Dosimetric properties of amorphous silicon electronic portal imaging device for verification of dynamic intensity modulated radiation therapy

Abstract

The purpose of this study is to investigate the basic dosimetric characteristics of flat panel amorphous silicon (a-Si) detectors (aS500, Varian medical systems) for the possibility of using a Si portal imager for absolute dosimetric verification of the delivery of dynamic intensity modulated radiation treatment (IMRT) fields. The measurements were performed with 6 MV X-ray beams from Varian Clinac 23EX. The electronic poral imaging device (EPID) was performed in AM maintenance program and portal dosimetry mode. The studies include field size dependence, dose rate and dose response, effect of dead time in frames acquisition, dose rate fluctuations, memory effect, relative and absolute dosimetry. The portal dose image was tested by comparing the EPID profiles with the ion chamber both for open and wedge fields. The portal dose image calculated by EPID portal dosimetry was compared with the EPID measurement for clinical IMRT fields. The results showed field size dependence of EPID, which was more sensitive than ion-chamber for larger field size but less than ion chamber in smaller field size. The EPID was linear with dose rate and integral dose. The effect of dead time in frame acquisition due to transfer to the CPU was found to start at 40 UM of field size studied which were 4 x 4, 10 x 10 and 15 x 15 cm[superscript2]. The dead time resulted in dynamic field caused error that increased with leaf speed and about 17.62% for a 1 cm leaf gap moving at 1 cm/s. The response of dose rate fluctuation at 50 and 500 MU were not different. The ghost effect was negligible by increasing the period more than 15 s between exposures. The comparison of profiles from EPID and ion chamber measurement showed the agreement for larger field size but slightly higher dose of EPID in the center of the beam for 4 x 4 cm[superscript2]. For wedge field, both of EPID and ion-chamber profiles showed the agreement in the center part but slightly shift in the penumbra region. The largerwedge angle gave more shift than small wedge angle. The pre-treatment verification for two fields of IMRT showed the agreement of absolute dose distribution between EPID calculation and EPID measurement within 3% difference in dose and 3 mm. Difference in distance. The profile in the direction of MLC movement showed good correlation between calculation and measurement. The research has demonstrated that an understanding of the relationship between pixel value reading and dose of fluence is a prerequisite for portal dosimetry. The EPID is suitable to be used for the IMRT pre-treatment verification