =
4 mm of Acrylic
Depth. TPS
Depth. Calc. Corr.
Dose measure no/corr.
PDD phant.
PDD TPS
Diff. %
Fact. depth. Hm/w’
30.0
30.6
46.92
93.47
94.34
0.9
1.010
33.5
34.1
45.87
91.38
92.80
1.5
1.016
47.1
47.7
43.06
85.78
87.17
1.6
1.017
56.4
57.0
41.18
82.02
83.26
1.5
1.016
70.0
70.6
38.45
76.59
77.66
1.4
1.015
86.8
87.4
35.00
69.73
71.09
1.9
1.021
104.1
104.7
31.83
63.40
64.81
2.2
1.024
prom.=
1.016
It shows the depth of measurement (depth TPS), the depth of water equivalent thickness (depth calculated corrected), the PDD and the cylindrical phantom used by TPS, the percentage
difference between them, and the depth correction factor Hm/w’.
151
Vargas NM, Píriz G, García M, Pérez N
Once the PDD of the cylindrical phantom was determined, we obtained the PDDs ratio. So, this depth correction factor (Hm/w’) with excessive scatter factor (ESC)
origin the Hm/w factor, allowing us to determine the absorbed dose at a point within the cylindrical phantom with
a percentage difference less than 1% compared to the
doses calculated by the TPS. (Figure 5 and Table 4).
For the evaluation of treatment with special techniques,
rotational beam hemifields single and double (opposite) at
different angles of rotation were verified. (Tables 5 and 6).
PDD phantom v/s TPS
100
PDD phant
PDD
90
PDD TPS
80
PDD Corr prof.
70
60
0,0
50,0
100,0
Depth (mm)
150,0
Figure 4. Comparison between PDDs used by TPS, measured in
cylindrical phantom with and without depth correction.
Measurements made without acrylic rods at points where
the central beam pass through the cavities presented
greater dosimetric changes due to reduced thickness of
tissue.
The correction factor for excess scatter could be considered negligible because the thickness of acrylic phantom are very low and generate an excess of scatter between 0.07% and 0.45%.
The PDD of the cylindrical phantom, in respect of PDD in
water phantom, showed differences between 1% and 2.3%
without making corresponding corrections; however, considering the corrections by Hm/w’ and ESC, the differences
were less than 0.8%. One average factor Hm/w=1.015
could be used to calculate dose at different depths, and the
results would differ by less than 1%.
Dose phantom v/s TPS
90,0
PDD phant
80,0
PDD
Discussion
PDD TPS
70,0
60,0
50,0
0,0 20,0 40,0 60,0 80,0 100,0 120,0
Depth (mm)
Figure 5. Measurements at different depths with acrylic rods
filling inside the cavities. The central beam directly affects the
spine and/or row of cavities to the point of measurement. Note
the similarity between the PDDs.
Table 4. TPS calculated dose and measured in NM-1420 Beta phantom, corrected for depth (Hm/w’) and excess scatter (ESC)
50 MU
Total thickness =
4 mm of Acrylic
Depth TPS
30.0
33.5
47.1
56.4
70.0
86.8
104.1
Depth calc. corr. Dose meas.
30.5
46.89
34.0
45.84
47.6
43.03
56.9
41.18
70.5
38.45
87.3
35.00
104.6
31.83
ESC
0.999
0.999
0.999
0.999
0.999
0.999
0.999
Hw/m’
1.016
1.016
1.016
1.016
1.016
1.016
1.016
Dose phant.
47.68
46.62
43.76
41.84
39.07
35.57
32.34
Dose TPS
47.4
46.6
43.5
41.8
38.9
35.7
32.5
Diff. %
-0.6
0.0
-0.6
0.1
-0.4
0.4
0.5
Table 5. Absolute percentage difference between calculated dose and measured in TPS cylindrical phantom
Cylindrical phantom
PLAN
80 open
120 open
150 open
180 open
Arc
INI
270
270
270
270
MU
FIN
350
30
60
90
217
223
228
233
Difference %
P1
-0.1
1.5
2.1
1.5
P2
-0.4
0.1
2.1
1.7
P3
-0.1
0.0
-0.2
P4
1.9
P3
-0.8
-0.5
-0.5
-1.1
P4
0.4
0.3
0.1
-0.1
Missing values were not considered because they were in high gradient areas.
Table 6. Differences between calculated and measured doses to do in opposing arcs
Cylindric phantom
PLAN
80 open x2
100 open x2
150 open x2
180 open x2
Arc
INI
0
0
0
0
MU
FIN
80
100
150
180
The P1, P2, P3, and P4 are measuring points (Figure 2a).
152
106
107
112
115
Difference %
P1
0.4
0.6
0.9
0.7
P2
0.1
0.7
0.5
0.3
Validation of a cylindrical phantom for verification of radiotherapy treatments in head and neck with special techniques
It is possible to incorporate such Pinpoint ionization
chamber or micro Pinpoint, adapting PMMA bar to the geometry of each chamber.
Conclusion
The cylindrical phantom NM-1420 Beta turned out to be
very practical, attainable and reliable to realize measurements in structures that resemble this geometry, with percentage differences less than 1% in fixed beams and less
than 2.5% in rotational beams with respect to the TPS
calculation.
It is very useful in verifying head and neck treatments,
with stationary technique, arc therapy, allowing them
to perform control in the various centers of the country
where are made treatments with special techniques like
arc therapy.
Acknowledgment
Thanks to the seminarians of Concepción University,
Medical Physics Unit of Cancer National Institute, Public
Health Institute of Chile and Oncology Service of Gmo.
Grant Benavente Hospital.
References
1. Podgorsak EB. Radiation Oncology Physics: A handbook for teachers and
students. Vienna: IAEA; 2005.
2. Khan FM. Physics of radiation therapy. Philadelphia: Lippincott Williams &
Wilkins; 2010.
3. IAEA - TECDOC-1151. Aspectos físicos de la garantía de calidad en radioterapia:
Protocolo de control de calidad. Vienna: IAEA; 2000.
4. International Comission on Radiation Units & Measurements (ICRU).
Prescribing, recording, and reporting photon beam therapy. Report 50.
Bethesda; 1993.
5. ICRU. Prescribing, recording, and reporting photon-beam intensity-modulated
radiation therapy (IMRT). Report 83. Oxford: Journal of ICRU. 2010:10(1).
6. British Journal of Radiology (BJR), Supplement 25. Central axis depth dose data
for use in radiotherapy. London: The British Institute of Radiology; 1996.
7. Attix FH. Introduction to radiological physics and radiation dosimetry. New
York: John Wiley and Sons; 1986.
8. PTW. Ionizing radiation detectors. PTW, Catalog; 2010.
9. Casson H, Kiley JP. Replacement correction factors for electron measurements
with a parallel-plate chamber. Med Phys. 1987:14; 216-7.
10. AAPM. Task Group 21, American Association of Physicists in Medicine Radiation
Therapy Committee: A protocol for the determination of absorbed dose from
high-energy photon and electron beams. Med Phys. 1983;10:741-71.
11. IAEA. TRS nº 398. Determinación de la dosis absorbida en radioterapia con
haces externos. Vienna: IAEA; 2005.
12. Almond PR, Biggs PJ, Coursey BM, Hanson WF, Huq MS, Nath R et al. AAPM’s
TG-51 protocol for clinical reference dosimetry of high-energy photon and
electron beams. Med Phys. 1999:26;1847-70.
153
Artigo Original
A digital subtraction radiography based tool
for periodontal bone resorption analysis
Uma ferramenta baseada na radiografia de subtração
digital para análise da reabsorção óssea periodontal
Homero Schiabel1, Eveline B. Rodrigues1 and Izabel R. F. Rubira-Bullen2
Department of Electrical Engineering (EESC) of University of São Paulo – São Carlos (SP), Brazil.
2
Bauru Dentistry School of University of São Paulo – Bauru (SP), Brazil.
1
Abstract
The aim of this paper was to describe an aided diagnosis scheme for periodontal bone resorption so that the dentist can make an early diagnosis of
the periodontal disease and establish the best treatment plan to increase the success of healing. Three ways of displaying the results are provided:
qualitative, simple quantitative and colored-percentage quantitative views. A total of 72 pairs of in vitro radiographic images were used. The main
procedure registers the images perspective projection aimed to align them in rotation and translation, and is followed by the application of a contrast
correction technique. The results from the subtraction were evaluated firstly by the comparison between the actual and the digital sizes corresponding
to the holes made by drills in phantoms. The mean error was 4.2%. The method was also applied to actual tooth radiographic images and could
detect clearly the effect of treatment of periodontal diseases. It is dependent on the reproducibility of the process of radiographs acquisition and
digitization, but the calculated mean error allows to conclude its better efficacy compared to usual procedures in this field.
Keywords: computer-assisted image processing, radiography, periodontics, periodontal disease.
Resumo
O objetivo deste trabalho foi descrever um esquema de auxílio ao diagnóstico da reabsorção óssea periodontal para que o dentista possa realizar um
diagnóstico prévio da doença periodontal e estabelecer o melhor plano de tratamento, de forma a melhorar o sucesso da cura. São fornecidos três
modos de exibição dos resultados: visualização qualitativa, quantitativa simples e quantitativa com porcentagem colorida. Um total de 72 pares de
imagens radiográficas in vitro foi utilizado. O principal procedimento registra a projeção perspectiva das imagens com o objetivo de alinhá-las em
rotação e translação, seguido da aplicação de uma técnica de correção com contraste. Os resultados da subtração foram primeiramente avaliados
pela comparação entre o tamanho atual e o digital, correspondentes aos buracos feitos pelas brocas nos simuladores radiográficos. O erro médio foi
de 4,2%. O método também foi aplicado para as imagens radiográficas atuais e foi possível claramente detectar o efeito do tratamento das doenças
periodontais. Tal fato depende da reprodutibilidade do processo de aquisição e digitalização das radiografias, mas o erro médio calculado permite
concluir sua melhor eficácia comparada aos procedimentos usuais neste campo.
Palavras-chave: processamento de imagem assistida por computador, radiografia, periodontia, doença periodontal.
Introduction
The two most important causes of tooth loss are cavities and periodontal diseases. A large percentage of the
adult population is reached by periodontal diseases that,
due to their chronic and aggressive character, become
the most important cause of dental extraction. The dental radiographic examination complements the clinical
examination. Bone resorption diagnosis is a difficult task
to be performed by dentists since it is only radiographic
visible when 30% to 60% of the mineral bone content
have been lost1.
The Digital subtraction technique (DSR) is applied to
solve the limitation in detecting radiographic changes. In
DSR, two digital radiographic images taken at different
times are subtracted so that common structures in both
images (structured noise) are removed and an image remains consisting only of differences (signal) between them.
Studies have shown that dentists using the digital subtraction radiography can detect alveolar bone changes
from 1% to 5% per unit volume2, since both images are
acquired under the same conditions of geometric projection, X-ray tube operational conditions and film developing.
With these techniques, even subtle changes in the bone
Corresponding author: Homero Schiabel – Departamento de Engenharia Elétrica, Escola de Engenharia de São Carlos, Universidade de São Paulo –
Av. Trabalhador São-carlense, 400 – Centro – CEP: 13566-590 – São Carlos (SP), Brasil – E-mail: [email protected]
155
Schiabel H, Rodrigues EB, Rubira-Bullen IRF
or dental tissue which are not perceptible for direct films
comparison can be revealed3.
Differences between two radiographic images after digital subtraction can be caused mainly by 3 factors: (a) local
anatomical deformations due to progression or regression
of the disease; (b) geometric changes due to projection
errors; (c) intensity changes due to exposure and different
parameters of film processing4. Anatomical differences
can be identified if the two other factors are eliminated.
Therefore, the geometric projection and the contrast regarding digital radiographic images to be subtracted must
be standardized.
Considering the first feature above, external mechanical devices have been used to reduce possible errors of
projection5,6. Image processing techniques allow the sequential reconstruction of images’ geometry and their contrast correction before the subtraction7. Many researchers
use manual registration of perspective projection by methods of selecting reference points marked on the image
by experienced specialists. In these methods, a geometric
transformation model lines up the images according to
the measurements of difference and correcting distortions
in geometry. There are semi-automatic methods of perspective projection registration, with manually selection of
reference points only on the first image and a computational algorithm automatically determining the corresponding
points in the second one8. Techniques of automatic registration perspective projection have been introduced into
the literature7,9.
This work rejoins the useful functionalities and tools for
the evaluation of bone resorptions due to periodontal diseases by the DSR technique. Therefore, a perspective projection registration method which corrects the geometric
projection and another correcting the contrast were developed. It is based on the identification of typical ‘edges’ in
dental images to allow adjusting the geometric projection.
The qualitative and quantitative subtractions help the dentist to locate the area and the type of lesion. Besides, a
quantitative colored–percentage subtraction is a new approach to make the result more quantitative, since in many
clinical situations the dentist needs to estimate exactly the
bone gain or loss.
Methodology
In vitro radiographs were obtained from a dry jaw perforated by three different drills (with diameters of 2.9, 1.7 and
1.3 mm). Four images were obtained for each exposure
time (0.17, 0.20 and 0.23 s). The first did not correspond
to a perforated sample, simulating inexistence of a lesion
(A); the second corresponded to the sample perforated by
the drill of 1.3 mm (B); the third was perforated by the drill
of 1.7 mm (C); and the fourth was perforated by a drill of
2.9 mm (D). A total of 72 pairs of images were obtained.
For comparisons, they were organized into two sets, each
one with 12 images, 4 for each exposure time (0.17, 0.20
156
and 0.23 s). In the first set, the perforation in the dry jaw
was made on the vestibular and lingual cortical, and in the
second set, the perforation in the jaw was only made on
the vestibular cortical. These perforations were performed
to check our procedure performance; therefore, they simulated bone resorption or bone gain depending on how they
were compared.
The images were compared with themselves relatively
to each exposure time and each set. They were all acquired from an X-707 radiographic system (Dental Yoshida
MFC Co. Ltd., Tokio, Japan), with 70 kVp and 7mA. Their
positioning was standardized to provide a constant ratio
for the distances among the X-ray tube, the film and the
object to be exposed. Supports are allowed to keep the
dry jaw always in the same location, as well as the distance between the film and the object constant. All the images were digitized by a UMAX PowerLook 1120 scanner
(resolutions of 600 dpi and 14 bits).
In order to perform the digital subtraction, the following procedures are achieved: (a) digitization of both
radiograms to be compared, (b) alignment of the subsequent images and (c) contrast adjustment.
Discrepancies in contrast can occur in practice, mainly due to variations in a series of factors involved in the
radiographic image acquisition. In order to compensate
such variations, the method compares images obtained
by using a same aluminum stepwedge attached to the
film during the X-ray exposure. The stepwedge has 4
steps so that the areas under each one are exposed to
different X-ray intensities. A comparison is performed between the average intensity in the areas corresponding
to those 4 steps in each image. Thus, the dentist must
select 3 circular regions, one in each step in both images,
in order to calculate the average of gray level of those
areas. The contrast of the subsequent image is increased
or decreased according to the difference in the percentage value.
For the alignment, one image can be translated or rotated relatively to the other; therefore, the common structures to both images should be lined up in such a way
that they are represented by the same pixels addresses.
This procedure is performed by a manual selection of four
reference points chosen by the dentist. The reference
points are anatomical marks clearly visible in both images
representing high contrast regions. The first (“pivot”) and
the three others are subsequent points. They must be
marked in equal regions of high contrast in both images.
The pivot point is a reference for the translation accomplishment and also a basis for the rotation calculation.
All points are firstly selected in the reference image and,
later, in the subsequent image.
To facilitate and improve the selecting process of these reference points by the dentist, some aid tools were
incorporated, as a magnifying glass, which increases the
visual information when selecting the points. In order to
eliminate the salt and pepper noise caused by the coordinates recalculation by means of rotation, the median filter
A digital subtraction radiography based tool for periodontal bone resorption analysis
is applied to the rotated image, but only in pixels that did
not receive any value from the rotation function. After the
registration of perspective projection and the correction
of resultant rotation artifacts in the subsequent image,
the images can be subtracted. Three types of subtraction
were implemented: qualitative, quantitative and quantitative colored-percentage. The latter is a new approach to
show the information in the subtracted image, as many
times the dentist needs to quantify the result.
The scheme performs the difference pixel by pixel of
two images, producing a subtracted image in that each
pixel represents the difference relatively to the corresponding pixels in the original images. In the qualitative
subtraction pixels corresponding to the positions, where
the two images are different will be white and they will
be black where they are equal (Figure 1). This type of
subtraction can be useful to visually determine the bone
resorption due to the periodontal disease, as it can immediately display the pixels affected.
In the quantitative subtraction, pixels are added up to
the mean value of the image’s total intensity (for instance, in an 8-bit image the value of the subtraction will be
added up to 128; for a 12-bit image, these pixels will be
added up to 2048). Thus, if the two pixels are the same,
the resultant image will be homogeneous, while darker
pixels in the first image will be also darker in the resultant
image, making any change of overlapping or the disease
more easily visible (Figure 2).
In the quantitative percent-colored subtraction, the
percentage of bone loss or gain is displayed by colors in
(a)
(b)
Figure 1. Qualitative subtraction: (a) with bone loss; (b) with
bone gain.
the subsequent image. This is made by calculating the
percentage of the corresponding pixels variation in the
original images. This variation can be configured by the
dentist selecting the values ranges in percentage. Thus,
the subsequent image is colored, accordingly with the
percentages and colors chosen by the dentist. An interval of positive or negative percentage values will show,
respectively, the gain or loss registered in the subsequent image relatively to the reference image. Figure
3(a) displays the medullary bone slightly wasted by the
dentists using drills, and Figure 3(b) shows an increase
in the same wasted area. Figure 3(c) shows the subtraction of Figures 3(a) and 3(b). Figure 3(d) shows the
quantitative percentage–colored subtraction, with the
difference in a selected interval of -10 to 0%, i.e., 10%
of bone loss, in yellow.
Results
The registration of perspective projection was firstly tested with different digitized images from test-objects. In
the second case, the object was slightly translated and
rotated relatively to the first. A pair of images was subtracted before making the subsequent image’s registration of
perspective projection. The result was an overlapping of
both objects, resulting in an image full of noise and inadequate for use. Later, the subsequent image was lined
up and subtracted from the reference image. The result
was suitable: the subtraction has provided an image with
(a)
(b)
Figure 2. Quantitative subtraction: (a) with bone gain; (b) with
bone loss.
157
Schiabel H, Rodrigues EB, Rubira-Bullen IRF
158
(a)
(b)
(c)
(d)
!
Figure 3. (a) Reference image slightly wasted; (b) subsequent
image with increased waste; (c) quantitative subtraction; (d) respective percentage-colored subtraction showing 10% of bone
loss in yellow.
Reference
image
Subsequent
image
Subtraction before Subtraction after the
the registration of
registration of
perspective
perspective
projection
projection
Figure 4. Tests performed for the subsequent image registration
of projection perspective.
Subtraction result
homogeneous intensity, since both digitized images were
equal and were not modified. After verifying the images
alignment, tests were performed with radiographic dental
images, as seen in Figure 4.
The drills’ holes diameters in the radiograms were measured by an Olympus BHA digital projection microscope, with ±0.5 mm of tolerance. Although the diameter in
the radiographic image has been increased compared to
that in the object (dry jaw), the result in the subtraction is
not affected, since the comparisons were made between
images from the same set. Furthermore, errors due to the
images digitization can be neglected since, if any variation
arises during this process, both images to be subtracted
are digitized by the same equipment and will present the
same variations.
In order to check effectively the result from the digital
subtraction scheme, a comparison was made between
the desired area, resultant from a digital subtraction of two
images, and the measured area. For such analysis, an algorithm was developed to count the pixels located in the
area corresponding to the hole. The number of pixels in
such a region in the reference image, as well as the ones
of the subsequent image, was counted. Later, the subtraction was performed and the remaining pixels were counted
again. The resultant pixels of the subtraction process were
compared with the desired values. The error of the difference between the desired and the obtained values was
calculated in percentage.
For the two sets of radiographic images (a total of 72
pairs), the average error was 7.7%. The value of this error
also includes a test performed by the dentist to verify the
variability in the manual method of marking points in the
registration of perspective projection. This rate represents
the error of all processes involved in the subtraction, i.e.,
the acquisition, digitization, registration of perspective projection and subtraction. The graph in Figure 5 shows the
relation between the desired values (dashed lines) and the
measured values in the subtraction (full lines) by the digital
subtraction scheme.
The hole diameter in the radiographic image was also
calculated and compared with the corresponding diameter in the digital image to found the error induced by the
digitizer. The mean difference between them was 3.5%.
The difference found between the diameter in the radiography and in the digital image was 3.5%, meaning that
this is the error induced by the digitizer. This error can be
deduced from the subtraction error found of 7.7%, lasting
only 4.2% including the contrast correction and registration of perspective projection errors. For this current work,
the percentage of the system inaccuracy is still very low
and much more efficient than the clinical analysis, which
can detect only a change of bone mineralization between
30 and 60%.
Altogether, 88 pairs of actual teeth radiographic images were subtracted. Previous Figures 3 and 4 have already illustrated results of the DSR, applied to actual patients
radiographic images.
6000
5000
4000
3000
2000
1000
0
0
Desired
Obtained
10
20
30
40
50
Number of images
60
70
80
!
Figure 5. Relation between desired and measured results in the
subtraction by the digital subtraction scheme.
Conclusions
The subtraction scheme proposed here is dependent on
the whole image acquisition process (imaging, film development and digitization). The imaging process must be
totally reproducible later. Thus, the same X-ray system and
A digital subtraction radiography based tool for periodontal bone resorption analysis
operational conditions should be used. A positioning standardization method providing a constant relation regarding
the distances between the X-rays tube, film and object to
be exposed is also recommended to be used in order to
assure that there are not great distortions relatively to the
geometric projection. Images digitization should keep the
reproducibility of parameters (same spatial and contrast
resolutions and noise level). It assures that both images to
be subtracted will be obtained and digitized with the same
characteristics so that such parameters will not affect the
subtraction final result.
The DSR is primarily an operation to improve the visual information, and the role of the dentist in detecting the
signal is still important. Moreover, the quantitative percentage–colored subtraction is a new approach of the subtraction result to make it more quantitative, since in many
clinical situations the dentist needs to evaluate the amount
of bone gain or loss. Thus, the dentist will be able to visualize by the colored areas in these percentages. This is an
improvement compared to other works in literature, since
it is not pseudo-color, but another way of registering quantitatively the bone gain or loss.
In many clinical situations requiring digital subtraction,
there is no specific system, and dentists are likely to use
generic image processing programs. Therefore, this procedure can work as an aid diagnosis tool and it will be
useful to the dentists for the evaluation of bone resorption
due to periodontal disease, contributing for an early diagnosis to establish the best treatment and to increase its
possibilities of success. This program is currently available
for free download at http://lapimo.sel.eesc.usp.br/lapimo/
LapimOdonto.htm.
References
1. Kornman KS. Nature of periodontal diseases: assessment and diagnosis. J
Periodontal Res. 1987;22(3):192-204.
2. Dove SB, McDavid WD, Hamilton KE. Analysis of sensitivity and specificity of
a new digital subtraction system: an in vitro study. Oral Surg Oral Med Oral
Radiol Endod. 2000;89:771-6.
3. Likar B, Pernus F. Evaluation of three contrast correction methods for
digital subtraction in dental radiography: an in vitro study. Med. Phys.
1997;24(2):229-307.
