Docstoc

Deliverable

Document Sample
Deliverable Powered By Docstoc
					Deliverable D3.3
 Del.   Deliverable name                                    WP   Nature   Disseminat     Delivery
 no.                                                                       ion level      date
D3.3* Report on the performance of clinical treatment        3      R         PU          24M
      planning systems for estimating comparative
      normal tissue dose distributions from current and
      emerging treatment modalities.

Task 3.5: Comparison and evaluation of treatment planning systems in low dose
region.
Commonly used treatment planning systems will be evaluated for accuracy of calculating
normal tissue dose distributions outside the treatment volume down to doses of 0.5Gy.
The evaluation will consist of intercomparisons between different systems and
algorithms, Monte Carlo simulation with PHITS (CUT), and MCNP (KULeuven) codes,
and dose measurements in anthropomorphic phantoms. The accuracy will be tested on
standard plans for head and neck, lung, prostate, CNS tumours and lymphoma in
children. The plans will all be based on the same CT data and defined PTV and OAR.
The task is divided into the following modality-determined sub-tasks:
      3.5(a): Photons (3-D conformal, IMRT, tomotherapy, stereotactic) (KULeuven)
      3.5(b): Protons (passive system, scanning system) (PSI)
      3.5(c): Carbon ions (where possible) (GSI)
Task leader: KULeuven
   •   Procedure for modelling of the beam

The experiment with irradiation of waa ter tank by photons was performed by the group,
lead by Prof. Frank Van den Heuvel at Katholieke Universiteit Leuven at the Leuven
Hospital, using an Varian Linac accelerator for radiation therapy using 18 MeV photons.

To get as good agreement as possible between PHITS simulations and the experimental
results, the geometric model of the accelerator and the parameters used during the
simulations of the experiment have to be as precise and accurate as possible. The head
geometry of the Varian Linac Accelerator, used for the PHITS simulations is shown in
Figure 1.




                 Figure 1. The PHITS geometry for Varian Linac Accelerator 18 MeV.



A water tank with water surface 90 cm from the isocenter was irradiated by the high
energetic photons. The experimental set up is illustrated in Figure 2. The total dose
distribution, induced by the photons, was measured by ion chambers, moved around in
the water tank. The dose was measured in lateral direction 10 cm below the water surface,
shown as yellow circles in Figure 2, and 20 cm below the water surface, shown as red
circles. The total relative dose deposition in the depth direction in the water was also
measured by ion chambers, which is schematically illustrated in Figure 2 as blue circles.



                                   Varian accelerator

                                             18 MeV electrons
            90




                                      Water tank
                                                            Depth dose distribution
                          10
                     20




                                                            Lateral dose distribution
            50




                               Figure 2. The experimental set up.
   •   Results of fitting PHITS to the clinical beams and comparison to the
       experimental data

PHITS simulations with different beam parameters were made to investigate the
sensitivity in the agreement between the simulation results and the experimental data.

To study the dependence of the simulation results on the electron beam energy, hitting
the target, electron energies of 17 MeV, 18 MeV and 19 MeV were used. The dose
distribution profile for the pencil electron beam, without Gaussian energy distribution,
with different energies for the photon beam of field size of 10x10 cm2 from Varian Linac
accelerator in water is shown in fig. 3, together with a comparison with the experimental
data.

Influences of the Gaussian distribution and the incident electrons beam size were also
considered. The simulations and comparison of the pencil electron beam of 18 MeV
energy without Gaussian distribution; pencil electron beam of 18 MeV energy with
Gaussian distribution with FWHM of 0.5 MeV/n; electron beam of the cylindrical shape
of 18 MeV energy with Gaussian distribution with FWHM of 0.5 MeV/n and
experimental data are shown in Figure 4.

To get accurate information about the geometry, material and composition of the Varian
Linac accelerator is essential to achieve good agreement between the simulations and the
measurements. There are still, however, some uncertainties present concerning the
materials and the geometry in our simulations. One of the uncertainties is the composition
of the flattening filter. For the 18 MeV accelerator, the flattening filter has a complex
geometry and composition, as it is shown in Figure 5. The filter is a combination of two
materials: iron (Fe) and tantalum (Ta). The material composition is still somewhat unsure.
The possibility that an alloy was used instead of a pure material can not totally be ruled
out. Therefore a simulation with a cylindrical electron beam of 18 MeV with Gaussian
distribution, FWHM of 0.5 MeV for flattening filter of different densities for the tantalum
part were made, which are shown in Figure 6.

