Voxel phantoms and Monte Carlo methods applied to in vivo measurements for simultaneous 241Am contamination in four body regions

Abstract

A Monte Carlo program, Visual Monte Carlo (VMC) in vivo, was written to simulate photon transport through an anthropomorphic phantom and to detect radiation emitted from the phantom.
VMC in vivo uses a voxel phantom provided by Yale University and may be used to calibrate in vivo systems. This paper shows the application of VMC in vivo to the measurement of 241Am deposited simultaneously in the thoracic region, the bones, the liver and in the rest of the body. The percentages of 241Am in the four body regions were calculated using the biokinetic models established by the ICRP, for a single intake via inhalation.
The four regions of the voxel phantom were then 'contaminated' in accordance with the calculated percentages. The calibration
factor of the in vivo system was then obtained. This procedure was repeated for the radionuclide distributions obtained 5, 30, 120, 240 and 360
days after intake. VMC in vivo was also used to calculate the calibration factor of the in vivo system in which the radionuclide was assumed to be deposited only in the lung, as is normally done. The activities calculated
with the radionuclide distributed in the four body regions as a factor of time, and the activities calculated with the radionuclide
deposited in the lung only are compared.

Full text

549
Radiation Protection Dosimetry Vol. 105, No. 1–4, pp. 549–552 (2003)
Published by Nuclear Technology Publishing
 2003 Nuclear Technology Publishing
VOXEL PHANTOMS AND MONTE CARLO METHODS APPLIED
TO IN VIVO MEASUREMENTS FOR SIMULTANEOUS
241
Am
CONTAMINATION IN FOUR BODY REGIONS
J. G. Hunt, B. M. Dantas, M. C. Lourenc
¸
o and A. M. G. Azeredo
Institute of Radiation Protection and Dosimetry
Av. Salvador Allende s/n, Recreio
Rio de Janeiro, CEP 22780–160, Brazil
Abstract — A Monte Carlo program, Visual Monte Carlo (VMC) in vivo, was written to simulate photon transport through an
anthropomorphic phantom and to detect radiation emitted from the phantom. VMC in vivo uses a voxel phantom provided by
Yale University and may be used to calibrate in vivo systems. This paper shows the application of VMC in vivo to the measurement
of
241
Am deposited simultaneously in the thoracic region, the bones, the liver and in the rest of the body. The percentages of
241
Am in the four body regions were calculated using the biokinetic models established by the ICRP, for a single intake via
inhalation. The four regions of the voxel phantom were then ‘contaminated’ in accordance with the calculated percentages. The
calibration factor of the in vivo system was then obtained. This procedure was repeated for the radionuclide distributions obtained
5, 30, 120, 240 and 360 days after intake. VMC in vivo was also used to calculate the calibration factor of the in vivo system
in which the radionuclide was assumed to be deposited only in the lung, as is normally done. The activities calculated with the
radionuclide distributed in the four body regions as a factor of time, and the activities calculated with the radionuclide deposited
in the lung only are compared.
INTRODUCTION
The Monte Carlo technique and voxel phantoms have
been applied to the problem of calibration of whole
body counters (WBCs) since 1995
(1–4)
. These papers
demonstrate that mathematical phantoms can comp-
lement physical phantoms for routine calibration pur-
poses. At present, the Institute of Radiation Protection
and Dosimetry (IRD) WBC
(5)
is using the Monte Carlo
calibration technique to provide activities that will be
reported as part of the current IAEA whole body inter-
comparison programme.
This paper describes the application of this technique
to the study of a biokinetic model proposed by the ICRP
in ICRP Report 56
(6)
. The case chosen involves a single
intake through inhalation of
241
Am type ‘M’, with an
AMAD of 5 ␮m. The biokinetic model proposed in
ICRP Report 56 was used to evaluate the activity
deposited in each tissue or organ for 5, 30, 120, 240
and 360 days after intake.
The Monte Carlo code, Visual Monte Carlo (VMC)
in vivo, together with the Yale whole body voxel phan-
tom, were then used to calculate the calibration factors
for a Canberra Ge detector placed over the lung, and
over the knee for such a case. The voxel phantom
tissues were mathematically ‘contaminated’ so as to
simulate the simultaneous deposition of
241
Am in the
thoracic region, the bones, the liver and the ‘rest of the
body’. In this case, the ‘rest of the body’ signifies the
remaining soft tissues, the muscle, the adipose tissue
and the skin.
Contact author E-mail: john@ird.gov.br
MATERIALS AND METHODS
The IRD whole body counter
The IRD–CNEN whole body counter uses an array
of four Canberra HPGe detectors for in vivo measure-
ments of low energy radionuclides. The detectors are
20 mm thick, with an active diameter of 50.5 mm, active
areas of 2000 mm
2
, and a 0.6 mm carbon composite
window. The measurements are performed inside a
2.5 m ⫻ 2.5 m ⫻ 2.62 m room with 15 cm steel walls,
each covered internally with a 3 mm lead layer, fol-
lowed by a 1.5 mm cadmium layer and finally a 0.5 mm
copper layer. These layers reduce the background count
rate for energies below 200 keV.
VMC in vivo
VMC in vivo was used to simulate a tissue contami-
Table 1. Percentage distribution of
241
Am in four body
regions in relation to the number of days after intake, cal-
culated using the ICRP 56 model.
Days after Distribution of
241
Am (per cent)
intake
Lung Bone Liver Rest
5 58 9 15 18
30 54 14 23 9
120 29 24 34 13
240 15 31 38 16
360 7 37 39 17