4. Papika S, Paulsen HU, Shi XQ, Welander U, Linder-Aronson S. Orthodontic
application of color image addition to visualize differences between
sequential radiographs. Am J Orthod Dentofacial Orthop. 1999;115(5):48893.
5. Burdea GC, Dunn SM, Immendorf CH, Mallik M. Real time sensing of tooth
position for dental digital subtraction radiography. IEEE Trans Biomed Eng.
1991;38(4):366-78.
6. Burdea GC, Dunn SM, Levy G. Evaluation of robot-based registration for
subtraction radiography. Med. Image Anal. 1999;3(3):265-74.
7. Zacharaki EI, Matsopoulos GK, Asvestas PA, Nikita KS, Gröndahl K,
Gröndahl HG. A digital subtraction radiography scheme based on automatic
multiresolution registration. Dentomaxillofac Radiol. 2004;33(6):379-90.
8. Yoon DC. A new method for the automated alignment of dental
radiographs for digital subtraction radiography. Dentomaxillofac Radiol.
2000;29(1):11-9.
9. Byrd V, Mayfield-Donahoo T, Reddy MS, Jeffcoat MK. Semiautomated
image registration for digital subtraction radiography. Oral Surg Oral Med
Oral Pathol Oral Radiol Endod. 1998;85(4):473-8.
159
Artigo Original
Gastric assessment by images processing
of ultrasound in LabVIEW platform:
preliminary results
Avaliação gástrica pelo processamento de imagens de
ultrassom na plataforma LabVIEW: resultados preliminares
T. Córdova1,2, M. Sosa1, J. J. Bernal1, A. Hernandez1, G. D. Gutiérrez1, D. Rodriguez1, S. Solorio3,
M. A. Hernandez3, M. Vargas1, I. Delgadillo1, G. Moreno1, J. G. Villalpando2 and C. R. Contreras2
Departamento de Ingeniería Física, Universidad de Guanajuato, Campus León – México.
Facultad de Ingeniería en Computación y Electrónica, Universidad De La Salle Bajío – México.
3
Unidad Médica de Alta Especialidad, Clínica T1-León, Instituto Mexicano del Seguro Social – México.
1
2
Abstract
Nowadays, the gold technique in gastric evaluations still is scintigraphy in spite of ionization radiation dose per patient undergoing this procedure.
Gastro images with ultrasound technique are controversial, because the stomach is a hollow cavity filled with gas in basal conditions or in fast state.
Fortunately, a stomach with food is recommended in gastric motility and gastric emptying assessment. So, a lack of air in stomach contributes in this
kind of study and in recordings of excellent images by ultrasound. In this study, a digital image processing of gastric ultrasound is presented. Whole
automated routine and implemented filters are described in order to use this procedure in gastric peristalsis and gastric emptying evaluations. Ten
volunteers were recruited and required to attend the measurements about dominant frequency, with values of at least 3 cpm. Although the behavior
stomach activity is observed in dynamic graph, an analysis in frequency space is performed.
Keywords: gastric, peristalsis, ultrasound, emptying, LabVIEW.
Resumo
Hoje em dia, a técnica de ouro nas avaliações gástricas ainda é a cintilografia apesar da dose de radiação de ionização por paciente submetido a esse
procedimento. As imagens gástricas com a técnica do ultrassom são controversas, pois o estômago é uma cavidade oca preenchida com gás em
condições basais ou em estado rápido. Felizmente, um estômago com comida é recomendado em motilidade gástrica, como também na avaliação do
esvaziamento gástrico. Portanto, falta de ar no estômago contribui para este tipo de estudo e para gravações de excelentes imagens por ultrassom.
Neste estudo, o processamento da imagem digital do ultrassom gástrico é apresentado. Uma rotina totalmente automatizada e filtros implementados
estão descritos para usar este procedimento no peristaltismo gástrico e nas avaliações de esvaziamento gástrico. Dez voluntários foram avaliados
em relação à frequência dominante, com valores de no mínimo 3 cpm. Embora a atividade estomacal comportamental seja observada em gráfico de
dinâmica, uma análise de frequência espacial é realizada.
Palavras-chave: gástrico, peristaltismo, ultrassonografia, esvaziamento, LabVIEW.
Introduction
The gastrointestinal system evaluation is, currently, as
important as other clinical procedures, like heart monitoring. If patients are not adequately treated, they may
die. This is especially important for some kinds of patients, for instance, diabetes patients with problems
of gastroparesis. The scintigraphy technique is now
the gold standard in this evaluation, despite ionizing
radiations that undergo the persons1. There are other
imaging techniques for this study, like the ultrasound,
that has been an alternative for assessment and monitoring of the gastric activity in the last years2-6, although
it has still not taken off, which could be due to a lack
of conclusive results and proper procedure, leading to
results highly correlated with the gold standard technique, scintigraphy.
A routine for processing ultrasound images of
stomach, implemented in LabVIEW platform, is presented. This has been used to perform evaluations of
Corresponding author: Teodoro Córdova – Departamento de Ingeniería Física – DCI, Universidad de Guanajuato, campus León – Loma del Bosque, 103 –
Lomas del Campestre – León, GTO, Mexico – E-mail: [email protected]
161
Córdova T, Sosa M, Bernal JJ, Hernandez A, Gutiérrez GD, Rodriguez D, Solorio S, Hernandez MA, Vargas M, Delgadillo I, Moreno G, Villalpando JG, Contreras CR
the peristaltic activity and gastric emptying in preprandial and postprandial conditions in healthy subjects
and patients.
Procedure
The above description is appropriated for gastro evaluations. The peristaltic information is seen from first measurement (Figure 7) in time domain, in which about three
contractions are shown. Nevertheless, physicians are
interested in the exact value of the dominant frequency,
An ultrasound equipment model Medison SONOACE
8000 SE was used in this study. A group of ten healthy
volunteers with no history of gastrointestinal diseases
and two patients were recruited and required to attend
the measurements after a night of fast. Each subject
swallowed 300 mL of water previous to first measurement; then, a solid test meal was ingested in order to
estimate the gastric area. Each subject was in supine
position during the auscultation and along the data acquisition time, that lasted one minute. It is important to
point out that this work was carried out according to
the Helsinki agreement for studies in human.
Protocol
Once the gastric area was identified, a video of 1740
frames was recorded in B Mode and in M modes (Figure
1). Then, the M mode video was split in each one of its
frames and the image processing was performed.
From the M mode video, a representative frame is
selected and the region of interest (RI) is studied (Figure
2); the superior and inferior bands are identified. In the
RI of the frame, the points x1,y1, x2 and y2 are determined and the same action is automatically executed
in each of the 1740 frames. Simultaneously, a new file
(Figures 2 and 3) of this new outage image is created in
order to perform the imaging processing over all of it.
Imaging processing
The upper and lower bands, in the image of the Figure
4, correspond to the upper and lower walls of the
stomach, respectively.
Here, a series of filters were implemented during image processing in order to reduce the area of
each band to a single line (IMAQ MathLookup +
IMAQ GrayMorphology + IMAQ MathLookup + IMAQ
BCGLookup + IMAQ LowPass + IMAQ Convolute +
IMAQ RejectBorder). When the filtering process is over,
one distance from inferior to superior band is measured through a subroutine of find Vertical Edge of the
IMAQ sofware; this leads to determine only one couple
of coordinates, selected from each one of the edge
points of the band. With that couple of coordinates,
the distance from one to other side of the stomach is
measured (Figure 5).
It is important to point out that only one distance
is measured from each couple of coordinates. If we
have 29 frames in one second, then, this corresponds
to have a rate frequency of 29 samples per second.
These points are simultaneously plotted (Figure 6).
162
Figure 1. A frame showing the B and M mode.
Figure 2. Representative frame.
Figure 3. LabVIEW Screen for the RI.
Gastric assessment by images processing of ultrasound in LabVIEW platform: preliminary results
x 107
Power spectral density [a.u.]
3.5
3
2.5
2
1.5
1
0.5
0
-0.5
Figure 4. Outage image of the represented frame.
Dominant Frequency 2.46 cpm
2
4
6
8 10 12 14
Frequency [cpm]
16
18
20
Figure 8. PSD of the gastric mechanical activity.
so a FFT (fast Fourier transform) is obtained from above
signal in order to present this value (Figure 8).
Figure 5. The measurement of the stomach wall can be obtained from de filtering of the left image.
Conclusions
This new ultrasound gastric imaging processing is presented as a powerful technique, with relatively easy
implementation, because the ultrasound equipments
are very common in hospitals and the LabVIEW routine
can be executed in any personal computer.
Acknowledgment
Figure 6. The left figure shows the superior and inferior band; the
right figure presents the square and gastric activity along the time.
Authors want to thank DAIP grant No. 000017/10 and
UDLSB grant of 2010.
Amplitude [a.u.]
References
25
1.
20
2.
15
3.
10
4.
5
0
5.
0
10
20
30
Time [Sec]
Figure 7. Gastric mechanical activity.
40
50
60
6.
Odunsi ST, Camilleri M. Selected interventions in nuclear medicine:
gastrointestinal motor functions. Semin Nucl Med. 1999;39(3):186-94.
Kusunoki H, Haruma K, Hata J, Ishii M, Kamada T, Yamashita N, et al.
Efficacy of Rikkunshito, a traditional japanese medicine (Kampo), in
treating functional dyspepsia. Inter Med. 2010;49(20):2195-202.
Newell SJ, Chapman S, Booth IW. Ultrasonic assessment of gastric
emptying in preterm infant. Arch Dis Child. 1993;69(1):32-6.
Gilja OH, Hausken T, Degaard S, Berstad A. Gastric emptying measured
by ultrasonography. World J Gastroenterol. 1999;5(2):93-4.
Bateman DN, Whittingham TA. Measurements of gastric empyting by
real-time ultrasound. Gut. 1982;23:524-7.
Holt S, McDicken WN, Anderson T, Stewart IC, Heading RC. Dynamic
imaging of the stomach by real-time ultrasound–a method for study
of gastric motility. Gut. 1980;21(7):597-601.
163
Artigo Original
Daily quality control in computed
radiography mammography using
the manufacturer phantom
Controle de qualidade diário em mamografia radiográfica
computadorizada usando o simulador do fabricante
Rosangela R. Jakubiak, Pricila C. Messias and Carlla M. Oliveira
Technological Federal University of Paraná (UTFPR), Academic Physics Department – Curitiba (PR), Brazil.
Abstract
The quality control (QC) in mammography system involves a large amount of test tools, which implies a large space for storage and a high number
of exposure. This work describes a QC system using a phantom, Fuji Computed Radiography (FCR) One Shot Phantom M Plus, that evaluates several
parameters with just one exposure. The software offers tests with annual, semi-annual, quarterly, weekly and daily periodicity, and analyzes the
conformities of the mammography equipment, image plate and cassettes. Because of the high number of tests, it was evaluated the daily test only
for seven months in two mammography equipments. The test, through the software and its image, allows the analysis of ten parameters in QC. The
evaluation of these parameters was realized by the average of the values provided by the software. Only one of the evaluated items showed not
conformity, but this was observed and the necessary corrections were realized. The monitoring of use of FCR Mammography QC software with the
FCR One Shot Phantom M Plus was realized and through this we could investigate that the quality program provided by the system is appropriate for
the radiology services that has the Fuji Computed Radiography system.
Keywords: mammography, quality control, computed radiography.
Resumo
O controle de qualidade no sistema de mamografia envolve uma grande quantidade de ferramentas em teste, o que sugere um amplo espaço
para armazenamento e um alto número de exposição. Este trabalho descreve um sistema de controle de qualidade utilizando um simulador para
Radiografia Computadorizada Fuji, One Shot Phantom M Plus, que avalia diversos parâmetros com apenas uma exposição. O software oferece
testes de periodicidade anual, semianual, trimestral, semanal e diária, e analisa as conformidades do equipamento de mamografia, da chapa de
imagem e dos cassetes. Por causa do elevado número de testes, avaliou-se o teste diário somente por sete meses em dois equipamentos de
mamografia. O teste, por meio do software e de sua imagem, permite a análise de dez parâmetros no controle de qualidade. A avaliação de tais
parâmetros foi realizada pela média dos valores fornecidos pelo software. Somente um dos itens avaliados mostrou não conformidade, mas isso
foi acompanhado, sendo realizadas as correções necessárias. O monitoramento do uso do software de controle de qualidade da mamografia
de radiografia computadorizada da Fuji com o One Shot Phantom M Plus foi realizado e, por meio dele, foi possível observar que o programa de
qualidade fornecido pelo sistema é adequado para serviços radiológicos que possuem o sistema de radiografia computadorizada da Fuji.
Palavras-chave: mamografia, controle de qualidade, radiografia computadorizada.
Introduction
Breast cancer is the second most frequent type in the world
and the most common among women, corresponding to
the 22% of the new cases per year. When it is diagnosed
and treated properly, the prognostic is relatively good1. To
obtain images that allow a reliable diagnostic, the radio diagnostic services must submit a guarantee quality program
for their equipments, containing tests and measurements2.
It has been habitual in private services the installation
in Brazil of digital radiography (DR) and computed radiography (CR) systems for mammography3. These technologies require new quality standards and test procedures
specifically for digital systems4.
Considering the installation of CR systems, originally
called digital radiography with photostimulable phosphor
(PSP)5, we recommend to check the proper functioning of
the system according to the manufacturer specification.
Corresponding author: Rosangela Requi Jakubiak – Universidade Tecnológica Federal do Paraná (UTFPR) – Av. Sete de Setembro, 3165 – CEP: 80230-901 –
Curitiba (PR), Brazil – E-mail: [email protected]
165
Jakubiak RR, Messias PC, Oliveira CM
The manufacturer should provide a quality control phantom and evaluation program with the PSP system6.
The Brazilian College of Radiology approved the phantom called FCR One Shot Phantom M Plus with Fujifilm
FCR Mammography QC Software, based on international
standards3. In this work, we’ll evaluate the daily test results
using the phantom and, through them, analyze the application in the practices realized and in the image quality.
Two mammography equipments Lorad Affinity-Hologic
were used in this study, realized at the Advanced Diagnostic
Imaging Clinic (DAPI) located in Curitiba, in state of Paraná,
which were provided with Fujifilm Computed Radiography
system. As additional tests of daily routine for quality control of these equipments, it was adopted the FCR
Mammography QC Software system associated to One
Shot Phantom M Plus. The software provides the realization of tests with different frequencies (annual, semi-annual, quarterly, weekly and daily). The results obtained are acquired through calculations performed by the software each
time that their images are registered and filed. Because of
the large number of tests, it will be presented just the evaluation of the daily test in a limited period between January
and July of 2010, a total of seven months7.
The software allows just one diary exposure. For this
reason, the survey was accomplished using the phantom
in alternate days in each equipment of mammography.
These equipments will be identified as the equipments
Room 1 and Room 2. The reports are manually issued,
because the software does not differentiate the data of
both equipments.
FCR Mammography QC Software
The software has as function to carry out tests based on
a quality control program and in this way to manage the
quality of the FCR Mammography system. The tests can
be used for evaluation of the mammography equipment,
cassettes, imaging plate (IP) and image reader7, that compose the system.
Daily test
The daily test is performed to evaluate the image quality,
with one exposure only, using the FCR One Shot Phantom
M Plus, and to check if the X-ray equipment, IP’s and cassettes used in clinic practice are according to this7. For this
test, it is used an exclusive cassette to the quality control.
After the exposure, are included, with the acquired image,
ten quality control parameters, specified in Figure 1, which
show a diagram of FCR One Shot Phantom M Plus.
To perform the exposure, the cassette is inserted and
the FCR One Shot Phantom M Plus is positioned under the
bucky of the mammographic equipment. The compression tray should be four centimeters from the bucky. The
baseline values are followed as a reference, according to
166
the stipulated through the installation test. For the baseline
survey, were performed two exposure with 28 kVp in the
mode auto time, the second indication from the manufacturer7. The technique achieved is of 28 kVp and 60 mAs,
the target/filter of Mo/Mo is selected and the photocell
must be in the position 1. The survey considered the two
mammographic equipments.
In the image, showed by Figure 2, are analized the
patterns: missed tissue at the chest wall edge, in which
must appear at least three bars; of geometric distortion,
in which the checkered pattern visualized on the images
edges should not indicate distortion, therefore all the lines
must appear straight and with an aspect off grid; of uniformity, in which must appear four circles in each corner
of the image, without cutting, and, if it happens, is indicative that the FCR One Shot Phantom M Plus is not correct
centralized and, so, a new exposure must be realized; the
image must not present artifacts7.
Figure 1. Diagram of FCR One Shot Phantom M Plus8. The
corresponding components to diagram and their respective
parameters analyzed are: 1. Missed tissue at chest wall edge
(right and left); 2. Contrast to noise ratio; 3. One Shot Phantom
sensitivity constancy; 4.Geometric distortion; 5. Artifacts; 6. Uniformity; 7. Dynamic range; 8. Spatial resolution; 9.Low contrast
detectability; 10. Linearity/Beam quality constancy.
Figure 2. Image acquired after the exposure using the FCR One
Shot M Plus.
Daily quality control in computed radiography mammography using the manufacturer phantom
Results
With the realization of the test using the FCR One Shot
Phantom M Plus in equipment, it was possible analyzing the
average values of each month for each parameter. The values obtained directly reflect the mammography image quality in the practice of the examinations realized in the room,
and in the condition of the equipment. The data of the respective reference of Room 1 and Room 2 follow in Table 1.
The values obtained in the equipments of the Room 1 and
Room 2 will be shown in Table 2 and 3, respectively.
From the data above, we can observe that the results
are within the upper and lower limits, therefore are in accordance with the recommendation of the manufacturer
and in agreement with the standard EUREF (European
Reference Organization for Quality Assured Breast
Screening and Diagnostic Services)9. During some days,
the linearity/beam QL in steps 3-4 and 4-5 presented no
conformity in both rooms, with values below the lower
limit, indicating bad positioning of the detector in relation to the sensor, but the problem was corrected putting centralized markers in the bucky. When it was found
no conformity of any criterion, surpassing the upper and
lower limit stipulated, a request of maintenance of equipments is indicated.
Conclusions
The application of FCR Mammography QC system with
the FCR One Shot Phantom M Plus occurs quickly and
safely. The results of several quality standards are achieved
and accompanied with just one daily exposure. The quality control through the analysis of the test and its image
demonstrates the conformity of image standards and the
system analyzed, as well as clinical practice. The results
obtained with One Shot Phantom M Plus, when compared
to others achieved with different test tools, are corresponding. The equipment in study presents reliable answers and
its implementation must be strongly encouraged, because
identifies the conformity standards of several parameters
with just one exposure. The tests with this type of tool
do not replace the quality control tests recommended by
other reports adopted and carried out with independent
equipment manufacturer, but the cost of the system is justified by the benefits.
Table 1. Reference data from Room 1 and Room 2
Content (C)
1. Missed tissue at chest wall edge
2. CNR
3. Sensitivity
4. Geometric distortion [mm]
6. Uniformity
7. SNR ratio (Bottom Left)
8. Spatial resolution [%]
9. Low contrast detectability
10. Linearity/
Beam quality constancy
Parameters (Par)
Lower limit
1.1 Thoracic edge - right
5
1.2 Thoracic edge - left
5
2.1 CNR
8.634
Upperlimit
Baseline
12.952
10.79
3.1 System sensitivity
72
150
111
4.1 Dimension (Scan direction)
101.6
105.8
103.7
4.1 Dimension (Scan direction)
101.8
106
103.9
6.1 Pixel value ratio (Top Right )
-38.05
-8.05
-23.1
6.2 Pixel value ratio (Top Left )
-34.48
-4.48
-19.5
6.3 Pixel value ratio (Bottow Right )
-19.35
10.65
-4.35
6.4 Pixel value ratio (Bottow Left )
-17.15
12.85
-2.15
6.5 SNR ratio (Top Right)
-27.89
2.11
-12.9
6.6 SNR ratio (Top Left)
-27.83
2.17
-12.8
6.7 SNR ratio (Bottom Right)
-18.6
11.4
-3.6
6.8 SNR ratio (Bottom Left)
-19.72
10.28
-4.72
7.1 Average QL at thinnest step wedge
3175
3375
3275
8.1 2 lp/mm
50.23
56.65
53.44
8.2 4 lp/mm
18.27
24.71
21.49
9.1 Light [%]
40
9.2 Dark [%]
40
10.1 QL gap step 1-2
676
716
696
508
548
528
790
830
810
722
762
742
167
Jakubiak RR, Messias PC, Oliveira CM
Table 2. Results obtained on equipment from Room 1
Mean values
C
Par
1.1
1.
1.2
2.
2.1
3.
3.1
4.1
4.
4.2
5.1
5.2
5.3
5.4
6.
5.5
5.6
5.7
5.8
7.
6.1
7.1
8.
7.2
8.1
9.
8.2
8.3
8.4
10.
8.5
8.6
Jan
Feb
Mar
Apr
May
Jun
Jul
4.3
4.8
10.42
114.5
103.7
103.9
-19.93
-20.03
-3.8
-4.7
-11.17
-13.08
-3.3
-5.3
3260
53.78
21.54
66.24
65.31
691.1
520
803.9
737
4.3
4.7
10.37
116.8
103.7
103.9
-20.1
-20.04
-3.9
-4.5
-11.29
-13.32
-3.3
-5.1
3250
53.71
21.56
65.65
64.68
692.4
521
805.8
737.3
3.9
4.4
10.24
120
103.6
103.9
-21.43
-21.39
-3.9
-5.9
-12.24
-14.23
-3.31
-6.07
3239
53.49
21.51
65.77
67.02
689.6
523.8
806.5
736.4
3.7
4
10.12
124.4
103.6
103.9
-22.63
-22.63
-4.11
-7.33
-13.08
-15.28
-3.31
-7.18
3224
53.19
21.58
66.11
65.22
689.6
525.8
808.2
737.2
3.7
4
10.09
125.2
103.6
103.9
-22.69
-22.63
-4.28
-7.38
-13.39
-15.35
-3.79
-7.29
3219
53.19
21.52
65.32
66
688.8
525.7
807.3
736.8
3.7
4
10.1
126.1
103.6
103.9
-22.84
-22.53
-4.45
-7.28
-13.55
-15.7
-4.05
-7.53
3214
53.11
21.59
65.87
64.45
687.9
525.2
806.9
736.7
3.7
4.1
10.11
126.1
103.6
103.9
-22.8
-22.45
-4.33
-6.87
-13.67
-15.46
-4.2
-7.4
3212
53.37
21.56
64.15
64
687.4
524.2
805.1
736.5
Table 3. Results obtained on equipment from Room 2
Mean values
C
Par
1.1
1.
1.2
2.
2.1
3.
3.1
4.1
4.
4.2
5.1
5.2
5.3
5.4
6.
5.5
5.6
5.7
5.8
7.
6.1
7.1
8.
7.2
8.1
9.
8.2
8.3
8.4
10.
