- Email: [email protected]
410 Verification of step-and-shoot breast compensator IMRT fields with an amorphous silicon EPID
Posters
$180
409 B e a m - c h a r a c t e r i s t i c s during beam s t a r t - u p : m i n i m u m n u m b e r of MU for I M R T . J. Steenhufisen Catharina-hospital, Radiotherapy, Eindhoven, The Netherlands I n t r o d u c t i o n : When performing IMRT multiple segments with small numbers of MU are delivered. To quantify the accuracy of the beam during these small segments, output, symmetry and flatness during beam start-up were investigated on 5 Elekta SLi linacs. On these machines steering of flatness and symmetry starts after a userdefinable delay of .5 to 1 s after beam-on, which equals 5 to 10 ME at 600 ME/rain. Methods: To quantify the linearity of the ionisation monitor 3 series of ionisation chamber readings for 10x10 cm z fields were taken for 6MV and 10MV photon beams over the range 1-200 MU. The output factor is defined as Gy/MU; the relative output as the output factor relative to the output factor for a 10 MU segment. By using the Wellh6fer Beam Imaging System (BIS) dosimetric images of the beam were acquired during the first 10 monitor units for 5 gantry angles (-180 °, -90 °, 0 °, 90 ° , 180°). Every 120 ms an image of 820x820 pixels was taken of a 40x40 cm ~ field at SSD = 73 cm with 3 cm water-equivalent build-up. Spatial resolution of the BIS is 0.37 mm. Results: The deviation of the relative output is smaller than 0.5% for segments larger than 2 MU at all linacs. For segments of 1 MU we found deviations of 1.5%. The symmetry and flatness of the beams were very large during the first 240 ms after 'beam on', but decreased within 2 MU to twice the value at 10 MU. While the output of the BIS is not equivalent to dose, we calculated all symmetry and flatness measurements relative to 10 MU. The integrated symmetry (defined as the symmetry of the integrated dose distribution) and integrated flatness decreased within 4 MU to twice the value at 10 MU. No influence of gantry angle or treatment machine on these results were found. Conclusion: During IMRT segments of 4 MU can be safely used for all gantry angles. Errors in dose are smaller than 1% and symmetry and flatness are less than twice that of 10 MU segments. 410 Verification o f s t e p - a n d - s h o o t b r e a s t c o m p e n s a t o r I M R T fields w i t h an a m o r p h o u s silicon E P I D C. Hood, P. Greer Newcastle Mater Misericordiae Hospital, Radiation Oncology/Medical Physics, Newcastle, Australia Background and purpose: Verification of IMRT treatments has, in the past, been a relatively complex and labour intensive process. By creating relative planar dose maps using an amorphous silicon electronic portal imaging device (EPID) and comparing these to fluence maps produced by the 3D treatment planning system, 2D dose distributions can be quickly and easily assessed. Method: Pinnacle Portal Dose Maps (PPDMs) at a depth of dm~× were produced for 10 forward-planned breast IMRT patients. The accuracy of the PPDMs were verified for relative dose distributions using a diode array system for beam profiling. An EPID image was produced for each beam and corrected for energy response off-axis and pixel sensitivity variation. This was then converted to a relative dose profile for comparison with the PPDM. Results: Figure 1 indicates results of the PPDM compared to the diode array (profiler) measurement and the EPID analysis. The differences are minimal and the assumption can be made that the PPDM is a relatively accurate indication of output dose from the linear accelerator. There also appears good agreement between the EPID profile and the PPDM. The reported profiles are only a 1 dimensional analysis, however
a full 2-dimensional program is currently being developed. Conclusion: This study has shown that PPDMs produce a reasonable estimate of the distribution from the linear accelerator and that EPlDs can closely match this relative distribution. The use of EPID as a dosimetry verification tool for IMRT treatments is both time-efficient and user-friendly. I t eliminates the need for multiple films and ion chamber measurements (once absolute doses are determined from the EPID) and provides both a visual and quantitative dosimetric analysis. Care must be taken to correct the EPID for small variations in photon energy away from the central axis as the EPID sensitivity is dependent on this.