To study the influence of the flattening filter on the simulation results a simulation with a
copper (Cu) flattening filter, which is usually used for the 6 MeV incident electron beam,
was made. The shape and the material properties are shown in Figure 7. The simulation
results and the comparison to an experimental data are shown in Figure 8.
                                                                                                                                              120
                                                                                                                                                                                         Depth dose


                                                                                                                                              100




                                                                                                           Relative dose, [MeV/source]
                                                Experiment                                                                                         80

                                                Pencil beam 17 MeV, no Gaussian
                                                                                                                                                   60

                                                Pencil beam 18 MeV, no Gaussian
                                                                                                                                                   40
                                                Pencil beam 19 MeV, no Gaussian
                                                                                                                                                   20



                                                                                                                                                    0
                                                                                                                                                        90               100                  110                 120              130
                                                                                                                                                                                     Depth position, [cm]



                                 140
                                                         Lateral dose 10 cm                                                        140
                                                                                                                                                                                   Lateral dose 20 cm


                                 120                                                                                               120
   Relative dose, [MeV/source]




                                                                                                            Relative dose, [MeV/source]
                                 100                                                                                               100


                                  80                                                                                                          80


                                  60                                                                                                          60


                                  40                                                                                                          40


                                  20                                                                                                          20


                                   0                                                                                                               0
                                       0    2       4        6       8       10        12        14                                                     0        2       4           6        8             10    12          14   16
                                                         Lateral position, [cm]                                                                                                    Lateral position, [cm]




                                       Figure 3. The depth dose and lateral dose profile of 17, 18, 19 MeV
                                       incident electrons (field size: 10x10cm2) in water simulated by PHITS and
                                       compared to the experimental data.


                                                                                                                                                   120
                                                                                                                                                                                          Depth dose


                                                                                                                                                   100
                                                                                                                     Relative dose, [MeV/source]




                                           Experiment                                                                                              80

                                           Pencil beam 18 MeV, no Gaussian
                                                                                                                                                   60

                                           Pencil beam 18 MeV,
                                           Gaussian distr., FWHM = 0.5 MeV/n                                                                       40


                                           Cylinder beam 18 MeV, r = 1.5 cm
                                                                                                                                                   20
                                           Gaussian distr., FWHM = 0.5 MeV/n
                                                                                                                                                       0
                                                                                                                                                            90               100                  110                   120              130
                                                                                                                                                                                      Depth position, [cm]

                                 140
                                                         Lateral dose 10 cm                                                                        140
                                                                                                                                                                                     Lateral dose 20 cm


                                 120                                                                                                               120
                                                                                                                     Relative dose, [MeV/source]
Relative dose, [MeV/source]




                                 100                                                                                                               100


                                 80                                                                                                                 80


                                 60                                                                                                                 60


                                 40                                                                                                                 40


                                 20                                                                                                                 20


                                  0                                                                                                                    0
                                       0    2        4           6       8        10        12        14                                                    0        2         4          6             8        10           12   14
                                                          Lateral position, [cm]                                                                                                     Lateral position, [cm]




                                       Figure 4. The depth dose and lateral dose profile of 18 MeV incident
                                       electron beam with/without Gaussian distribution of pencil and cylindrical
                                       shape (field size: 10x10cm2) in water simulated by PHITS and compared to
                                       the experimental data.
                                                            Figure 5. Flattening filter for the Varian Linac
                                                               Accelerator for 18 MeV electron beam.




                                                                                                                   120
                                                                                                                                            Depth dose


                                                                                                                   100
                                        Experiment
                                                                                     Relative dose, [MeV/source]




                                        Cylinder 18 MeV, r = 1.5 cm,                                               80

                                        Gaussian distr., FWHM = 0.5 MeV/n,
                                                           3
                                        rho(Ta)=16.69 g/cm (teor. density)                                         60



                                        Cylinder 18 MeV, r = 1.5 cm,                                               40

                                        Gaussian distr., FWHM =0.5 MeV/n,
                                        rho(Ta)=13.69 g/cm3                                                        20



                                                                                                                    0
                                                                                                                         90       100            110             120             130
                                                                                                                                        Depth position, [cm]

                              140
                                                  Lateral dose 10 cm                                               140
                                                                                                                                        Lateral dose 20 cm


                              120                                                                                  120
                                                                                     Relative dose, [MeV/source]
Relative dose, [MeV/source]