8.5
8.6
168
Jan
Feb
Mar
Apr
Mai
Jun
Jul
3.7
4.5
10.28
115.7
103.6
103.9
-21.60
-20.60
-2.48
-2.95
-12.05
-13.43
-2.8
-4.41
3243
53.77
21.63
67.46
63.69
683.2
520.6
793
733.6
3.7
4.5
10.55
113.3
103.6
103.8
-21.71
-20.29
-2.50
-2.80
-12.36
-13.47
-2.9
-4.74
3257
53.67
21.64
67.68
62.72
684.3
521.1
794.5
737.3
3.8
4.5
10.32
115.2
103.6
103.9
-21.28
-20.42
-2.67
-3.25
-11.81
-13.46
-3.1
-4.49
3249
53.80
21.56
67.88
63.99
683.5
520.1
795
733.2
3.9
4.5
10.50
112.2
103.6
103.9
-20.77
-20.29
-2.75
-3.36
-11.55
-13.67
-3.0
-4.38
3264
53.87
21.61
67.15
62.94
684.4
521.1
795.8
726.8
3.8
4.5
10.51
113.3
103.6
103.9
-21.26
-20.22
-2.99
-3.49
-11.59
-13.14
-2.8
-4.62
3258
53.84
21.61
68.15
64.45
684.5
520.4
795.5
738.1
3.8
4.6
10.57
115.1
103.6
103.9
-21.54
-19.89
-3.26
-3.19
-12.00
-12.96
-2.9
-4.27
3250
53.63
21.66
67.37
63.97
684.5
520.3
794.9
739.9
3.9
4.5
10.40
119.6
101.6
101.8
-21.86
-20.49
-3.45
-4.04
-12.50
-13.69
-3.4
-5.29
3234
53.49
21.65
67.44
63.45
685.1
521.1
796.1
732.9
Daily quality control in computed radiography mammography using the manufacturer phantom
Acknowledgment
Advanced Diagnostic Imaging Clinic
Technological Federal University of Paraná.
(DAPI)
and
5.
References
1. INCA [homepage on the Internet]. Available from: http://www.inca.gov.br.
2. Brasil. Ministério da Saúde. Agência Nacional de Vigilância Sanitária.
Radiodiagnóstico médico: Desempenho de equipamentos e segurança.
Brasília: Editora Anvisa; 2005.
3. Daros K. Comparison of the performance of FCR-Profect systems with
different mammographic equipment. World Congress on Medical Physics
and Biomedical Engineering, Munich, Germany; 2009.
4. Van Engen R, Young K, Bosmans H, Thijssen MAO. The European
6.
7.
8.
9.
Protocol for the quality control of the physical and technical aspects of
mammography screening. In: European guidelines for quality assurance in
mammography screening: Addendum on digital mammography. Nijmegen:
European Reference Organization for Quality Assured Breast Screening and
Diagnostic Services; 2003.
Carrol QB. Fuchs’s radiographic exposure processing and quality control.
Springfield: Charles C. Thomas Publisher; 2007.
American Association of Physicists in Medicine. Acceptance Testing and
Quality Control of Photostimulable Storage Phosphor Imaging Systems.
American Association of Physicists in Medicine, College Park; 2006.
Fujifilm. FCR Mammography QC Software (CR-IR 348CL) Operation Manual.
Fujifilm Corporation; 2007.
Fujifilm [homepage on the Internet]. Available from: http://www.
fujifilm.com.
European Communitie. European guidelines for quality assurance in
breast cancer screening and diagnosis. Luxembourg: Office for Official
Publications of the European Communitie; 2006.
169
Artigo Original
Detection of the smallest microcalcifications
for early diagnostic of breast cancer
Detecção das menores microcalcificações para o
diagnóstico precoce do câncer das mamas
Elizandra Martinazzi and S. O. Kepler
Institute of Physics of Federal University of Rio Grande do Sul (UFRGS) – Porto Alegre (RS), Brazil.
Abstract
Even though breast cancer is a cancer with relatively easy early diagnostic and has an appropriate treatment, it has high mortality rates in Brazil. This is
in part because the disease is diagnosed only in advanced stages, but also because the whole information contained in the mammograms is not used
by physicians and radiologists. There are many parameters to be considered in assessing the quality of a mammogram image. Among these parameters
are contrast, spatial resolution, the signal-to-noise ratio, and the efficiency of the applied dose. Even with the improvement of the quality of radiographs,
many structures, such as small microcalcifications, are not always identified by radiologists in the images. To determine the lowest detectable structures
in digital mammograms, we made a numerical analysis of a few digital mammography using simulators, determining the spatial and intensity resolutions,
and studying the noise and its distribution. With this, we could determine the detection levels, quantifying the probability that any point is due to statistical
noise or a real change in breast density. This is the first step towards early detection of microcalcifications. In our work, it was possible to detect even the
smallest microcalcifications of the simulator, 0.18 mm in diameter, with false alarm probability smaller than 1/1000.
Keywords: breast cancer, computer-aided diagnosis, mammography, radiology.
Resumo
Apesar do câncer de mama ser um câncer com diagnóstico precoce relativamente fácil e possuir tratamento adequado, o câncer de mama possui
taxas de mortalidade muito elevadas no Brasil. Isso deve-se provavelmente ao fato da doença ser diagnosticada somente em estágios avançados,
mas também pela não utilização de todas as informações fornecidas pelo exame mamográfico por parte dos radiologistas e médicos. Existem
muitos parâmetros a serem considerados para avaliar a qualidade de uma imagem mamográfica. Mesmo com a melhoria da qualidade dos
exames radiográficos, muitas estruturas, como as pequenas microcalcificações, nem sempre são identificadas pelos radiologistas nas imagens. Para
determinar as menores estruturas detectáveis nas mamografias digitais, fizemos uma análise numérica de algumas mamografias digitais, utilizando
simuladores mamográficos, determinando as resoluções espaciais e em intensidade, estudando o ruído e sua distribuição. Com isso, foi possível
determinar os níveis de detecção, quantificando a probabilidade de uma estrutura ser devido ao ruído estatístico ou uma alteração real na densidade
da mama, sendo esse o primeiro passo para a detecção precoce de microcalcificações. Estudamos as menores microcalcificações do simulador, de
0,18 mm de diâmetro, e determinando que possuem probabilidades de serem devido a ruído menores que 1/1000.
Palavras-chave: câncer de mama, diagnóstico com auxílio do computador, mamografia, radiologia.
Introduction
The significant increase in the incidence of breast cancer,
and consequently the mortality associated, has occurred
throughout the world in recent decades. The estimation of
new cases of breast cancer in Brazil was 49,240 for the year
2010, with a rate of 49 new cases for each 100 thousand
women, and, in the Southern Region, the risk is estimated
as 64 cases for each 100 thousand women, according to
data of the Brazilian National Institute for Cancer (INCA)1.
Breast cancer is the most common type of cancer
among women and the second most frequent in the world,
only behind lung cancer. It is much feared due to its high
frequency, the uncertainty of the success of treatment, the
possibility of recurrence, death, and, above all, by their
psychological effects2, which affect both the perception of
sexuality, as the very self image. It is relatively rare before
the age of 35 years, but above this age group its incidence
grows rapidly and progressively. In Brazil, it is responsible
for most deaths among women, with an estimate of 22%
new cases per year of cancer in women1. It also occurs in
men, but much less frequently.
Mammography
The objective of mammography is the early detection
of breast cancer screening in asymptomatic women3.
Corresponding author: Elizandra Martinazzi – Institute of Physics of UFRGS – Av. Bento Gonçalves, 9500 – CEP: 91501-900 – Porto Alegre (RS), Brasil –
E-mail: [email protected]
171
Martinazzi E, O. Filho KS
Mammography is the most usual technical detection method for non-palpable masses, intended for diagnostic
imaging of the breasts, using low doses of X-ray. However,
it is very important that it presents a good quality image
and diagnostic, having high contrast and resolution, a
high signal/noise, since the contrast of various tissues that
compose the breast is low.
The early discovery of this type of cancer through
mammography can provide the increase of the chances
of a successful treatment. For this, it is necessary to use
all the information contained in mammographic images.
However, retrospective analyzes of the mammographic
examinations revealed that a large number of breast cancer cases were already visible previously4, which demonstrates that not all the data contained in the mammography
was used, making necessary to increase the efficiency of
the diagnostic, never to replace the radiologist, but in order
to help radiologists render better clinical decisions.
Microcalcifications
One of early signs of malignancy in breast cancer may be
through non-palpable lesions, linked to certain forms and
patterns of distribution5. The detection of microcalcifications represents the earliest sign of possible breast cancer,
but the visual interpretation make the mammogram a fatiguing and time-consuming work because of the small size
of the microcalcifications6.
Mammography is still the most widely used technique,
widespread and promoted because its low cost for the detection of microcalcifications, which are composed basically by calcium, with great attenuation in relation to other
tissues. Their detection is relatively easy when on a uniform
region, but hampered by the low contrast and signal/noise
of complex images7.
For the assessment of mammography, one must observe: the sizes, number, shape, density and distribution
of microcalcifications, which are important factors to be
taken into account for the analysis criteria5 (Figure 1).
than the size of the detector, so that all pixels are equally
illuminated.
Noise
Noise is any external signal or statistical fluctuation superimposed on the signal of source. Even when exposing a
detector to a uniform X-ray beam, the actual number of
photons counted by the detector varies for each pixel of
the image. This occurs because of the statistical nature of
the emitting source, of the attenuation processes, so that
the number of photons absorbed varies from pixel to pixel,
and these are the source of the information9. The noise
can conceal the information in the image, for example, low
contrast structures. It also affects the spatial resolution, reducing the system capability of separate small and nearby
structures, as microcalcifications.
The signal-to-noise ratio is a fundamental limitation
on the perception of effects; it must be high to maximize
the information, but the attempt to increase the signal-tonoise ratio increases the dose damaging to the patient. It
Figure 1. Microcalcifications example.
Mammographic simulator
Mammogram simulators are used to evaluate the image
quality8, with different objects, of different sizes, representing small structures of the breast, as fibers, microcalcifications and tumor masses. Placed in the X-ray apparatus
where the breast should be placed and simulating the
size of an average breast, it generates an image basis to
be examined.
The simulator used in our analysis was the Phantom
Mama, formed by a body of acrylic with a plate of wax
containing the objects, which were used in the simulations
of microcalcifications with dimensions of 0.45, 0.35, 0.30,
0.25 and 0.18 mm (Figure 2).
Another object for tests is a PMMA acrylic plate, entirely homogeneous, of 2x25x30 cm. The plate must be larger
172
Figure 2. Objects of the mammographic simulator Phantom
Mama.
Detection of the smallest microcalcifications for early diagnostic of breast cancer
is necessary to balance an acceptable image for the diagnostic and the lowest exposure possible to the patient.
Noise sources include random noise, due to fluctuations in counts, and the noise of reading, caused by thermal noise, due to the movement of atoms, dark, since
even in the absence of the signal there may be formation
of pairs of electron-gap in the detector, generally temperature dependent9.
Dense regions are generally noisy and present low
contrast, while microcalcifications have properties of high
attenuation, but their small size tends to present low local
contrast.
The number of photons counted in images presents
statistical fluctuations, causing low contrast structures to
present difficulties in differentiating from noise. To assess
the probability that a point is a real detection, as a lesion
of the breast, or is due to noise, we study the noise and
its distribution in order to determine detection levels both
spatial and in intensity. The noise distribution is Poissonic,
as shown in Figure 3, which shows the values of background Regions of Interest (ROIs), i.e., without any object,
and in the direction not affected by the anodic effect.
The detector of the digital mammogram apparatus
used is 19x23 cm, forming images of 1917x2294 pixels,
with pixel size of approximately 0.1 mm on the side10.
The equipment used in this work was the DR model Senographe 2000D GE Medical Digital Systems
Mamography11, used in a radiographic clinic in Porto Alegre.
Simulator analysis
To study the effect of changes in energy and current in
the images, we exposed the mammographic simulator to
various current values and energy of the X-ray beam. To
analyze the spatial resolution and the sensitivity of images,
we examine how many standard deviations each one of
the simulator objects presented. For this, first, we must
study the background, that is, the regions which contain
Figure 3. Distribution of background counts obtained in a mammogram.
no objects, and determine what their fluctuations are, calculating the standard deviation of the background.
Once estimated the background, we measure the value of greatest absolute frequency, defined as mode of the
distribution, using a ring around the object of study.
Using ImageJ, a free software in the public domain
written in Java12, developed from NIH Image and available
to every office computer, it is possible to study the areas of
interest. These ROIs must be chosen parallel to the length
of the detector to avoid the reduction of intensity caused
by the anodic effect, which is an irregular distribution of
the X-ray intensity caused by the inclination of the anode.
With the histogram of the image, the software also shows
the values: maximum, minimum, mode, counts, mean and
standard deviation, determined for each region of interest.
Subsequently, we calculated the standard deviation of the
background, σ (background).
For each structure (object) of the simulator, we
determined:
|mode(structure) - mode (background)|
σ(background)
(1)
Results
Data were collected using the simulator Phantom Mama
images, irradiated in the Senegraphe 2000D mammographic equipment, with different values of voltage and current. The values of 28 kV with 138 mA correspond to the
automatic value chosen by the equipment for the simulator, and 26 kV and 60 mA parameters are recommended
by the manufacturer in the manual for the quality control
image. All images correspond to the effects simulating an
examination of the breasts in position craniocaudal (CC),
with molybdenum target and filter (Mo/Mo).
In the analysis of the values of background mode, in regions where there are no objects, the distribution of the values of pixels may be represented by a Gaussian function.
The histogram may be described by Gaussians, following
the Central Limit Theorem, which states that the distribution of many averages is a Gaussin distribution.
After we obtained the values of the mean and standard
deviation, we can plot the signal-to-noise ratio for each
image versus the power of the X-ray beam, given by the
product of voltage and energy in each image.
The points for the automatic exposure present the best
signal-to-noise ratio, proving that it was well selected by
the manufacturer and the equipment is operating within
the specifications.
Microcalcification simulators
Table 1 shows the values of the signal-to-noise ratio measured for smallest simulators of microcalcifications, with
0.18 mm (Figure 4).
The detection of a signal smaller than 3 σ occurs for
energy values of 28 kV and current of 80 mA. Table 1 lists
173
Martinazzi E, O. Filho KS
the voltages, current, average and standard deviation (σ)
of the background, with their respective values of signalto-noise ratio of the smallest sets of microcalcifications
(0.18 mm) in the Phantom Mama, in relation to the probability that the points are real or only noise of the image.
The second group of phantoms of microcalcifications,
with 0,25 mm, has the signal-to-noise ratio of 3.9 even for
the image with energy of 26 kV and current of 50 mA, already presenting likelihood of been real greater than 99.9%.
Table 1. Signal-to-noise ratio with their respective likelihood for
assemblies of the smallest microcalcifications (0,18 mm) to be
real or only noise of images, with relation to the values of tension
and current, average background and standard deviation σ.
kV
26
28
30
26
28
30
26
28
30
28
mA
50
50
50
80
80
80
125
125
125
138
Average
361
559.6
818.8
585.5
903.5
1325.4
920.4
1423.1
2087.1
1572.3
Sigma
7.3
9.6
11.9
9.8
12.7
16.1
12.6
17.5
22.2
15.5
S/R
1.9
2.3
3.0
2.2
3.2
3.4
2.5
3.4
4
2.3
Real (%)
94.3
97.8
99.7
97.3
99.9
99.9
99.9
99.9
99.99
99.9
Noise (%)
5.7
2.3
0.3
2.7
0.1
0.1
0.1
0.1
0.01
0.1
All other images, as well as the other sets of larger microcalcifications, have even greater probability of being real.
Conclusion
The smallest dimension microcalcification tested is 0.18
mm and is above the limit of detection, with signal-tonoise ratio of the order of 3.2 and likely to be real, and
not only noise of the image of 99.9% for exposures of
28 kV and 80 mA. Statistically, this means that part of
the information of real image is not being considered in
the evaluation or diagnosis, as they are detecting in general only masses of 0.5 mm or larger4, i.e., information
is not been used. The smallest microcalcifications that
the authors can see by eye have 0.25 mm, already with
signal-to-noise ratio of 3.9. If the probability distribution is
Poissonic, for signal=3σ, the probability of being due to
noise is already smaller than 0.1%.
The analysis of images is a process of finding, identifying and understanding the patterns that are relevant
for the function for which the image is designed. An important goal of the analysis of images using computation
is to provide the machine capacity to simulate the human senses, and, in addition, use their tools and speed
to make analysis less subjective, independent of human
abilities, having the capacity to recognize patterns at
previously established levels, as well as to considerate
probabilities.
Our next steps will be to analyze real breast mammograms, which have signal distribution much more complex
because of the density variations of breast tissues, and to
develop a software showing the regions that are above a
certain level of detection, which may be altered progressively. The objective is to use all the information contained in
mammograms and not only those of high signal/noise that
are easily identified by the naked eye.
Acknowledgment
Research supported by CNPq, CAPES and FAPERGS,
Brazil.
References
1.
2.
3.
Figure 4. Signal-to-noise ratio for microcalcification simulators
of 0.18 mm for different powers of the X-ray beam.
174
4.
Brasil. Ministério da Saúde Brasil. Instituto Nacional do Câncer (INCA).
Estimativas 2010: Incidência de câncer no Brasil.. Rio de Janeiro; 2010.
Soares RG. Aspectos emocionais do câncer de mama. Revista Virtual de
Psicologia Hospitalar e da Saúde. 2008;3(6):24-9.
Almeida CE, Magalhães MG, Casicava J, Peixoto JE, Canella E. Manual
de Técnicas Mamográficas. Projeto de capacitação profissional para
detecção precoce do câncer de mama por intermédio da mamografia.
Rio de Janeiro; 2007.
Nishikawa RM. Current status and future directions of computer-aided
diagnosis in mammography. Comput Med Imaging and Graph. 2007;31(45):224–35.
Detection of the smallest microcalcifications for early diagnostic of breast cancer
5.
6.
7.
8.
Brasil. Ministério da Saúde. Instituto Nacional do Câncer (INCA).
Mamografia: da prática ao controle. Rio de Janeiro: Universidade do
Estado do Rio de Janeiro (UERJ); 2007.
Marrocco C, Molinara M, Tortorella F. A computer-aided detection system
for clustered microcalcifications. Artif Intell Med. 2010;50(1):23-32.
Lemacks MR, Kappadath SC, Shaw CC, Liu X, Whitman GJ. A dualenergy subtraction technique for microcalcification imaging in
digital mammography—A signal-to-noise analysis. Med Phys.
2002;29(8):1739-51.
Brasil. Ministério da Saúde. Secretaria de Vigilância Sanitária. Portaria
nº 453, de 1º de junho de 1998: Diretrizes de proteção radiológica em
radiodiagnóstico médico e odontológico. Brasília; 1998.
9. Kepler SO, Saraiva MF. Astronomia e Astrofísica. São Paulo: Editora
Livraria da Física; 2004.
10. Suryanarayanan S, Karellas A, Vedantham S, Sechopoulos I, D’Orsi CJ.
Detection of simulated microcalcifications in a phantom with digital
mammography: effect of pixel size. Radiology. 2007;244(1):130-7.
11. GE Healthcare [internet]. Available from: http://www.gehealthcare.com.
12. ImageJ, National Institutes of Health (NIH) [internet]. Available from: http://
rsbweb.nih.gov/ij/.
175
Artigo Original
Development and evaluation of Standard
Operating Procedures (SOPs) for quality
control tests and radiological protection
activities in a Nuclear Medicine Service
Desenvolvimento e avaliação de Procedimentos Operacionais
Padrão para testes de controle de qualidade e atividades de
proteção radiológica em um Serviço de Medicina Nuclear
Alexandre R. Krempser1, Alexandre B. Soares2 and Rossana Corbo3
Programa de Engenharia Biomédica da Universidade Federal do Rio de Janeiro (PEB/COPPE/UFRJ) – Rio de Janeiro (RJ), Brazil.
2
Instituto de Física da Universidade Federal do Rio de Janeiro (IF/UFRJ) – Rio de Janeiro (RJ), Brazil.
3
Departamento de Radiologia da Universidade Federal do Rio de Janeiro (FM/UFRJ) – Rio de Janeiro, Brazil.
1
Abstract
The quality management in Nuclear Medicine Services is a requirement of national and international standards. The Brazilian regulatory agency in
health surveillance, the Agência Nacional de Vigilância Sanitária (ANVISA), in its Resolução de Diretoria Colegiada (Collegiate Directory Resolution)
nº 38, requires the elaboration of documents describing the technical and clinical routine activities. This study aimed to elaborate, implement
and evaluate Standard Operating Procedures (SOPs) for quality control tests and radiological protection activities in the Nuclear Medicine Service
of a university hospital. Eighteen SOPs were developed, involving tasks related to dose calibrator, gamma camera, Geiger-Müller detectors and
radiological protection activities. The performance of its application was evaluated for a period of six months. It was observed a reduction in 75% of
reported operational errors and 42% of the number of reported incidents with contamination by radioactive material. The SOPs were adequate and
successful in its application. New procedures involving clinical activities will also be developed and evaluated.
Keywords: quality management, nuclear medicine, quality control.
Resumo
O controle de qualidade em Serviços de Medicina Nuclear é uma exigência de normas nacionais e internacionais. A Agência Nacional de Vigilância
Sanitária (ANVISA), na Resolução de Diretoria Colegiada nº 38, solicita a elaboração de documentos que descrevam as atividades da rotina técnica
e clínica. Este estudo teve o objetivo de elaborar, implementar e avaliar os Procedimentos Operacionais Padrão (POPs) para testes de controle de
qualidade e atividades de proteção radiológica no serviço de medicina nuclear de um hospital universitário. Dezoito POPs foram desenvolvidos com
tarefas relacionadas ao calibrador da dose, à gama câmara, aos detectores Geiger-Müller e às atividades para proteção radiológica. O desempenho
da aplicação destes foi avaliado durante seis meses. Observou-se redução de 75% dos erros operacionais relatados e de 42% do número de
incidentes relatados com contaminação por material radioativo. Os POPs foram adequados e bem sucedidos em sua aplicação. Novos procedimentos
envolvendo atividades clínicas também serão desenvolvidos e avaliados.
Palavras-chave: gestão de qualidade, medicina nuclear, controle de qualidade.
Introduction
The quality management of procedures performed and radiological protection of patients and professionals should
be priority in any Nuclear Medicine Service (NMS)1,2.
The Brazilian regulatory agency in health surveillance,
the Agência Nacional de Vigilância Sanitária (ANVISA),
published with this purpose the Resolução de Diretoria
Colegiada – RDC nº 382, which requires the preparation of
documents describing the technical and clinical activities
undertaken in the NMS. These documents must consist of
Standard Operating Procedures (SOPs).
A SOP is a detailed description of all activities required to
perform a particular procedure, i.e., a standardized checklist
Corresponding author: Alexandre Rodrigues Krempser – Programa de Engenharia Biomédica (PEB/COPPE/UFRJ) – Av. Horácio Macedo, 2030 –
CEP: 21941-914 – Rio de Janeiro (RJ), Brasil – E-mail: [email protected]
177
Krempser AR, Soares AB, Corbo R
to perform an activity3. It has great importance in any functional process, whose basic objective is to ensure, through
standardization, the expected results for each task performed.
This is a quality management tool that strives for excellence in
service delivery, minimizing errors in routine actions4.
Basically, the process for quality assurance through
the SOP involves planning, development, verification
and implementation. Good clinical practice involves the
quality management of established and well-controlled
processes5.
The aim of this study was elaborate, implement and
evaluate Standard Operating Procedures for quality control
tests and radiological protection activities in the Nuclear
Medicine Service of a teaching hospital, in accordance
with national and international standards.
SOPs elaboration
First, it was conducted in the NMS a survey of the following items: (i) technical activities undertaken; (ii) equipment
installed; (iii) technical and operational equipment manuals;
(iv) documentation relating to the quality control tests performed; and (v) radiological protection activities. The service has two gamma cameras Millennium (GE Healthcare),
a dose calibrator CRC-10BC (Capintec Inc.) and two detectors Geiger-Müller MIR-7026 (Norton Helthcare) with
surface probe.