~4
.
.
.
.
g
I: .._J '.'~
.~
,,.~
,~
o
:-~
Figure 1. Comparison of the Pinnacle Planar Dose Map (PPDM), the EPID generated profile and the diode array measurement (Profiler). 411 A c c e p t a n c e criteria for p r e - t r e a t m e n t d o s i m e t r y of head and neck I M R T : statistical approach based on g a m m a distribution histograms. C. Fiorino~ E. De Martin, B. Longobardi, S. Broggi, G. Cattaneo, R. Calandrino IRCCS San Raffaele, Medical Physics, Milan, Italy Background and purpose: To provide a statistical analysis of gamma distribution histograms for planned and irradiated dose distribution comparison. To consequently assess acceptance criteria and dosimetric QA practice for 10 intensity levels Step and Shoot (SS10) IMRT with a Varian treatment planning and delivery system in case of head and neck cancer treatments. M a t e r i a l s and Methods: In this study pre-irradiation QA evaluation of the treatment plan was performed for 12 clinical patients, both for Single Fields (XV Omat films and 0.125cc PTW ion chamber in Bohsung phantom) and complete 3D treatment (EDR2 films and 0.6cc Farmer ion chamber in ESTRO CarPet phantom). Planned and irradiated dose distribution comparison concerned both absolute point dose measurements and dose profile verifications, but focused especially on gamma analysis (WellhoferScanditronix OmniPro 1.2 Software). Gamma distribution histograms were statistically evaluated in terms of mean values, gamma <1 values, gamma <1.5 values and gamma >2 values, both for agreement criteria of 3%/3mm and 4%/3mm. The results were used to establish which of the two sets of criteria should be applied in the gamma comparison process when head and neck tumours are concerned. Results and conclusions: 3%/3mm agreement criteria for the gamma test appear overly restrictive as far as head and neck tumours are concerned, due to complicated inverse
$180
409 B e a m - c h a r a c t e r i s t i c s during beam s t a r t - u p : m i n i m u m n u m b e r of MU for I M R T . J. Steenhufisen Catharina-hospital, Radiotherapy, Eindhoven, The Netherlands I n t r o d u c t i o n : When performing IMRT multiple segments with small numbers of MU are delivered. To quantify the accuracy of the beam during these small segments, output, symmetry and flatness during beam start-up were investigated on 5 Elekta SLi linacs. On these machines steering of flatness and symmetry starts after a userdefinable delay of .5 to 1 s after beam-on, which equals 5 to 10 ME at 600 ME/rain. Methods: To quantify the linearity of the ionisation monitor 3 series of ionisation chamber readings for 10x10 cm z fields were taken for 6MV and 10MV photon beams over the range 1-200 MU. The output factor is defined as Gy/MU; the relative output as the output factor relative to the output factor for a 10 MU segment. By using the Wellh6fer Beam Imaging System (BIS) dosimetric images of the beam were acquired during the first 10 monitor units for 5 gantry angles (-180 °, -90 °, 0 °, 90 ° , 180°). Every 120 ms an image of 820x820 pixels was taken of a 40x40 cm ~ field at SSD = 73 cm with 3 cm water-equivalent build-up. Spatial resolution of the BIS is 0.37 mm. Results: The deviation of the relative output is smaller than 0.5% for segments larger than 2 MU at all linacs. For segments of 1 MU we found deviations of 1.5%. The symmetry and flatness of the beams were very large during the first 240 ms after 'beam on', but decreased within 2 MU to twice the value at 10 MU. While the output of the BIS is not equivalent to dose, we calculated all symmetry and flatness measurements relative to 10 MU. The integrated symmetry (defined as the symmetry of the integrated dose distribution) and integrated flatness decreased within 4 MU to twice the value at 10 MU. No influence of gantry angle or treatment machine on these results were found. Conclusion: During IMRT segments of 4 MU can be safely used for all gantry angles. Errors in dose are smaller than 1% and symmetry and flatness are less than twice that of 10 MU segments. 