                              100                                                                                  100


                              80                                                                                    80


                              60                                                                                    60


                              40                                                                                    40


                              20                                                                                    20


                               0                                                                                     0
                                    0    2    4       6       8       10   12   14                                       0    2     4       6          8    10         12   14
                                                  Lateral position, [cm]                                                                Lateral position, [cm]




                                    Figure 6. The depth dose and lateral dose profile of 18 MeV incident
                                    electron beam of cylindrical shape of Gaussian distribution (field size:
                                    10x10cm2) in water, with flattening filter of two different tantalum densities,
                                    simulated by PHITS and compared to the experimental data.
                                                                 Figure 7. Flattening filter for the Varian Linac
                                                                     Accelerator for 6 MeV electron beam.




                                                                                                                          120
                                                                                                                                                      Depth dose
                                                                                                                                                                                             100


                                                                                                                          100




                                                                                                                                                                                                   Relative dose, [MeV/source]
                                                                                            Relative dose, [MeV/source]




                                              Experiment                                                                                                                                     80


                                                                                                                          80
                                              Pencil beam 18 MeV,                                                                                                                            60
                                              Gaussian distr., FWHM = 0.5 MeV/n
                                                                                                                          60
                                                                                                                                                                                             40
                                              Pencil beam 18 MeV,
                                              Gaussian distr., FWHM = 0.5 MeV/n,                                          40
                                                                                                                                                                                             20
                                              Flat.Filter of Cu, rho(Cu) = 8.96 g/cm3
                                              with shape as for 6 MeV                                                     20
                                                                                                                                                                                             0


                                                                                                                           0
                                                                                                                                90       100                110             120        130
                                                                                                                                                   Depth position, [cm]


                                    140
                                                         Lateral dose 10 cm                                               200
                                                                                                                                                   Lateral dose 20 cm


                                    120
      Relative dose, [MeV/source]




                                                                                            Relative dose, [MeV/source]




                                                                                                                          150
                                    100


                                    80
                                                                                                                          100
                                    60


                                    40
                                                                                                                          50

                                    20


                                     0                                                                                      0
                                          0   2      4       6       8       10   12   14                                       0    2         4        6         8       10      12   14
                                                         Lateral position, [cm]                                                                    Lateral position, [cm]




                                          Figure 5. The depth dose and lateral dose profile of 18 MeV incident electron
                                          beam of cylindrical shape of Gaussian distribution (field size: 10x10cm2) in
                                          water, with flattening filter of two different tantalum densities, simulated by
                                          PHITS and compared to the experimental data.




•   Analysis of results
The PHITS code is an accurate and helpful tool for simulation of the particle transport
and interactions. However, the simulations are very time (CPU) consuming, which
influenced the statistics in the presented diagrams. As mentioned before, the experiment
parameters have to be reproduced in PHITS as precise as possible to make a simulation
trustworthy. During the experiment small cylindrical ionization chambers of the diameter
of 0.45 mm and 3 cm long were used, which implies a great difficulty in getting enough
statistics from the simulation. Therefore the number of the simulation events has to be
increased to improve the agreement with the experimental results and get clearer
graphical representation.

The depth dose distribution, as can be seen in all the simulation diagrams, is shifted
regarding the experimental data, which might indicate that the flattening filter is
flattening the beam too much in the center. The shape of the lateral dose distribution has
also poor agreement with the experimental results, especially close to the edge of the
10x10 cm2 field.

The influence of the energy and the energy distribution for the incident electron beam is
not very significant, as can be seen in Figure 2 and Figure 3. However, the simulation
results show that the shape and size of the beam have a great impact on the shape of dose
distribution, as can be seen in the Figure 4. Therefore more work needs to be done in this
area.

The shape and material composition of the flattening filter have a great influence on the
simulation results though. A clear indication of that was illustrated by Figure 6 and,
especially, Figure 8. The best agreement with the experimental results, see Figure 8, was
achieved when implementing copper flattening filter into the PHITS simulations, which
is usually used for 6 MeV electron beam. This emphasizes the importance of the detailed
knowledge about the characteristics of the flattening filter, used during the experiment.

It is essential to get exact information about all the materials in the linac and to a analyze
further the sensitivity of the results to changes in geometries and materials in order to get
better agreement with the experiment and to be able to proceed with the studies of the
neutron dose further away from the field.

				
DOCUMENT INFO