Quality control tests were routinely performed. However,
there were no documents standardizing its implementation,
and results were recorded only in printed documents.
The second step was to divide the activities in groups
relating to procedures for dose calibrator, gamma camera,
detectors and radiological protection activities.
The third step was to define the correct procedure for
each quality control test, consulting recommendations in
national and international documents6,7 and even in equipment manuals.
The fourth step was the SOPs elaboration. Each SOP
is composed mainly of cover sheet, instructions sheet and
attachments sheet.
The items contained in each part are following
presented.
• Cover sheet:
• Name and logo of hospital.
• Title (activity/process).
• Name of the author.
• Date of elaboration and approval.
• Name of the reviewer.
• Number of the current version.
• Number of the document.
• Paging.
• Coverage.
• Distribution.
• Number of copies.
• Name of the radioprotection supervisor.
178
Instructions sheet:
• Objective.
• Acronyms.
• Equipments used.
• Step by step of the procedure.
• Calculation (when applied).
• Reference values.
• Instructions for completing the forms.
• Bibliographic references.
• List of attachments.
Attachments sheet:
• Figures.
• Tables.
• Examples.
Implementation and evaluation of SOPs
After the SOPs elaboration, they were reviewed, approved
and delivered to an intern in Medical Physics and to other in
Radiopharmacy, according to the procedure purpose, in order
to test the understanding and clarity of the instructions. After
this initial application, information was corrected or added.
A seminar was held to present the SOPs to the
Administration of NMS and the professionals involved in its
implementation and training on its use.
Printed copies were distributed among the NMS sectors and put in places easily accessible to those responsible for implementing the procedures.
Spreadsheets for data analysis were developed using Excel
software (Microsoft Corporation) to record the tests results. The
choice of this software was due to its versatility and easy development of mathematical calculations and data binding.
The SOPs were implemented and evaluated for six
months, as the implementation and performance results.
The indicators used for evaluation were: (i) number of occurrences of contamination by radioactive material and (ii) number of occurrences of operational errors in the equipments.
Results and discussions
Eighteen SOPS were elaborated and implemented, being
04 for the dose calibrator, 07 for the gamma camera, 01
for the detectors and 06 for radiological protection activities (Table 1).
Being a teaching hospital, the NMS has a large turnover of staff employed, students, residents and interns.
Thus, the implementation of SOPs facilitated the process
of training these professionals to perform their activities.
Table 2 shows the performance assessment results of
SOPs. There was a 75% reduction of the operational errors
recorded. The number of recorded incidents with contamination by radioactive material was reduced by 42%.
The registers were made in accordance with occurrences reported to the radiological protection supervisor of
NMS. However, even with seminars presenting the objectives of SOPs and the training of professionals involved in the
Development and evaluation of Standard Operating Procedures (SOPs) for quality control tests and radiological protection activities in a Nuclear Medicine Service
Table 1. List of elaborated SOPs.
Group
Dose
calibrator
Gamma
camera
Radiation
detectors
Nº
1
2
3
4
5
6
7
8
9
10
11
Procedure
Accuracy and precision test
Linearity test
Reproducibility test
Dial adjustment
Intrinsic uniformity test
Intrinsic linearity test
Extrinsic uniformity test
Linearity extrinsic test
Energy resolution test
Sensitivity test
Center of rotation test
12
Reproducibility test
13
Radiological
protection
activities
14
15
16
17
18
Evaluation of surface, skin, instruments and
clothes contamination
Skin, instruments and surfaces (countertops,
tables and floors) decontamination
Radiometry service
Elution of 99mTc generator
Calculation of the fraction of molybdenum in
the solution of eluted 99mTc
Radioactive waste management
!
Figure
1. Format designed for cover sheet.
practices of NMS, there was some fear of punishment by the
operational errors. Thus, the number of occurrences may differ from the actual number of incidents recorded during the
evaluation period. This discrepancy tends to disappear with
time, when the professionals understand that the purpose of
SOPs is to optimize the tasks executed in the NMS.
The employees with more than five years of service had
reservations about the use of SOPs. They did not meet the
steps properly. The operational errors were recorded in the
tasks performed by them.
The quality control tests results were inserted in the
spreadsheets for analysis and register. This form of registration
allows generation of practical graphs and charts to management consultation, and analysis of equipment performance.
The Figures 1 and 2 show the final format designed for
the cover sheet and instruction sheet, respectively.
Table 2. Average monthly number of incidents recorded before
and after the implementation of SOPs.
Indicator
Before After Reduction (%)
Number of occurrences of
12
07
42
contamination by radioactive material.
Number of occurrences of
08
02
75
operational errors in the equipments.
!Figure 2. Format designed for instructions sheet.
179
Krempser AR, Soares AB, Corbo R
Conclusions
The developed SOPs were adequate and successful in its
application.
New procedures will be developed and implemented
involving clinical activities, such as:
1. Pregnancy protocols (patients/employees).
2. Staff monitoring.
3. Specific protocols for image and non-image.
4. Patient assessment.
5. Education, preparation and release of patient.
6. Acquisition and image processing.
7. Data storage/transfer.
8. Adverse drug events and misadministrations.
The new SOPs will be evaluated for a period of one
year, considering the indicators as: (i) rate of repeat
tests stratified by subject (dose errors, technique, image acquisition protocols) and (ii) rate of intercurrence
(stratified in clinical intercurrence, techniques and of
180
radiation protection, with the record of corrective measures taken).
References
1. Comissão Nacional de Energia Nuclear. Diretrizes básicas de proteção
radiológica. CNEN-NN-3.01. Brasil; 2005.
2. Agência Nacional de Vigilância Sanitária. Regulamento técnico para
instalação e funcionamento de Serviços de Medicina Nuclear “in vivo”.
Resolução de Diretoria Colegiada (RDC) nº 38. Brasília: ANVISA; 2008.
3. Colenghi VM. O & M e qualidade total: uma integração perfeita. Uberaba:
Editora Qualitymark; 1997.
4. Hattemer-Apostel R. Standard operating procedures: a novel perspective.
Qual Assur J. 2001;5:207-19.
5. Woodin KE. Standard operations procedures (SOPs). In: The CRC´s guide to
coordinating clinical research. Boston: Thompson Center Watch; 2004. p. 59-72.
6. International Agency of Atomic Energy. Quality control in nuclear instruments.
TECDOC-6.02. Viena: IAEA; 1991.
7. Comissão Nacional de Energia Nuclear. Requisitos de radioproteção e
segurança para serviços de medicina nuclear. CNEN-NN-3.05. Brasil; 1996.
Artigo Original
Level of occupational exposure during daily
work in a Nuclear Medicine Department
Nível de exposição ocupacional durante trabalho diário
em um Departamento de Medicina Nuclear
Marcelo Schwarcke1,2, Domingos Cardoso3 and Nadya Ferreira2
1
Departamento de Física e Matemática da Universidade de São Paulo – Ribeirão Preto (SP), Brazil.
Departamento de Engenharia Nuclear do Instituto Militar de Engenharia (IME) – Rio de Janeiro (RJ), Brazil.
3
Comissão Nacional de Energia Nuclear/Instituto de Radioproteção e Dosimetria – Rio de Janeiro (RJ), Brazil.
2
Abstract
Workers of the Nuclear Medicine Department have a very complex geometric exposition. The source of irradiation is not collimated
and irradiated for all direction, the interaction with many structural tissue is inside the body before could be detected outside. The
professional who works in a Nuclear Medicine Department is exposed to this condition and different energies. This work proposes a good
approach to estimate the mensal dose level according to the dose rate during their daily routine. To measure the dose rate, a Babyline
81 ionization chamber was used, and the most frequent exams using 99mTc were chosen. A previous study was conducted to determine
the most frequent exams made in the Nuclear Medicine Department at the Central Army Hospital in Rio de Janeiro, and previous
environment monitoring determine the places with higher exposure that could interfere in the measurement of this paper. The Renal
scintigraphy with diethylenetriaminepentaacetic acid (DTPA) had an average dose rate of (2.50±0.25) µSv/h; for the Renal scintigraphy
with dimercaptosuccinic acid (DMSA), it was of (1.20±0.25) µSv/h; for Bone scintigraphy using two different protocols, it was (2.63±0.30)
µSv/h and (3.09±0.30) µSv/h. Exposition during elution, dose preparing and clinical procedure was considered a critical moment in the
daily routine of the employee. The dose rate obtained in this study demonstrated that the professional cannot exceed the public dose limit
in one day of his work routine. Therefore, for the Radioprotection Department, this is a good approach to make a radioprotection plan in
the Nuclear Medicine Department.
Keywords: nuclear medicine, dosimetry, professional, dose rate.
Resumo
Os profissionais do Departamento de Medicina Nuclear têm uma exposição geométrica muito complexa. A fonte da irradiação não é
colimada e irradiada para todas as direções, a interação com os diversos tecidos estruturais é feita dentro do corpo antes de poder ser
detectada na parte de fora. O profissional que trabalha no Departamento de Medicina Nuclear está exposto a essa condição e a diferentes
energias. Este trabalho propõe uma boa abordagem para estimar o nível de dosagem mensal de acordo com a taxa de dose durante sua
rotina diária. Para medir a taxa de dose, utilizou-se uma câmara de ionização Babyline 81, e os exames mais frequentes que utilizam
99m
Tc foram escolhidos. Um estudo anterior foi realizado para determinar os exames de maior frequência feitos no Departamento de
Medicina Nuclear do Hospital das Forças Armadas Central, no Rio de Janeiro, e o prévio monitoramento do ambiente determina os locais
com maior exposição que poderiam interferir na medição deste trabalho. A cintilografia renal com ácido dietileno triamino penta-acético
(DTPA) tinha uma taxa de dose média de 2,50±0,25 µSv/h; a cintilografia renal com ácido dimercaptosuccínico (DMSA), de 1,20±0,25
µSv/h; e a cintilografia óssea, utilizando dois protocolos diferentes, de (2,63±0,30) µSv/h e (3,09±0,30) µSv/h. A exposição durante a
eluição, a preparação da dose e o procedimento clínico foi considerada um momento crítico na rotina diária do funcionário. A taxa de
dose obtida neste estudo demonstrou que o profissional não pode exceder o limite de dose público em um dia de sua rotina profissional.
Portanto, para o Departamento de Radioproteção, esta é uma boa abordagem para criar um plano de radioproteção no Departamento de
Medicina Nuclear.
Palavras-chave: medicina nuclear, dosimetria, profissional, taxa de dose.
Corresponding author: Marcelo Schwarcke – Departamento de Física e Matemática da Universidade de São Paulo – Av. Bandeirantes, 3900 – Monte Alegre –
CEP: 14040-900 – Ribeirão Preto (SP), Brazil – E-mail: [email protected]
181
Schwarcke M, Cardoso D, Ferreira N
Introduction
The Nuclear Medicine Service uses to carry out its examination
a radionuclide that is chemically added to a compound able to
bind to the cells, forming the organ to be examined, showing
images of its physiology and in some cases its anatomy1,2. The
Tecnecium-99m (99mTc) is the most commonly used radionuclide in the Nuclear Medicine Service because it has a half-life
of 6.02 hours and an emission of a single photon of 140.0
keV, allowing a lower dose to the patient during the exam
and a better image quality3,4.
Several studies have been made about the quality of
radiopharmaceuticals administered to patient considering dose levels administered, precision of activity measurements and contamination by Molybdenum5. These
dosimetric indices are related to the patients and people
accompanying the examination of Nuclear Medicine and
their radiological protection levels6-9, but are not enough
to infer the doses received by the professional involved in
Nuclear Medicine Service. Furthermore, literature is sparse
on reports dose rate received by professional that handle
radioactive sample in the Nuclear Medicine Service. For
a more accurate dosimetric evaluation, it is necessary
to know the workers exposure in each stage of the daily
proceedings9-13.
Literature reports two strategies to determinate the
dose received by nuclear medicine professionals. One of
them is based on dose rate measurements at a fixed reference distance of patient and determining the time that the
professional remains at this distance. Other methodology is
the direct reading of personal electronic dosimeters during
the nuclear medicine procedures. In this study, a review of
the professional procedures in Nuclear Medicine Service is
presented to conduct an assessment of dose rate during
the various exams stage, aiming to decrease the dose received by workers in Nuclear Medicine Service.
The determination of the absorbed dose for the professional in each procedure in a Nuclear Medicine Service
is complex and involves individual conductions evaluation
during all nuclear medicine exam procedure, because we
must examine the individual conduct of the professional
to determine which employee routine is more hazardous.
The purpose of this study was to conduct an assessment
of dose rate during the various stages of examination,
as well as a review of the procedures of the legal profession service14,15, aiming to decrease the dose received by
workers.
Material and methods
The survey was conducted at a Nuclear Medicine
Department of Army Hospital in Rio de Janeiro. Four professionals were monitored during four months. For this
monitoring, it was used the personal dosimetry and an ionization chamber model Babyline 81, of the manufacturer
Eurisys Measures16.
182
For data acquisition during the examination, the detector was positioned at 1.0 m away from the gantry of camera range and 0.8 m from the ground, normally in gonads
region. The choice of this height is due to the fact that the
gonads region are at a parallel height to the patient during
the examination, the bladder of the patient retains most of
the radioisotope17 that was not added to the previous made
to determine the relationship of the measurement at a height
of 0.8 m and the height of 1.50 m, average height of the personal dosimeter in Brazil18, to compare with other studies.
The protocol for bone scintigraphy at Department was
the administration of 1110.0 MBq of 99mTc-MDP (99mTcMetilnodifosforico Acid)19,20, intravenously, waiting 3 hours
for the complete fixation of radiopharmaceutical. The
Department used two different protocols: a) Protocol 1 the image can be obtained using a fixed counts value and
a variable time acquisition; and b) Protocol 2 - the time of
acquisition was fixed and the counts value was variable. In
order to acquisition dose rate during the exam, we made 3
readings in each body section at each one minute during
all examination for each protocol.
We measure the dose rate during the exam of flow bone,
a previous exam for bone scintigraphy. For this, each acquisition data was made in a time interval of 15.0 s, and the
ionization chamber was positioned close to the technician.
The dynamic evaluation of the kidney is made using the
99m
Tc-DTPA (99mTc-Dietilenotriaminopentactico Acid)21,22.
For measurement, we decided to realize the reading of
the dose rate at intervals of 2.0 min during the exam realization. The protocol used by the Service is an average
administered activity of 407.0 MBq23 and 40.0 min for acquisition time.
Kidney anatomic evaluation is made using 99mTc-DMSA (99mTc-Dimercapto Succinic Acid)24, using an average
activity of 296.0 MBq and a waiting time of 5.0 h23. The
exam made is one static image for each gantry angle (0º,
180º, 135º and 225º), and we made three measurements
for each gantry angle.
Measurement inside manipulation room was made to determine the levels of exposure. For this, we choose acquisition
data during all process of radiopharmaceutical preparing. The
measurement was made in five phases: background levels
before manipulation and after all process, during generator
99
Mo/99mTc elution, during radiopharmaceutical preparing and
activity separation to be administered to the patient.
To make this measurement, the detector was positioned at 0.80 m from the ground and 1.60 m from the table
of manipulation. This distance was necessary because the
readings should be taken during the proceedings without
interfering in the routine and to minimize the scatter. Data
was obtained in five readings in each 15.0 s in each stage
of the process, and all readings were corrected25.
All rooms of the Nuclear Medicine Service were monitored by radiometric survey, using an ionization chamber
and an isotope identifier26. These values were used to determine the regions to be monitored and the factors that
could influence the measures during this work.
Level of occupational exposure during daily work in a Nuclear Medicine Department
Results
Bone scintigraphy
As mentioned, the analysis to the bone scintigraphy was
done in two phases. The first phase is the measurement of
bone flow, and the second phase is the study of two image
protocols used by the medical team.
In the flow bone phase, the professional who have
more exposition to the source is the nurse, responsible
for administer the radiopharmaceutical to the patient.
Five procedures were observed, and the average value of
dose rate was (5.17±0.52) µSv/h and the time of stay was
15.0 min. In this case, all measurements were made with
a fixed position of the professional during the job, being
possible to calculate the absorbed dose.
Professional uses the protocol 1 to make our image
normally using an average activity of 1417.1 MBq, resulting in a dose rate of (2.63±0.26) µSv/h and an average
time of acquisition of 66.3 min. To be used protocol 2, the
average activity administered was 1369.0 MBq, resulting
in an average dose rate of (3.09±0.3) µSv/h and a total
time of image acquisition of 42.6 min. In both protocol,
the professional remains 16.0 min in a distance less than
1.0 m from the patient.
Renal scintigraphy
In exams of renal scintigraphy with DTPA, the average activity administered was 444.0 MBq and the average dose
rate was (2.55±0.25) µSv/h, time of image acquisition was
37.0 min and 16.0 min of this time the professional remains
less than 1.0 m from the patient.
In renal scintigraphy with DMSA, the low activity administered to the patient, 45.0 MBq, result in a low professional exposure, than the average dose rate was
(1.20±0.12) µSv/h. The duration of exam was 18.4 minutes and 8.0 min of this time the professional stay less than
1.0 m from the patient.
A
Discussion
For bone scintigraphy, dose rate was 2.63-5.17 µSv/h.
Gomez-Palacios et al.12 find for the same condition dose
rate of 3.5-8.8 µSv/h, demonstrating great agreement between studies using the same irradiation geometry. This
conclusion is possible considering that Chiesa et al.10 find
for the same exam the accumulate dose of 0.3 µSv, and,
if we calculate the accumulate dose using ours methodology, we find values between 0.87-1.03 µSv. This difference is explained, because in this article the authors used a
fixed detector in the professional and the personal still in
movement, so all radiation were not detected.
For the comparison between protocol 1 and 2, the radioprotection service could not say which protocol have
better image, but the value of dose rate in protocol 2 is highest than in protocol 1. However, in this work is not possible to study new Protocols, in which the machine chose a
better protocol to be used. Probably, new machines reduce the exposition time of the professional, because reduce
the time with patient contact.
Results for renal scintigraphy with DTPA show similarity
of results obtained by Chiesa et al.10. The dose rate observed was (2.55±0.25) µSv/h, but this value could be modified
B
Background Morning
Background after preparing
Background end working
7
6
5
4
3
2
1
0
Background Morning
Background after preparing
Background end working
7
6
Dose Rate (µSv/h)
Dose Rate (µSv/h)
Levels of background radiation
During the procedures inside the room, the technician
handles the source to measure its activity and prepare the radiopharmaceutical. The range of dose rate in
which the activity is submitted during measurement is
15.5-5.5 µSv/h, and for radiopharmaceutical preparing
is 35.0-15.0 µSv/h.
Figure 1 shows a comparison between the levels of background radiation in two weeks studied, demonstrating
the difference between manipulators and theirs performance during the same active manipulated.
5
4
3
2
1
Monday
Tuesday
Wednesday
Tursday
0
Monday
Tuesday
Wednesday
Tursday
Figure 1. Difference between manipulators and the results at background levels. The graphic (A) demonstrates the performance of
technician A and graphic (B) demonstrates the performance of technician B.
183
Schwarcke M, Cardoso D, Ferreira N
during the time of exam, because in this kind of exam the patient need leave the room before to finish all image and return
to finish the sequence of image, and this difference results in
a decrease of the dose rate of ±0.3 µSv/h. The dose rate for
renal scintigraphy with DMSA was (1.20±0.12) µSv/h. This is
the exam with the lowest level of dose rate find in this work
and other papers, like Chiesa et al.10 and Mountford et al.9.
The background levels measured in the manipulation
room may contribute to the adoption of safe practice. In
the study of Smart13, it was demonstrated using a pocket
GM-tube dosimeter that preparing the radiopharmaceutical
to administration is the third higher exposure time during
the daily routine of the technician. This is demonstrated in
the Figure 1 and it is possible observe in each procedure of
the radiopharmaceutical preparation which one have more
probability to higher radiation exposure.
We observe from the numeric difference that two different
manipulators work in the same area with the same activity
and result in different background levels in the room. The lower level is the result of using all radioprotection shields; the
source is exposed only for a little time. The sources no shielded kept under the metallic structure for ventilation increase
the background levels, which results in radiation interaction,
causing a greater spread inside the room. An easy way to
solve this problem is to keep all possible sources shielded.
Conclusion
This study contributes to the professional responsible for
the radioprotection service to be able to implement a plan
of radioprotection based on numerical values. The results
demonstrated that the probability of the nuclear medicine
professional receiving in a single day of work more than the
limit dose for the public is negligible.
The highest potential dose for the professional in a daily
work is separate by function: for the nurse, the highest exposure is during radiopharmacy administration to the patient;
for the technician, the highest exposure occurs during the
radiopharmacy preparing; and, for the radiologist, the highest
exposure is during the patient interview after the exam.
Acknowledgment
Authors thank for the collaboration of Nuclear Medicine
Department of Army Central Hospital, Instituto Militar de
Engenharia and Universidade de São Paulo, CNEN for
technical support, and CNPq and CAPES for partial financial support.
References
1. National Research Council and Institute of Medicine of the National
Academies. Advancing Nuclear Medicine Through Innovation. Washington:
National Academies Press; 2007.
184
2. Sharp PF, Gemmell HG, Murray AD. Practical Nuclear Medicine. London:
Springer; 2005.
3. Webb S. The Physics of Medical Imaging. Bristol and Philadelphia: Institute
of Physics Publishing; 1998.
4. Sorenson JA, Phekps ME. Physics in Nuclear Medicine. Philadelphia:
Saunders Company; 1987.
5. Dantas BM, Dantas ALA, Marques FLN, Bertelli L, Stabin MG
Determination of 99Mo Contamination in nuclear medicine patient
submitted to a diagnostic procedure with 99mTc. Braz arch biol technol.
2005;48(2):215-20.
6. Stabin MG, Tagesson M, Thomas SR, Ljungberg M, Strand SE. Radiation
dosimetry in nuclear medicine. Appl Radiat Isot. 1999;50(1):73-87.
7. Shousha HA, Farag H, Hassan RA. Measurement of doses to the extremities
of nuclear medicine staff. Radiation Effects and Defects in Solids.
2010;165(1):16-22.
8. Cabral G, Amaral A, Campos L, Guimaraes MI. Investigation of maximum
doses absorbed by people accompanying patients in nuclear medicine
departments. Radiat Prot Dosimetry 2002;101(1-4):435-8.
9. Mountford PJ, O’Doherty MJ. Exposure of critical groups to nuclear
medicine patients. Appl Radiat and Isot. 1999;50(1):89-111.
10. Chiesa C, De Sanctis V, Crippa F, Schiavini M, Fraigola CE, Bogni A, et al.
Radiation dose to technicians per nuclear medicine procedure: comparison
between technetium-99m, gallium-67, and iodine-131 radiotracers and
fluorine-18 fluorodeoxyglucose. Eur J Nucl Med. 1997;24(11):1380-9.
11. Sudbrock F, Boldt F, Kobe C, Eschner W, Schicha H. Radiation exposure in
the environment of patients after application of radiopharmaceuticals Part
1: Diagnostic procedures. Nuklearmedizin. 2008;47(6):267-74.
12. Gomez-Palacios M, Terrón JA, Domínguez P, Vera DR, Osuna RF. Radiation
doses in the surroundings of patients undergoing nuclear medicine
diagnostic studies. Health Phys. 2005;89(2):S27-34.
13. Smart R. Task-specific monitoring of nuclear medicine technologists
radiation exposure. Radiat Prot Dosimetry. 2004;109(3):201-9.
14. International Atomic Energy Agency. Nuclear Medicine Resources Manual.
Vienna: International Atomic Energy Agency; 2006.