410 Verification o f s t e p - a n d - s h o o t b r e a s t c o m p e n s a t o r I M R T fields w i t h an a m o r p h o u s silicon E P I D C. Hood, P. Greer Newcastle Mater Misericordiae Hospital, Radiation Oncology/Medical Physics, Newcastle, Australia Background and purpose: Verification of IMRT treatments has, in the past, been a relatively complex and labour intensive process. By creating relative planar dose maps using an amorphous silicon electronic portal imaging device (EPID) and comparing these to fluence maps produced by the 3D treatment planning system, 2D dose distributions can be quickly and easily assessed. Method: Pinnacle Portal Dose Maps (PPDMs) at a depth of dm~× were produced for 10 forward-planned breast IMRT patients. The accuracy of the PPDMs were verified for relative dose distributions using a diode array system for beam profiling. An EPID image was produced for each beam and corrected for energy response off-axis and pixel sensitivity variation. This was then converted to a relative dose profile for comparison with the PPDM. Results: Figure 1 indicates results of the PPDM compared to the diode array (profiler) measurement and the EPID analysis. The differences are minimal and the assumption can be made that the PPDM is a relatively accurate indication of output dose from the linear accelerator. There also appears good agreement between the EPID profile and the PPDM. The reported profiles are only a 1 dimensional analysis, however
a full 2-dimensional program is currently being developed. Conclusion: This study has shown that PPDMs produce a reasonable estimate of the distribution from the linear accelerator and that EPlDs can closely match this relative distribution. The use of EPID as a dosimetry verification tool for IMRT treatments is both time-efficient and user-friendly. I t eliminates the need for multiple films and ion chamber measurements (once absolute doses are determined from the EPID) and provides both a visual and quantitative dosimetric analysis. Care must be taken to correct the EPID for small variations in photon energy away from the central axis as the EPID sensitivity is dependent on this.
~4
.
.
.
.
g
I: .._J '.'~
.~
,,.~
,~
o
:-~
Figure 1. Comparison of the Pinnacle Planar Dose Map (PPDM), the EPID generated profile and the diode array measurement (Profiler). 411 A c c e p t a n c e criteria for p r e - t r e a t m e n t d o s i m e t r y of head and neck I M R T : statistical approach based on g a m m a distribution histograms. C. Fiorino~ E. De Martin, B. Longobardi, S. Broggi, G. Cattaneo, R. Calandrino IRCCS San Raffaele, Medical Physics, Milan, Italy Background and purpose: To provide a statistical analysis of gamma distribution histograms for planned and irradiated dose distribution comparison. To consequently assess acceptance criteria and dosimetric QA practice for 10 intensity levels Step and Shoot (SS10) IMRT with a Varian treatment planning and delivery system in case of head and neck cancer treatments. M a t e r i a l s and Methods: In this study pre-irradiation QA evaluation of the treatment plan was performed for 12 clinical patients, both for Single Fields (XV Omat films and 0.125cc PTW ion chamber in Bohsung phantom) and complete 3D treatment (EDR2 films and 0.6cc Farmer ion chamber in ESTRO CarPet phantom). Planned and irradiated dose distribution comparison concerned both absolute point dose measurements and dose profile verifications, but focused especially on gamma analysis (WellhoferScanditronix OmniPro 1.2 Software). Gamma distribution histograms were statistically evaluated in terms of mean values, gamma <1 values, gamma <1.5 values and gamma >2 values, both for agreement criteria of 3%/3mm and 4%/3mm. The results were used to establish which of the two sets of criteria should be applied in the gamma comparison process when head and neck tumours are concerned. Results and conclusions: 3%/3mm agreement criteria for the gamma test appear overly restrictive as far as head and neck tumours are concerned, due to complicated inverse