J. G. HUNT, B. M. DANTAS, M. C. LOURENCO and A. M. G. AZEREDO
550
nation, to transport the photons through the tissues and
to simulate the detection of the radiation. As the radio-
nuclides of interest for radiation protection fall princi-
pally in the range 0.06⫺1.5 MeV, only photoelectric
and Compton interactions were considered. Runs of
5 ⫻ 10
6
histories were performed, which require around
30 minutes on a computer with an Athlon XP (1.8 GHz
clock) CPU.
Calculation of the calibration factors
The
241
Am calibration factors (in Bq cps
⫺1
) are calcu-
lated using the following equation
(3)
:
Calibration factor =
冘
N
冘
photopeak counts
⫻
1
DSF ⫻ P
,
where
冘
N is the total number of photons emitted in
the simulation,
冘
photopeak counts is the number of
photons counted in the photo peak, DSF is the ‘detector
simulation factor’ or efficiency, which in this work is
considered to be equal to 1.0, and P is the yield, which,
for
241
Am, is 0.357. A DSF equal to 1.0 implies that
there are no losses due to photon absorption in the
detector window and no losses due to the electronics.
Real measurements with the HPGe detectors show an
efficiency at 60 keV of around 0.95. If required, the cali-
bration factors given in Figure 2 and in the two tables
may be divided by 0.95 to give the true calibration fac-
tors for this detector model.
Figure 1. Visualisation of the body activity for three cases: 5 days after intake (top), 120 days after intake (middle) and 360 days
after intake (bottom) for the knee geometry.
The voxel phantom
For the simulation of in vivo measurement systems,
a voxel phantom with a format of 488 ‘slices’ each of
192 ⫻ 96 picture elements, was used. Each voxel is a
cube with 3.6 mm sides. The voxel phantom is derived
from a whole body magnetic resonance image (MRI)
scan and was obtained from the Yale University voxel
phantom library that is maintained by I. G. Zubal. The
Yale voxel phantom was modified at the IRD so as to
maintain only the tissues relevant for whole body coun-
ting. For this work, the phantom is divided into the fol-
lowing tissue types: lung, total bone and liver. The other
tissues and muscles are grouped together.
The ICRP biokinetic model
The case chosen for study involves a single intake
through inhalation, of
241
Am type ‘M’. The systemic
biokinetic model proposed in ICRP Report 56 was used
to evaluate the percentage of activity deposited in each
of the above mentioned tissue types for 5, 30, 120, 240
and 360 days after intake (see Table 1). The human res-
piratory tract model described in ICRP Publication 66
(7)
was also used.
RESULTS AND DISCUSSION
The graphical representation of the knee counting
geometry for 5, 120 and 360 days after intake is shown
in Figure 1. The relative activities of the lung, liver,
bone and ‘the rest’ can be seen. The calibration factors
for the lung and knee geometries for
241
Am deposition
in the four body regions were calculated for the five