15. ICRP. ICRP Publication 60. Oxford: Pregamon Press; 1990.
16. Eurisys Mesures. Owner’s manual board of ionization portable Babyline 81.
Manufacturer Eurisys Mesures, France; 1991.
17. Thrall JH, Ziessman HA. Medicina Nuclear. Rio de Janeiro: Guanabara
Koogan; 2003.
18. IBGE [homepage on the Internet]. Available from: http://www.ibge.gov.
br/home/estatistica/populacao/condicaodevida/pof/2008_2009_encaa/
tabelas_pdf/tab1_1.pdf.
19. American College of Radiology. Practice guideline for the performance of
adult and pediatric skeletal scintigraphy. ACR Council; 2007.
20. Donohoe KJ, Brown ML, Collier BD, Carretta RF, Henkin RE, Royal HD, et al.
Procedure guideline for bone scintigraphy 3.0. Reston: Society of Nuclear
Medicine; 2003.
21. Taylor AT, Blaufox MD, Dubovsky EV, Fine EJ, Fommei E, Granerus G, et
al. Procedure guideline for diagnosis of renovascular hypertension 3.0.
Reston: Society of Nuclear Medicine; 2003.
22. American College of Radiology. Practice guideline for the performance of
adult and pediatric renal scintigraphy. ACR Council; 2008.
23. Sapienza MT, Buchpiguel CA, Costa PLA, Watanabe T, Ono CR. Manual of
Procedures. São Paulo: Institute of Radiology at HCFMUSP; 2003.
24. Mandell GA, Eggli DF, Gilday DL, Leonard JC, Miller JH, Nadel HR, et al.
Procedure guideline for renal cortical scintigraphy in children. Society of
Nuclear Medicine; 2003.
25. Schwarcke M, Cardoso DDO, Ferreira N. Comparação entre detectores
utilizados para medidas ambientais em serviços de medicina nuclear. C&T
Revista Militar de Ciências e Tecnologia. 2009;26:26-32.
26. Thermo Fisher Scientific. Specification of product identiFINDER isotopo
identifier. USA; 2007.
Artigo Original
An occupational dose distribution study in a
positron emission tomography service
Estudo da distribuição de dose ocupacional em um
Serviço de Tomografia por Emissão de pósitrons
Ana Luiza S. L. Kubo and Cláudia Lúcia P. Maurício
Instituto de Radioproteção e Dosimetria (IRD/CNEN)/Serviço de Monitoração Individual Externa – Rio de Janeiro (RJ), Brazil.
Abstract
Positron emission tomography (PET) is a powerful diagnostic tool, especially for Oncology. In PET procedures, the hands exposition of the
workers is potentially higher than the thorax exposition due to the direct handling of the high-energy photons radionuclide. As the dose
distribution in the extremities is non-uniform, the conventional monitoring methods (dosimetric ring and bracelet) may underestimate the
skin dose equivalent in the most exposed part of the hand, which usually are the fingertips. In this study, two PET services had their workers
monitored during the tasks of preparation and injection of the radiopharmaceutical 18F-fluorodeoxyglucose (18F-FDG) in patients, using chips
of LiF:Mg,Cu,P thermoluminescent dosimeters (TLD-100H). Each employee worn TLD sets attached on the wrist and fingers of the dominant
hand, and on the thorax. The highest dose values were measured on the index finger, which received doses up to 0.4 mSv in a single
procedure of 18F-FDG dose preparation and 0.27 mSv in one injection. In a potential annual dose extrapolation, assuming that this technician
performs 840 PET scans (preparation and injection), with these doses values, in one year, his skin dose equivalent on the index finger would
be 564 mSv, exceeding the annual skin dose equivalent limit of 500 mSv. Despite the hands dose distribution is very sensitive of how to
hold the syringe, the dose near to the index fingertip are always the highest, can be, respectively, 4 and 12 times greater than in the position
where dosimetric rings and bracelets are commonly used for routine individual monitoring. Thus, extremity individual monitoring, in addition
to the mandatory whole body individual monitoring with thorax dosemeters, are important tools for occupational dose optimization and should
also be mandatory for PET technician.
Keywords: positron-emission tomography, occupational exposure, dosimetry, TLD.
Resumo
A tomografia por emissão de pósitrons é uma poderosa ferramenta de diagnóstico, especialmente para a Oncologia. Nos procedimentos
de tomografia por emissão de pósitron, a exposição das mãos dos trabalhadores é potencialmente maior do que a do tórax devido ao
manuseio direto de radionuclídeos com fótons de alta energia. Já que a distribuição da dose nas extremidades não é uniforme, os
métodos de monitoramento convencionais (anel dosimétrico e pulseira) podem subestimar a dose equivalente da pele na parte mais
exposta da mão, que costuma ser a ponta dos dedos. Neste estudo, dois serviços de tomografia por emissão de pósitrons tiveram seus
trabalhadores monitorados durantes as práticas de preparação e injeção do radiofármaco 18F-fluordeoxiglicose ( 18F-FDG) em pacientes,
usando dosímetros termoluminescentes LiF:Mg,Cu,P (TLD-100H) em forma de pastilhas. Cada funcionário utilizou um conjunto de
TLDs afixados ao punho e aos dedos da mão dominante, e no tórax. Os valores de dose mais altos foram medidos no dedo indicador,
que recebeu doses de até 0,4 mSv em um único procedimento da preparação de dose de 18F-FDG e 0,27 mSv em uma injeção. Numa
extrapolação da potencial dose anual, considerando que este técnico realize 840 exames de tomografia por emissão de pósitron
(preparação e injeção), com esses valores de doses, em um ano, seu equivalente de dose na pele do dedo indicador seria 564 mSv,
ultrapassando o limite anual de dose equivalente na pele de 500 mSv. Apesar de a distribuição de dose nas mãos ser muito sensível de
acordo com a forma de segurar a seringa, a dose próxima à ponta do dedo indicador será sempre a maior e pode ser, respectivamente,
12 e 4 vezes maior do que na posição em que as pulseiras e os anéis dosimétricos são comumente utilizados para monitoramento
individual periódico. Desse modo, o monitoramento individual da extremidade, além do monitoramento individual de corpo inteiro
obrigatório com dosímetros do tórax, são ferramentas importantes para a otimização da dose ocupacional e devem ser obrigatórias para
o técnico da tomografia por emissão de pósitrons.
Palavras-chave: tomografia por emissão de pósitron, exposição ocupacional, dosimetria, TLD.
Corresponding author: Ana Luiza S. L. Kubo – Instituto de Radioproteção e Dosimetria (IRD/CNEN) – Av. Salvador Allende s/n – CEP: 22780-160 – Rio de
Janeiro (RJ), Brasil – E-mail: [email protected]
185
Kubo ALSL, Maurício CLP
Positron emission tomography (PET) is a powerful diagnostic tool, particularly in Oncology, where its application
is growing continuously and becoming a routine procedure
in Nuclear Medicine Services. In this type of procedure,
doses are transferred from vials to syringes and, then, administered to patients, resulting in non-uniform occupational dose. This dose is mainly due to high-energy photons
resulting from the annihilation of positrons from the decay
of the 18F-fluorodeoxyglucose (18F-FDG)1, which nowadays
is the unique commercially available radiopharmaceutical
in Brazil for PET studies. The large and growing number
of patients undergoing PET procedures and workers involved in this practice warrants continued efforts to improve
the quality of diagnosis and to reduce the radiological risk
associated.
The radiation dose to a technician who performs PET
scans is usually greater than the dose for the same function in conventional nuclear medicine, considering the
same number of procedures. However, the doses in PET
can be quite variable, because in this practice professionals handle the radioactive material using syringes and vials
partially shielded, making a directional radiation field, which
does not occur with conventional procedures.
The routine monitoring of workers is an important part
of any radiation protection program and is performed,
among other reasons, to verify and demonstrate compliance with dose limits regulated, besides giving information
on work practices. In Brazil, external individual monitoring
with dosimeter located on the thorax is compulsory for
all workers in controlled areas. Extremity dosimeters are
recommended in activities where the hands dose can be
much higher than on the thorax, but is not compulsory. In
PET procedures, the radiation risk on the hands is much
higher than on the thorax. To worsen the occupational skin
dose evaluation, in this case, the dose distribution on the
hands is not homogeneous. The highest doses are usually
received by the fingertips2. The difficulty in estimating the
exposure of the most exposed part of the fingers is exacerbated by the conventional method used to determine the
dose received by the hands skin, using dosimetric bracelets or rings. Even rings underestimate the skin dose of
the finger most exposed part. Therefore, special attention
must be paid in positioning the extremity dosimeters3.
This paper presents results of a study of occupational
dose distribution in the two most critical activities of PET
services: 18F-FDG preparation and injection. The measurements are made with thermoluminescent dosimeters
(TLDs) on all fingers of the dominated hand, on the wrist
and on the thorax of each worker.
Methodology
The study of doses of occupationally exposed individuals
was done in two Brazilian PET services (S1 and S2). The
186
workers evaluated were those that performed tasks of
preparation of the radiopharmaceutical for its subsequent
administration in the patient, and of injection the radiopharmaceutical into the patient during PET procedures for oncologic purposes. Both practices were chosen because of
the proximity of professionals with radioactive material.
In both services studied, for dose preparation, the technician uses a manual device shielded with 30 mm lead to fractionate doses. The dose is then transferred to a syringe protected
by a shield with 6 mm tungsten, and the activity is measured in
a dose calibrator and taken to the patient to be injected. After
injection, the technician takes back the syringe in the shield to
the radiopharmacy. The same worker performs the tasks of
dose preparation and injection of the radiopharmaceutical.
To make the dosimetry of professionals, discs with
3.6 mm in diameter and 0.38 mm thick of Harshaw
LiF:Mg,Cu,P TLD (TLD-100H) were used. The dosimeters
were evaluated in a semi-automatic 5500 Harshaw reader.
Individual sensitivity factors are used for the TLD in order
to reduce the uncertainties. The operational quantity used
for the measurements is Hp(10) for thorax dosimetry and
effective dose estimation, and Hp(0.07) for finger and wrist
dosimetry and skin dose equivalent. Calibration in Hp(0.07)
and Hp(10) are performed, respectively, in ISO rod and slab
phantoms, according to ISO 4037-34 and ISO 127945.
The dosimeters were placed in blister of pills cut into
individual cavities. In S1, the dosimeters were placed,
with the aid of adhesive tape, in 8 points of the right hand
(Figure 1), one point in the right wrist (where the dosimetric
bracelet is commonly used) and one on the thorax (trapped
in the official dosimeter) of the worker in each procedure. In
S2, only 5 points were measured in the right hand, corresponding to points A, B, C, D and E in the Figure 1, which
are the points closer to the radioactive material during handling. TLDs were placed immediately before the beginning
of each practice (preparation and injection of radiopharmaceuticals) and removed soon after it ends.
During all measurements, parameters such as service,
place where the monitoring were performed, date, type of
procedure (preparation or injection), number and position
of each dosimeter, activity handled, exposure time, professional name (recorded in code) and any other special comments were recorded.
0,30
Hp (0.07) (mSv)
Introduction
S
1
0,25
0,20
0,15
0,10
0,05
0,00
A
B
C
D
E
F
G
Monitored point
H
I
Figure 1. Location of monitored points at the workers’ hand.
An occupational dose distribution study in a positron emission tomography service
Results
In S1, six procedures of preparation and six of injection, all
performed by one technician, were evaluated. In S2, five
preparations and five injections performed by one technician and one injection performed by another technician
were measured.
In S1, the mean time spent both in the dose preparation and in the injection of 18F-FDG was about 50 seconds.
The activity manipulated and injected in each patient was
about 370 MBq. In S2, the mean time spent in the preparation was approximately 160 seconds and in the injection, 37 seconds. Each syringe was filled, in this case, with
about 300 MBq.
Table 1 shows the results of the TLD measurements
evaluated points. The reported values are mean values per
procedure.
As expected, thorax doses, both in preparation and injection, were much lower than in the hands, because they
are always closer to the source. In preparation, the thorax
is additionally shielded by the bulkhead in front of the table
handling, giving doses even lower than in the injection. The
dose in the fingers during preparation is about two orders
of magnitude higher than in the thorax and, in the injection
procedure, one order higher. These results show clearly
that hands routine individual monitoring, besides thorax
dosimetry, is always recommended for PET technicians,
but it is rarely used today in Brazil.
The doses received by the workers’ fingers vary widely
depending mainly on the form in which the PET service
employee holds the syringe. However, all dose values
measured in this work during the radiopharmaceuticals
preparation is higher than in the injection process, some
cases reaching values four times higher. The highest doses
were obtained always on the index finger, especially at the
point closest to the fingertip (B).
Taking into account both the preparation and the injection, the hands point with the highest dose evaluated (B)
may received a dose four times higher than the point where
the dosimetric rings are commonly worn (E) and 11 times
higher than the point where dosimetric bracelets are worn
(I). Then, for extremity individual monitoring of PET workers
services, the use of dosimetric bracelet should be avoided
and much attention should be given to the positioning of
the dosimetric ring to do not underestimate the hands skin
dose. Figure 2 shows the hands dose distribution of the
three PET technicians.
The mean dose values on the hands depend also
on the activity manipulated and on the experience of the
technician. Greater experience, less time spent to perform
each function, then less dose.
For preparation, at S2, even with higher times, the
doses are still lower at the measured hand; this is probably
due to the fact that all measurements were made only on
the dominated hand, which is not always the one located
closer to the source, i.e., the most exposed. Measures in
both hands are now in progress.
Table 1. Mean measured dose per procedure.
Monitored
point
Hand
A
B
C
D
E
F
G
H
I
Thorax
Preparation
S1
0.29
0.40
0.36
0.26
0.26
0.20
0.19
0.15
0.06
0.001
S2
S1
Hp(0.07) (mSv)
0.08
0.15
0.16
0.27
0.11
0.22
0.10
0.17
0.07
0.17
0.07
0.07
0.05
0.02
Hp(10) (mSv)
< LID*
0.01
Injection
S2
0.02
0.15
0.14
0.08
0.08
< LID*
*<LID = Value lower than the lower detection limit of the system.
Figure 2. Dose distribution in the hands of three different technicians performing the task of injection at PET services.
Using the Table 1 data, an extrapolation of the potential annual dose received by the S1 technician was made,
whereas he performs about 840 PET examinations per year,
including preparation and injection. The results are presented
in Table 2, showing the possibility of the skin of his index
finger (point B) receiving equivalent doses (estimated by the
TLD Hp(0.07) measurements) higher than the extremity annual skin equivalent dose limit of 500 mSv/year6,7. The annual
skin dose equivalents in all other points of the fingers are
lower than the dose limit, but surpass the skin (and extremity)
annual dose equivalent investigation level of 150 mSv/year6,7.
187
Kubo ALSL, Maurício CLP
Table 2. S1 technician potential annual dose on PET preparation
and injection procedures.
Monitored
point
A
B
C
D
E
F
G
H
I
Thorax
Potential annual dose (mSv)
Preparation
Injection
244.19
128.13
334.07
230.05
301.65
183.30
220.12
142.43
214.85
138.37
165.94
61.36
155.05
60.29
128.21
37.70
52.92
19.28
0.88
8.67
Total
372.32
564.12
484.95
362.56
353.22
227.29
215.34
165.90
72.20
9.55
On wrist, the dose is below the skin dose equivalent investigation level. Considering the value of Hp(10) measured
on the thorax as an estimate of the effective dose, this
one exceeds the annual effective dose investigation level
of 6 mSv/year, but not its annual limit.
Conclusions
The external radiation doses measured on the thorax and
hands of workers doing procedures of 18F-FDG preparation
and injection for PET examinations confirm, as expected,
that the values are high and not homogeneous. Thus, extremity individual monitoring, in addition to the mandatory
whole body individual monitoring with thorax dosimeters,
are important tools for occupational dose optimization and
should also be mandatory for PET technician.
Despite the great variation in the distribution of the skin
doses in the hands for different PET employees, which depends on the different way they hold the 18F-FDG syringes,
the point of highest exposition on the hands are normally
the index fingertip (B). The highest skin dose equivalent
does not happen on their wrist (I) or their middle finger
188
base (E), where extremity dosimeters are usually worn.
This work evaluated only the dominated hand, but, as observed during the measurements, sometimes the highest
dose may occur on the other hand. Then, it is necessary to
continue this work to better map both hands of the professionals in PET procedures to check the points that receive
the highest doses, in order to aid better position of the
dosimetric ring.
In PET service, constant optimization of radiation protection is essential, because it is possible that some doses
surpass annual individual dose limits. Occupational exposures can be minimized through good planning, good
practice, education program and patient cooperation. As
the dose is directly related to exposure time and manipulated activity, it is necessary to focus on basic recommendations of radiation protection, including time, distance
and shielding. Appropriated routine external individual
monitoring can give very important information for dose
optimization programs.
References
1. Ginjaume M, Pérez S, Duch MA, Ortega X. Comparison of TLD-100 and
MCP-Ns for use as an extremity dosemeter for PET nuclear medicine staff.
Radiation Measurements. 2008;43:607-10.
2. Wrzesien M, Olszewski J, Jankowski J. Hand exposure to ionising radiation
of nuclear medicine workers. Radiat Prot Dosimetry. 2008;130(3):325-30.
3. European Commission. Technical recommendations for monitoring
individuals occupationally exposed to external radiation - RP160; 2009.
4. International Organization for Standardization. ISO 4037-3, Part 3. X and
gamma reference radiation for calibrating dosemeters and doserate meters
and for determining their response as a function of photon energy; 1999.
5. International Organization for Standardization. ISO 12794. Nuclear energy
– Radiation protection – Individual thermoluminescence dosimeters for
extremities and eyes; 2000.
6. Comissão Nacional de Energia Nuclear. Diretrizes básicas de proteção
radiológica, CNEN-NN-3.01. Brasil; 2005.
7. International Commission on Radiological Protection. The 2007
recommendations of the International Commission on Radiological
Protection, Publication 103. 2007;37(2-4).
Artigo Original
BSc in Medical Physics at the
University of Campinas
Bacharelado em Física Médica na
Universidade Estadual de Campinas
Roberto J. M. Covolan1, Rosângela F. Coelho2 and Gabriela Castellano1
2
1
Neurophysics Group, Gleb Wataghin Physics Institute (UNICAMP) – Campinas (SP), Brazil.
Medical Physics Division, Center for Biomedical Engineering (UNICAMP) – Campinas (SP), Brazil.
Abstract
Several Medical Physics university programs have started in Brazil in the last decade. They are mainly taken at the undergraduate level. At UNICAMP,
we combined the teaching expertise of several units, including the Physics Institute, the Institute of Biology, the School of Medical Sciences and
the Center for Biomedical Engineering, to create a course that starts with Core Physics, Mathematics, Computing and Chemistry, then follows
with specific courses in Medical Physics intertwined with advanced courses belonging to a strong BSc in Physics, and concludes with a one-year
traineeship at the University Hospital. The UNICAMP Medical Physics undergraduate course began in 2003, lasts five years and has formed 67
students up to 2010, most of them have got stable working positions as Medical Physicists. This article presents the conception, structure and the
first outcomes of this course.
Keywords: Unicamp, medical physics, undergraduate courses.
Resumo
Diversos programas universitários de Física Médica se iniciaram no Brasil na última década. Tais programas são principalmente realizados nos
cursos de graduação. Na Universidade Estadual de Campinas (UNICAMP), combinou-se a técnica de ensino de diversas unidades, incluindo o
Instituto de Física, o Instituto de Biologia, a Faculdade de Ciências Médicas e o Centro de Engenharia Biomédica, para criar um curso que começa
com Física Básica, Matemática, Computação e Química. Segue-se com cursos específicos em Física Médica, mesclando-os com cursos avançados
pertencentes a um ótimo bacharelado em Física, e conclui-se com um estágio de um ano no Hospital Universitário. O curso de graduação em Física
Médica da UNICAMP começou em 2003, tem duração de cinco anos e já graduou 67 estudantes até 2010, a maioria deles com cargos profissionais
estáveis como físicos médicos. Este artigo apresenta a concepção, a estrutura e os primeiros resultados desse curso.
Palavras-chave: UNICAMP, física médica, cursos de graduação.
Introduction
The UNICAMP Medical Physics undergraduate course was
conceived in 1999 by several university units, coordinated
by the Gleb Wataghin Physics Institute (IFGW), in order to
meet the growing demand for physicists in the medical
field. Many studies were conducted on the composition of
the curriculum and course programs, until the final program
was reached in 2001. The course was implemented in 2003,
when the first class of Medical Physics began at UNICAMP.
Because it is an interdisciplinary program, also participated
in its design and currently participate in its implementation
lecturers from the Institute of Biology (IB), from the School of
Medical Sciences (FCM) and from the Center for Biomedical
Engineering (CEB), everyone with specific contributions related to their research and work fields. The course’s final year
consists of traineeship at the UNICAMP University Hospital in
the areas of Nuclear Medicine, Radiotherapy, Radiology and
Radiation Protection conducted by CEB Medical Physics
division. Thus, over the five-year duration of the course, students have contact with professionals with extensive experience in each of the covered topics. So far, UNICAMP has
formed four Medical Physics classes, totalizing 67 students.
Course description
Course summary
The disciplines that make up the BSc in Medical Physics
are organized in three blocks corresponding to three different levels that are subsequently accessed by the students
as they progress along the course.
Corresponding author: Roberto J. M. Covolan – Gleb Wataghin Physics Institute – Rua Sérgio Buarque de Holanda, 777 – Cidade Universitária Zeferino Vaz –
CEP: 13083-859 – Campinas (SP), Brazil – E-mail: [email protected]
189
Covolan RJM, Coelho RF, Castellano G
The student starts taken classes in a four-semester
basic course that is offered to those enrolled in the hard
sciences. The program of this short course, which offers
a background in Physics, Mathematics, Computing and
Chemistry, is shown in Table 1.
After that, the student is prepared to take advanced
classes of the block that includes specific courses of the
BSc in Medical Physics intertwined with advanced courses of the BSc in Physics. These courses are presented
in Table 2. Again, this block takes two semesters to be
completed.
The third and final block consists basically of a twosemester traineeship in Medical Physics, which is performed at the UNICAMP University Hospital (Table 3). This
block has also an introductory course at the hospital called
“Clinical Aspects of Medical Physics” which includes topics of medical ethics and professional relationships, oncology, diagnosis and treatment, and biosecurity.
Radiotherapy: Monitoring of treatments, monitoring
and implementation of treatment planning from simulation
to approval by the physician, dosimetry and quality control
of equipment used in therapy, including brachytherapy for
High Dose Rate.
Nuclear Medicine: Monitoring and implementation of activities of radiation levels assessment and surface contamination, monitoring and implementation of instrumentation
Table 2. Advanced courses taken by Medical Physics students.
Code
BD580
§
F 502
F 520
†
F 540
†
F 589
§
MC202
†
†
Mandatory traineeship
The disciplines of the cited mandatory traineeship (Table 3)
are held during the 5th year of the course and conducted in
the hospital for 720 hours under the supervision of a medical physicist from the respective area.
BD681
§
F 320
F 550
†
F 620
†
F 689
*
§
Table 1. Basic courses offered to students of hard sciences.
Code
*F 128
*F 129
*FM003
*MA111
*MA141
*MS149
*F 228
*F 229
*MA211
*MA327
*MC102
*F 328
*F 329
*MA311
†
ME210
†
MS211
*F 315
*F 428
*F 429
*MA044
†
QG101
†
QG102
Course
1st semester
General Physics I
Experimental Physics I
Profession Related Seminars
Calculus I
Analytical Geometry and Vectors
Mathematics Complements
2nd semester
General Physics II
Experimental Physics II
Calculus II
Linear Algebra
Algorithms and Computer Programming
3rd semester
General Physics III
Experimental Physics III
Calculus III
Probability
Numerical Calculus
4th semester
General Mechanics I
General Physics IV
Experimental Physics IV
Mathematics IV
Chemistry I
Experimental Chemistry I
Credits
20 credits
04
02
02
06
04
02
22 credits
04
02
06
04
06
20 credits
04
02
06
04
04
22 credits
04
04
02
04
04
04
F 604
F 740
§
F 752
§
F 758
§
MC920
†
*
**
F 837
§
F 852
§
F 853
§
F 854
§
Courses in common with the BSc in Physics.
190
Credits
26 credits
04
04
04
04
04
06
20 credits
04
04
04
04
04
20 credits
04
04
04
04
04
20 credits
04
04
04
04
04
*Common core courses; †Courses in common with the BSc in Physics; §Courses specific
of BSc in Medical Physics; **Elective courses: 04 credits among: †F 602 Electromagnetism
II; †F 789 Quantum Mechanics II; §F 856 Biophotonics; †F 885 Elementary Particles and
Fields; †F 888 Solid State Physics
Table 3. Traineeship at the University Hospital.
Code
MD760
§
MD947
§
MD948
§
*Common core courses.
†
Course
5th semester
Fundamentals of Cellular and Molecular
Biology
Electromagnetism
Mathematical Methods for Physics I
Methods for Experimental Physics I
Structure of Matter
Data Structure
6th semester
Fundamentals of Human Anatomy,
Histology and Physiology
Thermodynamics
Radiation: Interaction and Detection
Mathematical Methods for Physics II
Quantum Mechanics I
7th semester
Statistical Physics
Methods for Experimental Physics III
Magnetic Resonance Applied to Medicine
Radiobiology and Radioprotection
Introduction to Digital Image Processing
8th semester
Elective course
Medical Physics Laboratory
Physics of Radiology
Physics of Nuclear Medicine
Radiotherapy Physics
Course
Credits
28 credits
9th semester
Clinical Aspects of Medical Physics
04
Supervised Traineeship on Medical Physics I:
24
Radiology, Radiotherapy & Nuclear Medicine
10th semester
24 credits
Supervised Traineeship on Medical Physics
II: Radiology, Radiotherapy & Nuclear
24
Medicine
Courses specific of BSc in Medical Physics
§
BSc in Medical Physics at the University of Campinas
quality programs, medical image processing, and study of
clinical protocols under development.
Radiology and Radiological Protection: Tracking surveys,
monitoring and implementation of quality programs in radiology and in radiation measurement equipment, monitoring
and implementation of radiometric surveys and testing of
radiation leakage in X-ray equipment, individual monitoring
management and radioactive waste management.
All information included in Tables 1-3 were taken from
the UNICAMP undergraduate courses catalogue of 20111.
Graduate Programs
At UNICAMP, the Medical Physics student can pursue
graduation studies at the Department of Biomedical
Engineering (School of Electrical and Computing
Engineering) or at the Physics Institute. The Physics
Institute has only one graduate program. There are, however, lecturers who can guide students at MSc and PhD
research in Medical Physics subjects.
Conclusions
The proposal to create a BSc in Medical Physics at
UNICAMP came from the idea of combining the strong
background in Mathematics and Physics, usually provided
to students of the BSc in Physics, with the extensive educational resources that the University has in the Biomedical
field and Health Sciences.
As mentioned, in four years since the Medical Physics
BSc program started at UNICAMP, 67 students graduated. From these, 29 (43%) have gone to specific traineeship
in Physics of Radiotherapy, which is the preferred area for
those who enter the Medical Physics course.
In the last three years (2008-2010), from the 60 radiotherapy traineeship positions available in Brazil (20 positions/year), 38% have been taken by UNICAMP students.
The remaining students are either working in Physics
of Radiology (7%), Physics of Nuclear Medicine (6%),
Radiation Protection (3%) or other activities including graduate studies. Most of the former students have got stable
working positions. For us, this is the best indication that
our course is achieving its purpose.
However, a series of difficulties still remain, not
specifically related to our course, but mostly related to
working positions for medical physicists in Brazil. Only
Radiotherapy Centers are required by law to hire medical
physicists. For Radiology and Nuclear Medicine, Medical
Physics is still a relatively new area in Brazil. The work
of medical physicists in these activities has been established by law since 1998 and 2005, respectively, but
just as part-time job and, in some cases, it is accepted that the work be carried out by other professionals.
Physicians and clinic owners have not yet understood the
importance of the work of medical physicists, particularly
in what concerns quality control. Therefore, there is still a
large field to be explored by medical physicists in Brazil,
and certainly the job market of this area will grow up in the
near future. Nuclear energy and radioisotope production
are areas which are growing in the country and will certainly need good medical physicists to work in radiation
protection and quality control.
Acknowledgment
We would like to thank the Brazilian governmental agencies CNPq and FAPESP for ﬁnancial support.
References
1. UNICAMP Undergraduate courses catalogue (2011) [internet]. Available from:
http://www.dac.unicamp.br/sistemas/catalogos/grad/catalogo2011/index.html
191
Artigo Original
Residence program on Medical Physics at
Hospital das Clínicas of São Paulo University
Programa de residência em Física Médica no Hospital
das Clínicas da Universidade de São Paulo
Laura N. Rodrigues, Laura Furnari, Marco A. da Silva, Rodrigo A. Rubo, Gabriela R. dos
Santos and Gisela Menegussi
Radiotherapy Department, Hospital das Clínicas of São Paulo University – São Paulo (SP), Brazil.
Abstract
The main goal of the Residence Program on Medical Physics at Clinicas Hospital is to provide a specialization course of 24 months. The
candidate selection is made in 2 steps: a general examination with 50 questions of multiple choices and a specific examination which
does not presuppose the candidates must have certain knowledge in the area. After this second written exam, an interview is promoted in
order to evaluate other aspects not covered by the previous exams, including some ethical issues, the ability to deal with patients and the
multidisciplinary aspect among other professionals in the hospital. The Board Committee also evaluates the candidates’ curriculum according
to rules established by the Residence Coordination (participation in scientific meetings, trainings performed during graduation in Radiotherapy
and scientific publications). The residence is thus directed to students which concluded bachelor’s degree or a degree in Physics, and aims
to train professionals skilled in Radiotherapy area, including the obtaining of further qualified professional title offered by the Brazilian
Association of Medical Physics. The students attend lectures, seminars and the routine of the hospital. They also perform experimental work
and develop a monograph to be presented at the end of the residence’s period. In the last 5 years, almost 50 candidates have been attending
to the selective process for only 2 available positions per year; the candidates come from different parts of the country, including students
from the North, Northeast, Central-West, South and Southeast regions. It should be noticed in the last 5 years that most of the approved
candidates are students coming from Graduation on Medical Physics, especially in the São Paulo state. For the last 10 residents, 20% were
hired by companies in the Radiotherapy area; the remaining residents were hired by hospitals, 20% have found jobs in the Northeast region
and 60% in São Paulo state.
Keywords: education, human resources, radiotherapy.
Resumo
A principal meta do Programa de Residência em Física Médica no Hospital das Clínicas é fornecer um curso de especialização de 24 meses.
A seleção do candidato é feita em duas etapas: uma prova geral com 50 questões de múltipla escolha e uma prova específica que não
pressupõe que os candidatos tenham algum conhecimento na área. Após a segunda prova escrita, uma entrevista é promovida para avaliar
outros aspectos que não foram cobertos nos exames anteriores, inclusive algumas questões éticas, a capacidade de lidar com pacientes e o
aspecto multidisciplinar com outros profissionais no hospital. Uma Banca de Examinadores também avalia os currículos dos candidatos, de
acordo com as normas estabelecidas pela Coordenação da Residência (participação em reuniões científicas, treinamentos realizados durante
a graduação em Radioterapia e publicações científicas). Desse modo, a residência é direcionada a estudantes que concluíram o bacharelado
ou tenham uma graduação em Física, objetiva treinar profissionais peritos na área de Radioterapia, inclusive com a obtenção de um título
de especialista oferecido pela Associação Brasileira de Física Médica. Os estudantes participam de palestras, seminários e da rotina do
hospital. Eles também realizam um trabalho experimental e desenvolvem uma monografia, que deve ser apresentada ao final do período de
residência. Nos últimos cinco anos, quase 50 candidatos têm participado do processo seletivo para apenas duas vagas disponíveis por ano;
os candidatos vêm de diferentes partes do País, inclusive estudantes das regiões Norte, Nordeste, Centro-Oeste, Sul e Sudeste. Deve-se
relatar que, nos últimos cinco anos, a maioria dos candidatos aprovados são estudantes graduados em Física Médica, especialmente no
estado de São Paulo. Quanto aos últimos dez residentes, 20% foram contratados por empresas na área da Radioterapia; os outros foram
contratados por hospitais, 20% conseguiram empregos na região Nordeste e 60% no estado de São Paulo.
Palavras-chave: educação, recursos humanos, radioterapia.
Corresponding author: Laura Natal Rodrigues – Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo (HCFMUSP) – Av. Dr. Eneas
de Carvalho Aguiar, 255, 3º andar – CEP: 05403-000 – São Paulo (SP), Brasil – E-mail: [email protected]
193
Rodrigues LN, Furnari L, Silva MA, Rubo RA, Santos GRD, Menegussi G
Introduction
The history of Medical Physics in Brazil started in 1969
by the introduction of a graduation course at the Physics
Institute of São Paulo University. At that time, due to the
lack of specialized laboratories and appropriate equipments, some Radiotherapy centers shared their equipments, as well as their own physical installation to the
university. In this sense, the Hospital das Clínicas of São
Paulo University School Medicine is the first reference center for training in the Medical Physics area which was established by Alipio Luiz Dias Neto at the Nuclear Medicine
Center of São Paulo University, which belongs to Hospital
das Clínicas.
The program aims to provide Radiotherapy Physics professionals who are capable of acting as experts in the field
of Medical Physics with a specialization in Radiotherapy.
This course, offered by the Institute of Radiology of the
Hospital das Clínicas (InRad - HCFMUSP), is one of the
few programs in the country that offers a comprehensive
course in its area, comprising 3,840 hours of a specialized program. It should be point out that the Radiotherapy
Physics contributes significantly to the quality assurance
and radiation safety for all hospitals in the country.
Methodology
The residence is directed to students that concluded bachelor’s degree or a degree in Physics, aims to train professionals skilled in Radiotherapy area, including the obtaining
of further qualified professional title offered by the Brazilian
Association of Medical Physics.
The course consists of two parts: theoretical (20% of
the total of 3,840 hours) and practical (remaining hours),
both developed at the Institution. Qualified professionals in
classrooms equipped with multimedia will proffer the theoretical part and student teachers will support the scientific
collection available in the library of the Institution. In addition, students participate in scientific events organized by
professionals of the Institution. Professionals with experience in Radiotherapy Physics will supervise the practical
training with experience in Radiotherapy Physics.
Students use the following equipments at the practice: 3 linear accelerators; a CT-simulator; a high-dose-rate
brachytherapy; and physical facilities of the Radiotherapy
Department at Hospital das Clínicas.
At the Radiotherapy Department, there are 5 different
treatment planning systems: Eclipse, iPlan, BrainScan,
XiO and Oncentra. These systems allow treating patients
with special procedures, such as Intensity Modulated
Radiation Therapy (IMRT), Radiosurgery (SRS) and Total
Body Irradiation (TBI) among the conventional techniques.
The Department also has a whole set of dosimetric systems, such as: 3D automatic water phantom, 2D matrix
linear array, ionizations chambers of different sizes with
associated electrometers, termoluminescent dosimeters,
194
semiconductor detectors (diodes and diamond), anthropomorphic phantom, slabs of virtual water, radiochromic
films, among others.
The evaluation criteria comprehend written examinations, seminars and cases discussion. Tests and seminars accomplish the evaluation of the theoretical content.
It is also taken into account the student’s performance in
presenting work at scientific meetings. The seminars are
evaluated by the content itself, performance, teaching
resource, knowledge of the subject, skills of content and
treatment planning in accordance with treatment protocols
adopted by the Institution. The tests are applied in order
to measure if students carried the theoretical knowledge
to the practice performed. They include open questions to
allow better exposure of the knowledge learned.
The assessment of the practical content is performed
by professionals involved in the Residence Program (medical physicists and radiation oncologists), according to the
following criteria: initiative, interest, critical capacity, commitment, responsibility, ethical behaviour, attendance,
punctuality, personal presentation, scientific knowledge,
teamwork, relationship with the Radiotherapy team and
with patients. The issues raised by the supervisor are discussed with students and transformed into opportunities
for improvement. These professionals also assess various
cognitive aspects, such as performance in the daily routine
of the Department of Radiotherapy (simulation, planning of
patient, dosimetry and quality control of equipment).
Every three months an assessment is sent to the
Centre for Training of Hospital das Clínicas containing the
following ranks: theoretical subjects, supervised training,
development of the monograph and evaluated aspects like
ethical and professional attitude. The annual assessment is
performed through a global and final evaluation.
In all assessments are required a cut-off score greater
than or equal to 7.
The completion of course work (monograph) starts
in April when the theme is set with the supervisors. The
development of the research project takes place preferably in May and June, including consultations on relevant
scientific literature and the subject of analysis and correction guidelines.
Data collection is performed in the months of August
to October and is supervised by the supervisor and the
remaining staff, aiming to ensure compliance with the
technical, scientific and ethical precepts. The data analysis phase occurs in November under the supervision of
the supervisor.
The monograph concludes with discussion and conclusion or concluding remarks in December, covering more
queries in the scientific literature. The work of completion
is presented to the team of Medical Physics and member
appraisers in January. The presentation contains introduction, objective, method, results, discussion and conclusion
or closing remarks. The abstracts of papers are published
on the website HCFMUSP (www.hcnet.usp.br). The minimum threshold is always with a score equal to 7.
Residence program on Medical Physics at Hospital das Clínicas of São Paulo University
Syllabus
A theoretical module (40 hours) is given to all residents of
Hospital das Clínicas for all programs, including the Medical
Physics. It covers the following topics: public health policies and health system; introduction to research methodology; and health education.
The completion of the remaining theoretical module
(20% of the total number of hours) contains the following
subjects: dosimetry and radiation physics; clinical treatment planning; radiation protection; brachytherapy; radiobiology; fundamentals of anatomy.
The practical training comprehends 3 major modules,
such as: dosimetry of ionising radiation, treatment planning
and brachytherapy. The bibliography includes the IAEA
Syllabus1, Khan’s2 and Van Dyk’s books3; Jani’s book4;
Bentell’s book5; Godden’s book6; and Hall’s book7. All the
pertinent publications by ICRU, ICRP, NCRP and IAEA
documents are also included in the bibliography employed
by the Medical Physics Program.
Conclusion
In the last 5 years, almost 50 candidates have been attending to the selective process for only 2 available positions per
year. The candidates come from different parts of the country,
including students from the North, Northeast, Central-West,
South and Southeast regions. It should be noticed in the last
5 years that most of the approved candidates are students
coming from Graduation on Medical Physics, especially in the
São Paulo state. For the last 10 residents, 20% were hired
by companies in the Radiotherapy area; the remaining residents were hired by hospitals, 20% have found jobs in the
Northeast region and 60% in São Paulo state.
References
1. Podgorsak EB. Radiation Oncology Physics: A handbook for teachers and
students. Vienna: International Agency of Energy Atomic; 2003.
2. Khan FM. The Physics of Radiation Therapy. New York: Lippincott Williams
& Wilkins; 2009.
3. Dyk JV. Modern technology of radiation oncology. Volumes 1 and 2. New
York: Medical Physics Publishing Corporation; 2005.
4. Jani S. CT simulation for radiotherapy. New York: Medical Physics Publishing
Corporation; 1993.
5. Bentell GC. Radiation therapy planning. New York: McGraw-Hill Professional;
1995.
6. Godden TJ. Physical aspects of Brachytherapy. Medical Physics Handbook,
No 19. London: Taylor & Francis; 1988.
7. Hall EJ, Giaccia AJ. Radiobiology for the radiologist. New York: Lippincott
Williams & Wilkins; 2005.
195
Artigo Original
Analysis of the BOLD responses on EEGfMRI acquisition in patients with epilepsy
Análises da resposta BOLD em aquisições de EEGfMRI (eletroencefalografia e imageam por ressonância
magnética funcional) em paciente com epilepsia
Brunno M. de Campos1, Ana C. Coan1, Guilherme C. Beltramini2,
Roberto J. M. Covolan2 and Fernando Cendes1
Department of Neurology of the State University of Campinas (UNICAMP) – Campinas (SP), Brazil.
2
Gleb Wataghin Physics Institute of UNICAMP – Campinas (SP), Brazil.
1
Abstract
The technique of magnetic resonance imaging (MRI), characterized by high spatial resolution, associated with electroencephalography (EEG),
characterized by high temporal resolution, can be a powerful tool to study neurological disorders, including epilepsy. Electroencephalography is a
mechanism to record electrical brain activity, using principles of Electronics, Physics and Physiology. Functional MRI (fMRI) relies on the different
magnetic properties of blood depending on its oxygen content. The goal of functional imaging is to obtain images that are sensitive to brain function.
To this end, we aimed to understand the mechanisms of neural activity and the processes related to it. This paper describes the methodology of the
combined EEG-fMRI method in a tertiary hospital, and assesses the results of 16 exams regarding their concordance with the clinical history and
clinical applicability of the technique. The results of the exams were statistically analyzed with the software SPM8. BOLD (Blood Oxygenation Level
Dependent) responses were analyzed considering the clinical history of each volunteer. The studies that showed statistically significant activation
areas consistent with clinical history were considered compatible (seven exams) while the studies with results not consistent were considered
incompatible (one exam). Six compatible exams were considered clinically relevant, because they add information regarding the definition of the
epileptogenic zone or epileptic syndrome. The studies with absence of epileptiform activity on EEG (three exams) or significant BOLD activations (three
exams) were considered null. Two exams were excluded due to excessive head motion. EEG-fMRI is a promising technique that can be important to
improve the understanding of neurological disorders, including epilepsy. The method may be used in the future as an important diagnostic tool for
refractory epileptic patients as it may add information about the localization of epileptogenic zone or definition of epileptic syndrome.
Keywords: EEG-fMRI, neurology, fMRI, epilepsy.
Resumo
A técnica de imagem por ressonância magnética (MRI), caracterizada pela alta resolução espacial, associada com a eletroencefalografia (EEG), caracterizada
pela alta resolução temporal, pode ser uma ferramenta poderosa para estudar distúrbios neurológicos, inclusive a epilepsia. A eletroencefalografia é um
mecanismo feito para registrar a atividade elétrica cerebral, que utiliza princípios de Eletrônica, Física e Fisiologia. A imagem por ressonância magnética
funcional (fMRI) conta com diferentes propriedades magnéticas do sangue, dependendo do seu conteúdo de oxigênio. A meta da visualização funcional
é obter imagens que sejam sensitivas à função cerebral. Para esse fim, objetivamos entender os mecanismos da atividade neural e seus processos
relacionados. Este trabalho descreve o método combinado de eletroencefalografia e imagem por ressonância magnética funcional (EEG-fMRI) em um
hospital terciário, e avalia os resultados de 16 exames quanto à concordância deles com o histórico clínico e a aplicabilidade clínica da técnica. Os
resultados dos exames foram analisados estatisticamente com o software SPM8. As respostas BOLD foram analisadas levando em consideração o
histórico clínico de cada voluntário. Os estudos que mostraram áreas de ativação estatisticamente significantes consistentes com o histórico clínico
foram considerados compatíveis (sete exames), enquanto os estudos com resultados não consistentes foram considerados incompatíveis (um exame).
Seis exames compatíveis foram considerados clinicamente relevantes, pois adicionam informações a respeito da definição da zona epileptogênica ou
síndrome epilética. Os estudos com ausência de atividade epileptiforme na eletroencefalografia (três exames) ou ativações BOLD significantes (três
exames) foram considerados nulos. Dois exames foram excluídos devido ao excessivo movimento da cabeça. A eletroencefalografia e a imagem por
ressonância magnética funcional é uma técnica promissora que pode ser importante para melhorar o entendimento dos distúrbios neurológicos, inclusive
a epilepsia. O método pode ser usado futuramente como uma importante ferramenta de diagnóstico para pacientes com epilepsia refratária, já que pode
acrescentar informações sobre a localização da zona epileptogênica ou sobre a definição da síndrome epilética.
Palavras-chave: eletroencefalografia e imagem por ressonância magnética funcional, neurologia, imagem por ressonância magnética funcional, epilepsia.
Corresponding author: Brunno Machado de Campos – Cidade Universitária “Zeferino Vaz” Barão Geraldo (UNICAMP) – Vital Brasil, 251 – CEP: 13083-970 –
Campinas (SP), Brasil – E-mail: [email protected]
197
Campos BM, Coan AC, Beltramini GC, Covolan RJM, Cendes F
Introduction
General introduction
The technique of magnetic resonance imaging (MRI),
characterized by high spatial resolution, associated with
electroencephalography (EEG), characterized by a high
temporal resolution, can be a powerful tool to study neurological disorders, including epilepsy.
Epilepsy is the most common neurological disorder,
with prevalence of 2% of world population. It can cause
serious consequences, including injuries, psychological
problems, mental disorders and even sudden death. Since
it is such a common disease and has a great impact on the
lives of the affected ones, the establishment and development of techniques that can better guide the assessment
of each case is of great value, increasing the chances of
curing or improving the quality of patient’s life.
The simultaneous acquisition of functional magnetic resonance imaging (fMRI) and EEG can be in a near future a
valuable technique to the analysis and localization of neural
activity and a complementary diagnostic method in clinical
routine of patients with epilepsy.
This paper describes the methodology of the combined
EEG-fMRI method in a tertiary hospital and assesses the
results of 16 exams regarding their concordance with the
clinical history and clinical applicability of the technique.
Functional magnetic resonance imaging
The goal of functional imaging is to obtain images that are
sensitive to brain function. To this end, we aim to understand the mechanisms of neural activity and the processes
related to it. There are electrophysiological methods that
directly measure neural activity, but they are very invasive.
FMRI is noninvasive and reveals the neural activity by the
assessment of hemodynamic changes associated with it1.
The study of brain functions requires acquisition methods with similar speeds to the physiological changes
of interest. The technique of EPI (Echo Planar Imaging)
scans the whole brain with strong gradients, covering the
k-space in a rectangular manner on the order of seconds,
which is the same order of magnitude of the hemodynamic
response.
Since brain does not store energy, the ATP (adenosine
triphosphate) must be formed mainly by oxidation of blood
glucose. Experiments show that this consumption is concentrated in regions with increased neural activity.
Increased blood flow causes an increased transport of
glucose and oxygen to the site of activation to meet the
energy needs of nerve cells. The increase in local perfusion
rate leads to dilution of venous deoxyhemoglobin (deoxygenated hemoglobin) and thus increase the concentration
of oxyhemoglobin (oxygenated hemoglobin)1.
Since oxyhemoglobin is diamagnetic (displays weak
repulsion in a magnetic field) and deoxyhemoglobin is
paramagnetic (displays attraction to a magnetic field),
the presence of deoxyhemoglobin creates a higher local
magnetic field. It is known that the magnetic susceptibility
198
of deoxygenated blood is around 20% higher than that of
oxygenated blood. This introduces local inhomogeneities
and, therefore, variation in proton precession frequencies,
which reduces T2* (transverse relaxation time including
field inhomogeneity)2.