THE MONTE CARLO TECHNIQUE AND VOXEL PHANTOMS
551
time intervals, and the results are shown in Figure 2,
which also shows the calibration factors obtained for a
100% deposition of
241
Am in the lung and a 100% depo-
sition of
241
Am in the total bone.
In Table 2, the cps that would be obtained through a
real measurement in the case of a single intake of 1 kBq
of
241
Am are shown. These values are obtained by divid-
10
2
10
1
10
0
0 50 100 150 200 250 300 350 400
knee geometry
lung geometry
6.1 kBq cps
-1
for
241
Am in lung-lung geometry
1.6 kBq cps
-1
for
241
Am in lung-lung geometry
Days after intake
kBq cps
-1
Figure 2. Comparison of calibration factors (kBq cps
⫺1
) for
241
Am in relation to time after intake. The black circles refer to
the lung geometry, and the white circles to the knee geometry.
Table 2. Calculated cps for the lung and knee geometries as a function of time after an intake of 1 kBq of
241
Am type
‘M’.
Days after Total Lung geometry Knee geometry
intake activity
(Bq)
Calibration Cps Lung Lung activity Calibration Cps Bone Bone activity
factor activity calculated factor activity calculated
(kBq cps
⫺1
) (Bq) using (kBq cps
⫺1
) (Bq) using
1.6 kBq cps
⫺1
6.1 kBq cps
⫺1
(Bq) (Bq)
5 91.8 2.54 0.036 53 58 26.4 0.0035 8.3 21
30 71.5 2.73 0.026 39 42 32.4 0.0022 10 13
120 60.3 4.39 0.014 18 22 20.0 0.0030 14 18
240 57.2 6.95 0.0082 8.6 13 13.8 0.0041 18 25
360 55.6 9.58 0.0058 3.4 9.3 12.7 0.0044 21 27
REFERENCES
1. Mallett, M. W., Hickman, D. P., Knuchen, D. A. and Poston J. W. Development of a method for calibrating in vivo measure-
ment systems using magnetic resonance imaging and Monte Carlo computations. Health Phys. 68(6), 773–785 (1995).
2. Ishikawa, I. and Uchiyama, M. Estimation of the counting efficiencies for individual subjects in
137
Cs whole body counting,
using voxel phantoms. Radiat. Prot. Dosim. 71(3), 195–200 (1997).
3. Hunt, J. G., Mala
´
tova
´
, I. and Folta
´
nova
´
,S.Calculation and measurement of calibration factors for bone-surface seeking low
ing the total body activity by the respective calibration
factor. Table 2 also includes the activities for the lung
and knee geometries obtained from the ICRP model,
and the corresponding activities obtained by using the
calibration factors obtained for the lung phantom
(1.6 kBq cps
⫺1
) and for the bone phantom
(6.1 kBq cps
⫺1
).
A comparison of the activities for the lung and knees,
and the corresponding activities obtained by using the
calibration factors obtained for the lung phantom and
for the bone phantom, show that the question of activity
present in other organs introduces an error in the meas-
ured activity. For the lung geometry, the result is a 10%
overestimation of the activity for the first 2 months after
intake. For the knee geometry, the result is an approxi-
mately 30% overestimation of the activity 30 days
after intake.
CONCLUSIONS AND FUTURE WORK
If the measured cps values for a real case of
241
Am
incorporation over the time interval of 360 days are
similar to those shown in Table 2, then it can be con-
sidered that the ICRP Report 56 model has been suc-
cessfully checked.
The results confirm that the time after intake is very
important for the interpretation of the values measured
with a WBC. For example, at 5 days, the overestimation
of bone activity by using a bone phantom for calibration
is by a factor of 2.5, falling to 30% during the first
month. For the lung, the overestimation increases from
10% after 5 days to a factor of 2.7 after 360 days.

J. G. HUNT, B. M. DANTAS, M. C. LOURENCO and A. M. G. AZEREDO
552
energy gamma emitters and determination of
241
Am activity in a real case of internal contamination. Radiat. Prot. Dosim.
82(3), 215-218 (1999).
4. Hunt, J. G., Bertelli, L., Dantas, B. M. and Lucena, E. Calibration of in vivo measurement systems and evaluation of lung
measurement uncertainties using a voxel phantom. Radiat. Prot. Dosim. 76(3), 179–184 (1998).
5. Oliveira, C. A. N., Lourenc
¸
o, M. C., Dantas, B. M., Lucena, E. A., Bertelli, L. and Laurer, G. R. The IRD/CNEN whole body
counting facility: background and calibration results. Radiat. Prot. Dosim. 29(3) 203–208 (1989).
6. International Commission on Radiological Protection. Age-dependant doses to members of the public from intake of radio-
nuclides: Part 1. ICRP Publication 56. Ann. ICRP 20(2) (Oxford: Pergamon Press) (1989).
7. International Commission on Radiological Protection. Human respiratory tract model for radiation protection. ICRP Publi-
cation 66. Ann. ICRP 24(1–3) (Oxford: Pergamon Press) (1994).