In other words, the area of activation has a decreased
concentration of deoxyhemoglobin and consequently an
increase in the magnetic signal. Hence, this is called BOLD
(Blood Oxygenation Level Dependent) signal and is the basis of fMRI.
Electroencephalography
Neurons have the ability to communicate quickly and accurately via action potentials (rapid membrane depolarization that propagates through the axon until its terminal). On
average, a neuron can form one thousand synapses and
receives more than ten thousand connections. When the
action potential reaches the axon terminal, the presynaptic
neuron releases neurotransmitters that will affect the membrane potential of the postsynaptic cell. This postsynaptic
potential can be either excitatory or inhibitory. The neuronal
information integration of the signals received from other
cells occurs at the axon hillock (structure between the cell
body and the axon), in which another action potential will
be fired only if the threshold potential is crossed.
Electroencephalography is a mechanism to record
electrical brain activity, using principles of Electronics,
Physics and Physiology. The electrical activity recorded
in EEG derives mainly from synchronized pyramidal cell
postsynaptic potentials. Action potentials do not contribute to EEG signals because the fields generated by them
decay faster with distance and they have a short duration
(1-2 msec), overlapping much less in time – postsynaptic
potentials last about 10-250 msec. Moreover, the EEG signal can be measured at a considerable distance from the
source if the responsible neurons are regularly arranged
and activated in a fairly synchronous way. These properties
hold for pyramidal cells3.
The potential difference between the electrodes and a
reference electrode is measured, and a conductive gel is
used to decrease the electrical resistance between electrode and scalp. After signal measurement, the signal is
amplified and recorded.
Method and equipment
Clinical procedures
The procedures involved in preparing the experiment can
be divided into three steps: 1) selection of a patient or volunteer, who meets the requirements and has all the physical and legal conditions for submission to the procedure;
2) preparation, assembly and obtainment of reference results of the EEG out of the scanning room; 3) positioning of
the patient inside the MRI scanner (Philips Achieva 3T) and
simultaneous fMRI and EEG (Brain Products, München,
Germany) acquisition.
Analysis of the BOLD responses on EEG-fMRI acquisition in patients with epilepsy
The positioning of the cap with 64 electrodes follows
the international standard 10-20 system. This process requires a strong electrical coupling between electrodes and
scalp (impedance below 5 kilo-ohms). On average eight,
EPI sequences of six minutes are performed, depending
on the patient conditions and possible emergencies. The
technical characteristics of EPI were TE=30 ms, TR=2s,
240x240x117 mm³ FOV, 39 slices, 3x3x3 mm³ voxel and
180 volumes.
Computational procedures
The EEG is corrected for gradient and ballistocardiogram
artifacts using the software Brain Vision Analyzer2 (Brain
Products, München, Germany). The gradient artifact is
the sum of the interference generated by the magnetic
gradients and the radiofrequency pulses. They generate
electric current in the electrodes due to electromagnetic
induction, masking the patient’s signal. The correction
method is the AAS (Average Artifact Subtraction), which
subtracts out the sliding average (21 intervals window)
from every TR interval, since the pulse sequence is exactly the same in each of these intervals. The heartbeat artifact is basically generated by small oscillations due to the
patient’s blood flow and heart rate, and is corrected in the
same way.
The EEG is then examined by a specialist in order to locate
epileptiform activity, whose timing and duration are recorded.
The preprocessing and statistical analysis of fMRI data
are performed in SPM8 software (Wellcome Trust Centre
for Neuroimaging, London, England). The fMRI images are
preprocessed (realigned, slice timing corrected, normalized and smoothed) and the timings of the epileptiform
activity are used as task in an fMRI paradigm to look for
BOLD changes in the signal.
The results of 16 exams were statistically analyzed with the
program SPM8. BOLD responses were analyzed considering the clinical history of each volunteer. The results were
then divided in compatible, incompatible or null (Table 1).
The studies that showed statistically significant activation
areas consistent with the clinical history were considered
compatible (seven exams). The studies with BOLD activations with no apparent relation to the clinical history were
considered incompatible (one exam). The studies were
considered null for two different reasons: i) absence of epileptiform activity on EEG recorded inside the MRI (three
exams) or ii) absence of significant BOLD activations (three
exams). Since excessive head movement during the acquisition can result in false positive or negative results, we
limited the accepted movement to 3 mm (voxel size in the
EPI sequence) in the three coordinate axes. Two exams
were excluded due to excessive head motion.
The null results related to the absence of epileptiform
activity were expected and compatible with previous
studies. It is known that the probability of an individual with
epilepsy to have a normal EEG is close to 50%. This problem can be attenuated by the adequate choice of patient
who may have previous routine EEG with frequent epileptiform abnormalities, and by appropriate number of EPI sequences, increasing as much as possible the time of the
EEG-fMRI acquisition. The null results related to absence
of BOLD activation were also expected and described in
the literature4. It may be related to individual responses or
to the necessity of different hemodynamic response function for different individuals. However, in our specific cases,
we believe it occurred because of the small number of epileptic markers in these exams (two to six markers)4.
The exam with incompatible result is also eventually
expected4. But, in this case, it may be related to problems in the acquisition or data processing, or it may also
be secondary to clinical data controversy of this specific
patient.
A second classification of the studies was based on
the clinical usefulness of EEG-fMRI results. The tests with
incompatible or null findings and the tests discarded by
head movement were considered not clinically useful (total of nine exams). One study with compatible result was
considered with no clinical utility, once it did not add or
reinforce any information that could have improved patient
diagnosis or treatment. All the other compatible exams
were considered clinically relevant.
The clinically relevant studies add two types of information: location of the epileptogenic zone and definition
of epileptic syndrome. The epileptogenic zone was well
defined in three patients with ictal recordings (seizures registered during the exam acquisition). In these patients, the
BOLD maps revealed significant activation in areas already
Table 1. Results classification.
Patient
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Ictal markers
0
1CPS
0
0
0
1CPS 4EleSz
1CPS 22EleSz
0
0
0
0
0
0
1CPS 7EleSz
0
1 CPS
Interictal
markers
0
3
4
10
3
10
49
30
6
116
0
189
0
89
2
3
EEG-fMRI
results
Null
Compat.
Null
Compat.
Incompat.
EDEHM
Compat.
Compat.
Null
Compat.
Null
Compat.
Null
Compat.
Null
EDEHM
Clinical utility
No
Yes
No
No
No
No
Yes
Yes
No
Yes
No
Yes
No
Yes
No
No
CPS: Complex Partial Seizure; EleSz: Electrical Seizure; EDEHM - Excluded Due to
Excessive Head Motion; Compat: Compatible; Incompat: Incompatible.
199
Campos BM, Coan AC, Beltramini GC, Covolan RJM, Cendes F
suspected to be the epileptogenic zone in each case. The
areas were consistent with structural MRI in all cases and
were also concordant with ictal SPECT (single photon
emission computed tomography) in two cases. Two of these patients were submitted to surgical resection including
the BOLD activation area with significant improvement of
seizure control. It is important to emphasize that all patients were closely observed during the EEG-fMRI acquisition and the seizures were registered safely, with no injury
of none of these individuals.
The other three patients with clinically relevant studies
had difficult clinical and electroencephalographic definition
of the epileptic syndrome: primary generalized epilepsy
versus frontal lobe epilepsy. These two epileptic syndromes have different treatments, so the precise diagnosis is
very important. The EEG-fMRI acquisition revealed focal
BOLD activation in two of them and it reinforced the possibility of frontal lobe epilepsy. The third patient had a BOLD
activation map compatible with primary generalized epilepsy, as previously defined by the literature5.
The BOLD maps (Figure 1 and Figure 2) were based on
the time and duration of each interictal or ictal epileptiform
event seen on EEG of each patient. As an example, Figures
1 and 2 show BOLD activation and deactivation maps of
patients 2 (Figure 1) and 7 (Figure 2). These specific exams
were based on one complex partial seizure (ictal event) occurred during EEG-fMRI acquisition of each individual.
For patient 2, the activation (Figure 1-A) was observed in right parietal region (precuneus) and was
compatible with the area of MRI structural abnormality
(focal cortical dysplasia) and the region of epileptic abnormality observed on EEG. Another activation area was
observed in the right frontal region (precentral gyrus)
and was consistent with seizure propagation as seen
in this patient semiology. Sparse areas of deactivation
(Figure 1-B) were observed mainly in right and left frontal
regions.
For patient 7, the activation in Figure 2-A was observed
in right frontal region and was compatible with the area
of MRI structural abnormality (focal cortical dysplasia) and
the region of epileptic abnormality observed on EEG. This
activation was also compatible with the significant area
observed on an ictal SPECT. Deactivation in Figure 2-B
was observed on extensive areas including both frontal,
temporal and caudate nuclei regions.
Conclusion
The EEG-fMRI study is a safe and promising technique
that can be important to improve the understanding of
neurological disorders, including epilepsy. Its utility can be
improved by appropriate selection of patients, adequate
preparation of each individual, including head fixation, and
increasing acquisition time. Also, EEG-fMRI may be used
in the future as an important diagnostic tool for refractory
epileptic patients as it may add information about the localization of epileptogenic zone or definition of epileptic
syndrome.
Acknowledgment
The authors would like to acknowledge financial support
from FAPESP (Fundação de Amparo à Pesquisa do Estado
de São Paulo), grant 2009/54552-9, and CNPq (Conselho
Nacional de Desenvolvimento Científico e Tecnológico),
grant 140379/2008-8.
Figure 1. Statistical parametric map of patient 2. (A) Positive
BOLD or activation. (B) Negative BOLD or deactivation. (L) is Left
and (R) is Right. (T) is a T score of the T test.
References
1.
2.
3.
4.
Figure 2. Statistical parametric map of patient 7. (A) Positive
BOLD or activation. (B) Negative BOLD or deactivation. (L) is Left
and (R) is Right. (T) is a T score of the T test.
200
5.
Buxton RB. Introduction to functional magnetic resonance imaging:
Principles and techniques. New York: Cambridge University Press; 2000.
Haacke EM, Brown RW, Thompson MR, Venkatesan R. Magnetic resonance
imaging: Physical principles and sequence design. New York: John Wiley
& Sons; 1999.
Niedermeyer E, Silva FHL. Electroencephalography: Basic principles,
clinical applications, and related fields. Philadelphia: Lippincott Williams
& Wilkins; 2005.
Salek-Haddadi A, Diehl B, Hamandi K, Merschhemke M, Liston A,
Friston K et al. Hemodynamic correlates of epileptiform discharges:
An EEG-fMRI study of 63 patients with focal epilepsy. Brain Res.
2006;1088(1):148-66.
Hamandi K, Salek-Haddadi A, Laufs H, Liston A, Friston K, Fish DR et
al. EEG-fMRI of idiopathic and secondarily generalized epilepsies.
Neuroimage. 2006;31(4):1700-10.
Artigo Original
Development of a shielding to protect
patients against photoneutrons produced by
linacs in radiotherapy treatments
Desenvolvimento de uma blindagem para proteger
pacientes contra fotonêutrons produzidos por aceleradores
de partículas lineares em tratamentos radioterápicos
Hugo R. Silva1, Wilson F. Rebello2, Ademir X. Silva1 and Alessandro Facure3
Programa de Engenharia Nuclear/COPPE da Universidade Federal do Rio de Janeiro (UFRJ) – Rio de Janeiro (RJ), Brazil.
2
Seção de Engenharia Nuclear do Instituto Militar de Engenharia (IME) – Rio de Janeiro (RJ), Brazil.
3
Comissão Nacional de Energia Nuclear (CNEN) – Rio de Janeiro (RJ), Brazil.
1
Abstract
This work focused on radiological protection of patients submitted to radiotherapy using high energy linear accelerators, in which their
healthy tissues receive undesirable doses due to photoneutrons. For that, a shield against the produced photoneutrons was developed
using the computer code Monte Carlo N-Particle version X (MCNP-X). This shield showed to be positioned in a simple way, at the outside
part of the linear accelerator’s head, reducing the doses. This shield was named external shielding. The simulation was performed using
a computational model of the head of a Varian 2300 C/D linear accelerator, plus the external shield. In order to verify the effects of this
shielding, the values of ambient dose equivalent were calculated. These values were compared with the accelerator operating with and
without the external shielding. The results of this study indicated that the external shielding showed great efficiency in reducing the
ambient dose equivalent due to photoneutron, resulting in an average reduction above 60% for the various simulated configuration,
without increasing the ambient dose equivalent due to the photos at the plane of the patient. It was concluded that the implementation
of an external shield at the accelerator’s head increases the protection of the patients against undesirable photoneutrons doses and may
avoid new focus of cancer produced by the radiotherapy.
Keywords: simulation, MCNPX, linear accelerator, shielding against radiation.
Resumo
Este trabalho teve como objetivo a proteção radiológica de pacientes submetidos à radioterapia utilizando aceleradores lineares de alta energia, nos
quais os tecidos saudáveis recebem doses indesejáveis devido aos fotonêutrons. Para isso, uma blindagem contra os fotonêutrons produzidos foi
desenvolvida, usando o código de computador Monte Carlo N-Particle, versão X (MCNP-X). Tal blindagem mostrou que é posicionada facilmente
na parte de fora do cabeçote do acelerador linear, reduzindo as doses. Essa camada protetora foi chamada de blindagem externa. A simulação foi
realizada utilizando um modelo computacional do cabeçote de um acelerador linear Varian 2300 C/D mais a blindagem externa. Para verificar os
efeitos dessa blindagem, os valores de equivalente de dose ambiente foram calculados. Esses valores foram comparados com o acelerador operando
com e sem a blindagem externa. Os resultados deste estudo indicaram que a blindagem externa mostrou grande eficácia em reduzir o equivalente
da dose do ambiente devido ao fotonêutron, resultando em uma média de redução acima de 60% para as diversas configurações simuladas, sem
aumentar o equivalente da dose do ambiente devido às fotos no plano do paciente. Concluiu-se que a implementação de uma blindagem externa
no cabeçote do acelerador aumenta a proteção dos pacientes contra doses de fotonêutrons indesejados e pode prevenir novos focos de câncer
produzidos pela radioterapia.
Palavras-chave: simulação, MCNPX, acelerador linear, blindagem contra radiação.
Corresponding author: Hugo Roque da Silva – Programa de Engenharia Nuclear – Ilha do Fundão, Caixa Postal 68509 – CEP: 21945-970 – Rio de Janeiro
(RJ), Brasil – E-mail: [email protected]
201
Silva HR, Rebello WF, Silva AX, Facure A
Introduction
H*(10)n(mSv/Gy)
H*(10)f(mSv/Gy)
1,00E+03
H*(10) (mSv/Gy)
The production of unwanted neutrons has been a major
problem for patients undergoing radiotherapy, particularly
when the equipment operates at energies greater than
7 MV and/or IMRT mode. Also, 60% of the patients which
undergo some type of treatment against cancer, submitting
to radiotherapy1, note a disturbing and important enough
issue to be addressed. In order to minimize the undesirable doses due to neutrons produced in the sections of
the treatment, it was developed by Silva and colleagues2
a shielding against these photoneutrons that could reduce
considerably the ambient dose equivalent due to neutron
H*(10)n. With the use of collimators Jaws and multi-leaf
(MLC) to model and conform the therapeutic beam, the
ambient dose equivalent due to photons H*(10)f greatly reduces at the patient’s plan, but, for the ambient dose equivalent due to neutron H*(10)n, the calculated values remain
almost constant. Therefore, it is observed that the primary
shielding, Jaws and MLC, provide excellent electromagnetic radiation shielding, however have no satisfactory shielding for neutrons, instead, end up producing more neutrons,
especially when the therapeutic beam is higher than 7 MV.
Figure 1 shows the calculated values in the computational
model of the head of the linear accelerator Varian 2300 C/D
operating at 18 MV for H*(10)n and H*(10)f 3.
1,00E+04
1,00E+02
1,00E+01
1,00E+00
1,00E-01
1,00E-02
1,00E-03
0
50
150
100
200
250
Distance to the isocenter (cm)
Figure 1. Comparison of the calculated values of ambient dose
equivalent due to neutrons and photons, given in mSv for each dose
in Gy deposited at the isocenter. (All for the Varian 2300 C/D).
Table 1. Settings of the simulations related to the fields of apertures of collimators JAWS and MLC
Configuration
1a H*(10)n
2a H*(10)n
3a H*(10)n
Jaws
5 x 5 cm2
30 x 30 cm2
5 x 5 cm2
MLC
5 x 5 cm2
5 x 5 cm2
5 x 5 cm2
External shielding
5 x 5 cm2
5 x 5 cm2
40 x 40 cm2
Methodology
Results
For the first configuration, Figures 3 and 4 present the
values of H*(10)n in the axis Y (longitudinal direction to
202
Figure 2. Coordinates of the simulations in terms of patient.
4
Unshielded
With shield
3.5
3
2.5
mSv/Gy
The external shielding, consisting of borated polyethylene,
was developed using computational simulation. It was idealized and implemented using the Monte Carlo N-Particle
version X code (MCNP-X). For this, the external shielding
was simulated at the head of Varian 2100 C/D operating
at 18 MV, assuming a beam with 1E15 electrons of 18.8
MeV each focused on a target consisting of tungsten and
copper4; the gantry was simulated at 0°, with therapeutic
beam focused on the perpendicular plane of the patient.
The simulation was conducted in three configurations,
regarding to the opening of the fields. These settings are
presented in Table 12. It was used the MCNP F5 command
to simulate point detectors and calculate H*(10)n. The detectors were placed at coordinates A (0,0,0), B (0,20,0),
C (0,40,0), D (0,60,0), E (0,80,0), F (0,100,0), G (0,120,0),
H (20,0,0), I (40,0,0) and J (60,0,0) of this code. Figure 2
illustrates the calculated points5. To evaluate the effect of
shielding, the calculated values of H*(10)n were compared
with values obtained by Rebello and colleagues6, also by
computer simulation, with the equipment operating without external shielding.
2
1.5
1
0.5
0
0
20
40
60
80
Position (cm)
100
120
Figure 3. First configuration, H*(10)n at points along the Y axis.
Development of a shielding to protect patients against photoneutrons produced by linacs in radiotherapy treatments
the patient) and X (transverse direction to the patient),
respectively, calculated by MCNP-X with fields: JAWS
5 x 5 cm², MLC 5 x 5 cm² and external shielding 5 x 5
cm², with the head of the linear accelerator operating
with (this work) and without shielding6.
For the second configuration, Figures 5 and 6
present the values of H*(10) n in the axis Y (longitudinal
direction to the patient) and X (transverse direction
to the patient), respectively, calculated by MCNP-X
with fields: JAWS 30 x 30 cm², MLC 5 x 5 cm² and
external shielding 5 x 5 cm², with the head of the linear accelerator operating with (this work) and without
shielding 6.
For the third configuration, Figures 7 and 8 present
the values of H*(10) n in the axis Y (longitudinal direction
to the patient) and X (transverse direction to the patient), respectively, calculated by MCNP-X with fields:
JAWS 5 x 5 cm², MLC 5 x 5 cm² and external shielding
40 x 40 cm², with the head of the linear accelerator
operating with (this work) and without shielding 6.
3
4
2.75
Unshielded
With Shield
2.5
3
2.25
2.5
mSv/Gy
2
mSv/Gy
Unshielded
With shield
3.5
1.75
1.5
2
1.5
1.25
1
1
0.75
0.5
20
0.5
25
30
35 40
45
Position (cm)
50
55
0
20
60
Figure 4. First configuration, H*(10)n at points along the X axis.
35
40
45
Position (cm)
50
55
60
2.8
Unshielded
With shield
2.6
Unshielded
With shield
7
2.4
6
2.2
2
5
1.8
mSv/Gy
mSv/Gy
30
Figure 6. Second configuration, H*(10)n at points along the X axis.
8
4
3
1.6
1.4
1.2
2
1
0.8
1
0
25
0.6
0
20
40
60
80
Position (cm)
100
120
Figure 5. Second configuration, H*(10)n at points along the Y axis.
0.4
20
25
30
35
40
45
Position (cm)
50
55
60
Figure 7. Third configuration, H*(10)n at points along the Y axis.
203
Silva HR, Rebello WF, Silva AX, Facure A
4
Unshielded
With shield
3.5
3
mSv/Gy
2.5
2
1.5
1
0.5
0
0
20
40
60
80
Position (cm)
100
120
Figure 8. Third configuration, H*(10)n at points along the X axis.
it has reached an average value reduction of around
60%. The analysis of this last configuration is extremely
important, because it demonstrates that the simple installation of the external shielding, with its single opening of 40 x 40 cm 2 (this opening represents the greatest
possible opening of the primary beam and, therefore,
does not interfere with treatment), would generate a
significant reduction of H*(10)n, ensuring less patient
exposure to neutrons generated.
After these results, it can be considered, initially, that
the external shielding is able to reduce the dose absorbed
by healthy tissues of the patients, indicating a positive way
in order to stimulate a deeper study of this new system.
The external shielding can be further considered as an important safety item to be used in linear accelerators. For
future work, it will be suggest an analysis of the effect of
external shielding in internal dosimetry of organs close to
the isocenter and the evaluation of the influence of shielding effects in regions far from the plane of the patient,
particularly in the area of the maze, considering, inclusive,
multiple angles of gantry inclination.
Conclusions
At points away from the isocenter, the effect by external
shielding was very satisfactory for the three configurations. It was observed at the points evaluated the reducing of the ambient dose equivalent with an average
of 70% for the first configuration (Jaws, MLC and external shielding 5 x 5 cm²), 78.34% for the second configuration (Jaws 30 x 30 cm², MLC and external shielding
5 x 5 cm²) and 60.23% for the third configuration (Jaws,
MLC 5 x 5 cm² and external shielding 40 x 40 cm²). In
the first and second configuration, there was a greater
reduction in the values of H*(10) n, which is explained
by the fact that the external shielding was set with
the same size of fields for opening the Jaws and/or
MLC, allowing an improvement to shield the neutrons.
Instead, in the third configuration, in which the field
opening external shield was up, there was a smaller reduction of H*(10) n when compared with other settings;
however, it can be considered not less significant since
204
References
1. Facure A. Doses ocupacionais devido a nêutrons em salas de aceleradores
lineares de uso médico. [Tese de Doutorado]. Rio de Janeiro: PEN/COPPE/
Universidade Federal do Rio de Janeiro - UFRJ; 2006.
2. Silva HR. Desenvolvimento de uma blindagem contra fotonêutrons para
proteção de pacientes submetidos a radioterapia. [Dissertação de Mestrado].
Rio de Janeiro: Instituto Militar de Engenharia – IME; 2010.
3 Rebello WF, Silva AX, Facure A. Multileaf shielding design against
neutrons produced by medical linear accelerators. Radiat Prot Dosimetry.
2008;128(2):227-33.
4. Mao XS, Kase KR, Liu JC, Nelson WR, Kleck JH, Johnsen S. Neutron sources
in the Varian Clinac 2100C/2300C medical accelerator calculated by the
EGS4 code. Health Phys. 1997;72(4):524-9.
5. Teles LFK, Braz D, Lopes RT, Silva AX, Osti N. Simulação por Monte Carlo dos
feixes de 15 e 6 MV do CLINAC 2100 utilizando o código MCNP 4B. Santos:
International Nuclear Atlantic Conference – INAC; 2005.
6 Rebello WF. Blindagem para a proteção de pacientes contra nêutrons gerados
nos aceleradores lineares utilizados em radioterapia. [Tese de Doutorado]. Rio
de Janeiro: COPPE/Universidade Federal do Rio de Janeiro - UFRJ; 2008.
Artigo Original
Comparative study of two methodologies for
structural shielding design of imaging facilities
Estudo comparativo de dois métodos para o cálculo
estrutural de barreiras em instalações radiológicas
Lana T. Taniguti1 and Paulo R. Costa2
1
Bioscience Institute - Physics and Biophysics Department, State University of São Paulo – Botucatu (SP), Brazil.
2
Physics Institute - Nuclear Physics Department, University of São Paulo – São Paulo (SP), Brazil.
Abstract
The present study aimed at showing which implications can be found in structural radiation shielding design, depending on the calculation method
adopted. Two methods were analyzed: one that considers the sum of thickness contributions, and other that considers the sum of unshielded air
kerma contributions. To compare the results, a case analysis was done. A hypothetical radiographic room, which contains a table of exam and a chest
bucky, was considered. The thickness contribution method presented the highest results, reaching a maximum relative difference of 85% from the
results of the 147 National Council of Radiation Protection and Measurements, and 57% from the unshielded air kerma contributions method.
Keywords: shielding against radiation; radiation protection; air kerma; radiology.
Resumo
O presente estudo teve como objetivo mostrar quais implicações podem ser encontradas no cálculo estrutural de barreiras, dependendo do método
de cálculo utilizado. Dois métodos foram analisados: um que considera a soma das contribuições de espessura e outro que considera a soma das
contribuições de kerma no ar sem barreiras. Para comparação dos resultados, realizou-se uma análise de caso. Considerou-se uma sala radiográfica
hipotética, a qual continha uma mesa para exames e um bucky torácico. O método de contribuição de espessura apresentou os maiores resultados
alcançando uma diferença relativa máxima de 85% dos resultados do relatório 147 do National Council of Radiation Protection and Measurements,
e 57% do método das contribuições de kerma no ar sem barreiras.
Palavras-chave: barreiras contra radiação; proteção contra radiação; kerma no ar; radiologia.
Introduction
In 1925, which was the year of the first International
Congress of Radiology, radiation protection practices began to be sketched, mainly, the need of shielding radiation
sources to prevent unnecessary exposure to patients and
workers1.
Recently, structural shielding design of radiological facilities intends to protect workers and individual members
of the public, decreasing the dose to restricted levels established by national regulations2,3.
Under this perspective, the National Council of Radiation
Protection and Measurements (NCRP) published a structural shielding design methodology in report 147, which
became a reference in the area4-7.
Therefore, this study aimed at comparing two methods
for calculating the final shielding thickness, using the 147
NCRP methodology.
Methodology of the NCRP 147
Equation 1 shows Archer’s formulation8 to calculate the
shielding thickness (x).
[ [
1
. ln
x=
F(W).L(W)
G(W)
F(W)
G(W)
1+
F(W)
BL(W) +
(1)
where:
α(W), β(W) and γ(W) are fitting parameters, which are
dependent of the attenuation properties of the considered
shielding material and also of the workload spectra (W).
Transmission factor (B) consists of the ratio of the
shielded air kerma (K(x)) by the unshielded air kerma (K(x=0)),
as shown in Eq. 24,8. The shielded air kerma is related
to the planned restriction of the area concerned, which
Corresponding author: Lana Tahara Taniguti – Physics Institute of State – University of São Paulo – Rua do Matão, Travessa R, 187 – São Paulo (SP), Brazil –
E-mail: [email protected]
205
Taniguti LT, Costa PR
means radiation restriction level (P) corrected by the occupancy factor (T).
B=
K(x)
K(x = 0)
=
P 1
.
T K(x = 0)
(2)
In addition, unshielded air kerma depends on equipment output (KW1), workload distribution (W), average number of patients examined in a week (N), and source’s distance (d), as indicated in Eq. 34,8.
Thickness contribution method
Figure 1 schematically shows the thickness contribution
method for a radiographic room.
This method uses Eq. 1 to calculate each shielding
thickness contribution, using specific workload distributions. For example, Figure 1A shows the use of the X-ray
tube for abdominal images, which utilizes a workload W1,
while Figure 1B shows its use for chest examinations,
whose workload is W2.
The final shielding thickness will be the sum of all thickness contributions for the analyzed barrier.
K 1 . W. N
K(x = 0) = w 2
d
(3)
In practice, for the equipment that is used with the
X-ray tube directed to more than one position, there are
two methods for calculating the required shielding thickness, especially for secondary barriers.
The first method consists initially on calculating the shielding thickness contribution using Eq. 1 for each X-ray tube position. Afterward, these individual thickness contributions are
summed in order to find the final shielding thickness.
The second method is operated by calculating the unshielded air kerma contribution, using Eq. 3 for each X-ray
tube position. The total unshielded air kerma is obtained
summing all individual unshielded air kerma contributions.
This value is used to find the transmission factor, which is
applied to calculate the final shielding thickness.
A
X-Ray tube position 1
Air kerma contribution method
Figure 2 schematically shows the air kerma contribution
method for a radiographic room.
This method uses Eq. 3 to calculate each unshielded air
kerma contribution. The sum of these contributions will be
used to calculate the transmission factor, B, to finally find the
shielding thickness necessary to protect the desired area.
Since α, β and γ are dependent on workload spectra, the
sum of all workload distributions, of each X-ray tube positioning, needs to be considered. The resulting values are summarized at the α, β and γ parameters for all barriers, which in
NCRP 147 is mentioned as RadRoom (all barriers)4.
Case study: radiographic room
A case analysis was done in order to compare both methods. The considered facility consisted on a radiographic
A
x1(W1)
K(x=0)1
B
B
x2(W2)
x(W1 + W2) = x1(W1) + x2(W2)
K(x=0)2
C
Figure 1. Illustration of the thickness contribution method scheme.
206
K(x=0) = K(x=0)1 + K(x=0)2
C
Figure 2. Illustration of air kerma contribution method scheme.
Comparative study of two methodologies for structural shielding design of imaging facilities
room, which examines 125 patients per week, presented
as one of the examples at NCRP 147 (Example 5.34). This
hypothetical room was used to evaluate the result differences on the application of the two methods.
Figure 3 shows the X-ray tube positions, which were
considered to be used during the room routine, and
Figure 4 represents the case analyzed, showing the calculation parameters used.
A computer algorithm was developed9, and a computer
calculation software was used to execute the calculations.
Only secondary shielding thicknesses were calculated,
since primary barriers calculations for both methods result
in equal thickness values.
Results
Table 1 shows the shielding thickness results for the radiographic room presented in the Methodology, using the thickness and the air kerma contribution methods. Results of the
NCRP 1474 are also presented to compare the results.
Figures 5 and 6 indicate a comparative analysis of
the results, separated by lead and concrete thicknesses,
respectively.
Figure 3. The three X-ray tube positions considered to the case
analysis, indicating the respective use factors.
Discussion
Table 1 shows that all final shielding thickness, calculated
by the thickness contribution method, resulted in higher
values when compared to NCRP 147 results, which can
also be seen in Figures 5 and 6. This can be explained
by the fact that NCRP 147 calculation considers only one
X-ray tube position, although the radiographic room presents three ones.
Nevertheless, barriers A, E and F presented lower values compared to those presented in NCRP 147. This is
explained by the use of conservative distances by NCRP
147. In barrier A, for example, this publication used a distance of 3 m from the scatter radiation source (patient) to
the point to be protected, instead of 4.1 m presented by
the architectonical plant.
Therefore, the thickness contribution method also
showed the highest thicknesses needed for shielding the
room, when compared to unshielded air kerma contribution method. This fact demonstrates that calculating individual thicknesses, and summing all of them in the end
of the process represent a final shielding thickness higher
than calculated by others methods, such as the sum of
unshielded air kerma contributions.
For thickness contribution method applied to the studied radiographic room (Methodology), thicknesses differences reached 0.8 mm of lead and 28 mm of concrete
from NCRP 147 results. This corresponds to a relative
difference of 62 and 85%, respectively. The magnitude of
these differences is mainly due to the distances between
the secondary radiation sources from the interest point.
Figure 4. Illustration of the radiographic room used as an example for analyzing the two methods for structural shielding design.
Pink squares correspond to the barriers’ names.
207
Taniguti LT, Costa PR
Table 1. Results calculated by the thickness contribution method, and values presented by NCRP 1474.
Barrier
A
B
C
D
E
F
G
H
I
Ceiling
Wall around film box
Film Box
Floor around primary barrier
Wall around primary barrier
Wall around primary barrier
Wall
Door
Control wall
Thickness Contribution Method
48
0.7
2.1
61
0.4
1.1
0.7
0.6
0.4
Thickness (mm)
Unshielded Air Kerma Contribution Method
36
0.6
2.0
41
0.3
0.7
0.5
0.4
0.3
NCRP 147 results
44
0.5
1.3
33
0.4
1.0
0.3
Lead thickness [mm]
Conclusions
Thickness Contribution Method
Unshielded Air Kerma Contribution Method
NCRP 147
2.5
2
1.5
1
0.5
0
B
C
E
F
Barrier
G
H
I
Concrete thickness [mm]
Figure 5. Comparative plot of the lead thickness results of both
methods evaluated in this study, and the results presented at
NCRP 1474. Barriers G and H do not have NCRP 147 results because the publication do not present their shielding thicknesses.
70
60
50
40
30
20
10
0
Thickness Contribution
Method
Unshielded Air Kerma
Contribution Method
NCRP 147
A
Barrier
D
Figure 6. Comparative plot of the concrete thickness results of
both methods evaluated in this study, and results presented at
NCRP 1474.
As shorter as this distance is, more evidenced is the resulting thickness difference. This occurs because other
contributions of X-ray tube positions were not considered
by NCRP 147.
Also, from Table 1, unshielded air kerma contribution
method showed, in general, higher results when compared
to NCRP 147 values. This fact can also be explained by
the previous argument. However, these differences are
lower than those presented by the thickness contribution
method. This is due to the use of the summed workload
distribution RadRoom (all barriers) by NCRP 147.
208
The present study demonstrated the existence of differences in the final shielding thickness, depending on the
calculation method adopted. The results show that the
sum of the unshielded air kerma contribution method
presented optimized results compared to the sum of the
thickness contribution method.
Thickness differences between both methods reached
0.4 mm of lead and 20 mm of concrete for the considered
radiographic room (Methodology). These can be relevant
at final architectonical and engineering design of a radiological facility.
Acknowledgments
This study was supported by the National Council for
Scientific and Technological Development (CNPq), in
Brazil.
References
1. Archer BR. History of Shielding of Diagnostic X-ray Facilities. Health Physics.
1995;69(5):750-8.
2. Archer BR. Recent History of the Shielding of Medical X-Ray Imaging
Facilities. Health Physics. 2005;88(6):579-86.
3. Costa PR. Modelo para determinação de espessuras de barreiras protetoras
em salas para radiologia diagnóstica. [Thesis]. São Paulo: Energetic and
Nuclear Researches Institute of State University of São Paulo; 1999.
4. National Council on Radiation Protection and Measurements. Structural
Shielding Design for Medical X-Ray Imaging Facilities. Report nº 147.
Bethesda: NCRP Publications; 2004.
5. Turner JE. Front Matter, in Atoms, radiation and radiation protection.
Germany: Wiley-VCH; 2007.
6. Stabin MG. Radiation protection and dosimetry: an introduction to health
physics. New York: Springer; 2007;
7. Cember H, Johnson TE. Introduction to Health Physics. Fourth Edition.
McGraw-Hill; 2008.
8. Archer BR, Thomby JI, Bushong SC. Diagnostic X-ray shielding design
based on an empirical model of photon attenuation. Health Physics.
2008;44:507-17.
9. Taniguti LT. Cálculo Estrutural de Barreiras: desenvolvimento de um
algoritmo computacional de interface online. [dissertation]. Botucatu:
Bioscience Institute of State University of São Paulo; 2010.
Artigo Original
X-ray spectroscopy applied to the study
of the radiation transmission through
nanomaterials
Espectroscopia de raios x aplicada ao estudo da
transmissão de radiação através de nanomateriais
Roseli Künzel and Emico Okuno
Instituto de Física da Universidade de São Paulo (USP) – São Paulo (SP), Brazil.
Abstract
In this study, we compare the energy absorbed by nanostructured and microstructured materials as a function of the x-ray beam energy and
material concentration. For this purpose, we used CuO microparticles, with a mean particle size of about 56 μm, and nanoparticles with size in the
range 10 – 100 nm. These particles were separately incorporated into a polymeric resin in proportions of 5% and 30% relative to the resin mass.
Plates with about 5x5 cm² in area and uniform thickness were produced for each material. The x-ray generator was a Philips, model MG 450, with
a tungsten anode tube. Measurements were performed for beams generated at 25, 40 and 100 kV tube voltages. Data were registered with an
Amptek XR-100T-CdTe detector. Results show that nanostructured materials absorb more radiation than the microstructured ones for both material
concentrations in the resin. For example, for a 5% particle concentration and material thickness of (6.0±0.2) mm, the difference between air kerma
values is about 16% for 25 kV, 8% for 40 kV and about 2% for 100 kV.
Keywords: nanoparticles, microparticles, oxides, x-rays.
Resumo
Neste trabalho, comparamos a absorção da radiação por materiais nanoestruturados e microestruturados em função da energia do feixe de
radiação e da concentração de partículas na amostra. Com esse propósito, utilizamos micropartículas de CuO, com tamanho médio de partícula
de aproximadamente 56 µm, e nanopartículas com tamanho na faixa de 10 – 100 nm. Essas partículas foram incorporadas separadamente a
uma resina polimérica em proporções de 5% e 30% com relação à massa da resina. Para cada material foram produzidas placas com área de
aproximadamente 5 x 5 cm2 e espessura uniforme. Os feixes de raios X foram emitidos por um equipamento da marca Philips, modelo MG 450,
dotado de um anodo de tungstênio. As medições foram realizadas para feixes gerados com diferença de potencial de 25, 40 e 100 kV. Os dados
foram registrados com o detector XR-100T-CdTe da marca Amptek. Os resultados obtidos mostram que o material nanoestruturado absorve uma
maior parcela da radiação quando comparado ao microestruturado para as duas concentrações de partículas de CuO na resina estudadas neste
trabalho. Por exemplo, para o caso da concentração de 5% de partículas de CuO na resina, para uma amostra com espessura de (6,0±0,2) mm, a
diferença entre os valores de kerma no ar é aproximadamente 16% para 25 kV, 8% para 40 kV e 2% para 100 kV.
Palavras-chaves: nanopartículas, micropartículas, óxidos, raios X.
Introduction
Nanomaterials are defined as those materials whose structural elements have dimensions from 1 to 100 nm1. The
chemical and physical properties of many conventional
materials can change for the same compound in the nanostructured form1. These materials can be lightweight,
flexible and can also exhibit enhanced mechanical performance providing interesting possibilities for structural
applications1,2. Currently, nanoparticles are widely investigated for several technological applications, including diseases diagnostics, tumor treatment procedures and design of lead-free radiation protection devices3-5.
Traditionally, radiological protection of the operating
personal under clinical interventional procedures is based
on lead personal protective equipment (PPE). These lead
protection devices are relatively heavy, and the toxicity of
these materials is also an environmental issue6. In this way,
new materials are being investigated in order to produce
lead-free radiation protection devices4,6-8. The study of the
properties of radiation absorption by nanomaterials could
provide an alternative in order to replace lead garments for
use in the clinical routine4,6.
In this work, we compare the x-ray attenuation by nanostructured and microsctrutured materials. For this purpose, CuO particles, both with nanosize and microsize,
Corresponding author: Roseli Künzel – Instituto de Física da Universidade de São Paulo – Rua do Matão, Travessa R – CEP: 05508-090 – São Paulo (SP),
Brasil – E-mail: [email protected]
209
Künzel R, Okuno E
were separately incorporated in a polymeric resin in proportions of 5% and 30%. X-ray spectra were measured for
beams generated at 25, 40 and 100 kV tube voltages.
In this work, CuO nanoparticles with grain size in the range between 10 and 100 nm, produced by the Brazilian
Industry Nanum, and CuO microparticles (Vetec Química
Fina, Brazil) with 56 µm average particle size were incorporated, separately, to a polymeric resin in proportions
of 5% and 30%, relative to the resin mass. Plates with
about 5×5 cm2 in area and uniform thickness were produced for each material. Plates with thickness (6.0±0.2)
mm were produced with 5% CuO powder (nanosized
and microsized) + 100% resin. In the case of 30% CuO
powder + 100% resin, plates with (10.3±0.3) mm were
produced for both nanosized and microsized materials.
The x-ray beams were emitted by a Philips equipment MG 450 model, connected to a constant potential
generator. This tube has a 2.2 mm thick Be window with
a fixed tungsten anode tube (22o anode angle). The radiation beams were registered by using a XR-100T-CdTe
detector (Amptek, Inc., Bedford, MA) with 9 mm2 nominal active area and 1 mm nominal thickness. Output
pulses were processed by a digital pulse processor PX4
Amptek system. This equipment has a 100 µm Be window and is cooled by Peltier cells. A tungsten collimator
with 1000 µm aperture and 2 mm thickness was used
in front of the detector. The alignment between the focal
spot and detector was carried out with a laser device.
Figure 1 presents a schematic setup of the experimental arrangement used in the measurement of the transmitted radiation through the above described materials.
X-ray beams were emitted at 25, 40 and 100 kV tube
voltages. The energy calibration of the spectrometer
was performed using the measured spectra of γ-rays
and x-rays emitted by 241Am, 133Ba and 152Eu radioactive
sources.
Primary and transmitted x-ray spectra through
the materials were corrected for all possible partial interactions of photons with the detector active layer9.
Figure 2 presents the corrected primary x-ray spectra measured at 25 and 40 kV tube voltages; Figure 3, at 100 kV
tube voltage. The primary x-ray spectra were measured
using only the 4 mm Be filtration. The x-ray spectra measured at this condition shows the L and K characteristics
of X-rays emitted by the tungsten anode.
Figures 4 to 6 compare the x-ray spectra absorbed by
microstructured CuO and nanostructured CuO separately
incorporated to a polymeric resin in a proportion of 5%
relative to the resin mass. The thickness of the plates was
6.0 mm. According to these Figures, the nanostructured
material absorbs more low energy photons when compared to the microstructured material. The air kerma value
calculated at 25 kV, illustrated in Figure 4, is about 16%
higher for the microsized material when compared with the
nanosized one. Figure 5 also reveals a higher absorption
by the nanosized material for low energy photons. The
Figure 2. Primary x-ray spectra measured at 25 and 40 kV tube
voltages.
Pb Collimator
Focal
spot
W Collimator
Samples
Detector
60 cm
420 cm
Figure 1. Illustration of the experimental setup used in the spectra measurements.
210
Figure 3. Primary x-ray spectra measured at 100 kV tube
voltages.
X-ray spectroscopy applied to the study of the radiation transmission through nanomaterials
difference in the air kerma value for the x-ray beam generated at 40 kV tube voltage is about 8%. On the other hand,
the difference in the air kerma value for the x-ray beam
generated at 100 kV tube voltages, illustrated in Figure 6,
is only 2%.
Figure 7 compares the x-ray attenuation for samples
produced with the separate incorporation of 30% CuO
nanoparticles or microparticles in a polymeric resin. Both
samples have (10.3±0.3) mm thickness. Results show that
the nanostructured material absorbs about 5% more radiation than the microsctructured one.
The difference observed in the x-ray absorption by
nanostructured and microstructured materials is attributed in the literature to the particle size effect7. As the particle size decreases, the number of particles increases.
Considering the material used in this work, one microparticle is equivalent in scale to about 600 nanoparticles.
Therefore, the distribution of the nanoparticles in the resin
is different of that presented by the microparticles, resulting in a more uniform dispersion in the resin. This effect
can be responsible for the higher absorption of the low
energy x-ray photons by the nanostructured CuO material when compared to the same proportion of CuO material but with microsized particles.
Conclusion
In this work, we present results of the x-ray absorption by nanostructured and microstructured materials
Figure 4. Comparison of a 25 kV x-ray spectra transmitted
through 6.0 mm of a sample with 5% nanosized CuO+100%
resin (red line) and 6.0 mm of 5% microsized CuO+100% resin
(black line).
(black line).
(black line).
resin (red line) and 10.3 mm of 30% microsized CuO+100%
resin (black line).
211
Künzel R, Okuno E
incorporating 5% and 30% of CuO powder to a polymeric
resin. For the same material concentration, sample thickness and x-ray beam energy, the parameter that changes
from the microstructured to the nanostructured sample is
the particle size and consequently the quantity of particles. Results show that, for the materials concentrations
studied in this work, the difference between the x-ray attenuation by nanostructured and microstructured materials is only evident for low energy x-ray beams. For the
same material concentration and thickness, the radiation
absorption is higher for the nanostructured material when
compared with the microsctrutured one for low energy
x-ray photons. On the other hand, for higher energy x-ray
beams, namely those generated at 100 kV tube voltages,
the influence of the particle size on the x-ray absorption
is not evident.
Acknowledgment
The authors acknowledge the financial support of FAPESP,
through process 2010/06814-4, and CNPq.
212
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
Poole Jr C, Owens FJ. Introduction to nanotechnology. New York: WileyInterscience; 2004.
Lines MG. Nanomaterials for practical functional uses. Journal of Alloys
Compounds. 2008;449:242-5.
Cho SH, Jones BL, Krishnan S. The dosimetric feasibility of gold
nanoparticle-aided radiation therapy (GNRT) via brachytherapy using lowenergy gamma-/x-ray sources. Phys Med Biol. 2009;54(16):4889-905.
Haber FE, Froyer G. Transparent polymers embedding nanoparticles for
x-rays attenuation. Journal of the University of Chemical Technology and
Metallurgy. 2008;43(3): 283-90.
Hainfeld JF, Slatkin DN, Focella TM, Smilowitz HM. Gold nanoparticles: a
new X-ray contrast agent. Br J Radiol. 2006;79(939):248-53.
Scuderi GJ, Brusovanik GV, Campbell DR, Henry RP, Kwon B, Vaccaro AR.
Evaluation of non-lead-based protective radiological material in spinal
surgery. Spine J. 2006;6(5):577-82.
Botelho MZ, Künzel R, Okuno E, Levenhagen RS, Basegio T, Bergmann
CP. X-ray transmission through nanostructured and microstructured CuO
materials. Appl Rad Isot. 2011;69(2):527-30.
Taylor EW. Organics, polymers and nanotechnology for radiation hardening
and shielding applications. SPIE. 2007;6713(671307):1-10.
Künzel R, Herdade SB, Costa PR, Terini RA, Levenhagen RS. Ambient
dose equivalent and effective dose from scattered x-ray spectra in
mammography for Mo/Mo, Mo/Rh and W/Rh anode/filter combinations.
Phys Med Biol. 2006;51(8):2077-91.

RBFM v5n2.indb

Transcrição

Documentos relacionados

Effects of frustration in Cr8Fe molecular magnet Karlo Penc

COMPANHIA BRASILEIRA DE DISTRIBUIÇÃO AUTHORIZED

BEST OPTIONS (a) and (b) at a cost of US$ 120000.00

The effect of hydrostatic pressure on the normal and

Enjoy our exclusive weekend packages! Aproveite os

Arquivo PDF_IMC_Ingles - Indústria Metalúrgica Costinha

confederação brasileira de luta de braço e halterofilismo

RBFM v6n1.indb

Baixar este arquivo PDF

RBFM v5n1.indb

Status of Participation in the Information System on Occupational