ESTRO 37 Abstract book

S952

ESTRO 37

3 Illawarra Health and Medical Research Institute, IHMRI, Wollongong, Australia

Purpose or Objective The aim of this work is to demonstrate the

feasibility of performing in-vivo dosimetry and image co- registration with a dual detector. A 2D array detector based on silicon diodes is placed above an EPID. Requiring no adjustment to the EPID, near tissue equivalent dosimetry could be performed while the EPID can still function as an imaging device, aiding in patient positioning error detection. Material and Methods In order to validate the dual detector approach, a lung phantom embedded with a 2cm diameter spherical water equivalent target was positioned on the patient couch at isocentre. A LINAC was used to deliver 250 MU for a field of 5x5 cm 2 with 6 MV photon beam. The EPID was set up at SSD of 150 cm directly beneath the LINAC couch. The 2D array detector (MP) composed of 11×11 small 1.5x1.5 mm 2 epitaxial diodes with 10 mm pitch, embedded in a 0.6 mm thick kapton carrier is placed above the EPID. The array detector had 5 mm thick water equivalent material underneath and build-up of 5 mm or 15 mm. The effect of build-up on SNR, defined as the ratio of centre to out of field response was evaluated. The TPS Pinnacle TM version 9.0 software (Philips Radiation Oncology Systems, Fitchburg, WI) was used to calculate dose profiles at the array position. To investigate the detector sensitivity of a variation in the dose distribution due to a misalignment of a target, the plan has been delivered to the phantom when the target was at isocentre and laterally shifted by 7 mm. The dose profiles are co-registered to the EPID images taken in the same setup. Results When the target is at isocentre, dose profile comparison between the measured dose and the TPS agrees within 4.5% for inter-umbra region (Fig. 1a). The difference graph (Fig.1b) shows the difference between profile measured at isocentre and the profile measure with the phantom shifted laterally. Fig. 2 shows EPID images taken of the target when at isocentre and laterally shifted. It confirms that the displacement of the dose measured in the shifted configuration is from the misalignment of the target in the treatment field. Similar results were observed with the thinner build-up. The 15 mm build-up resulted in a 65.4% increase in SNR of the array detector, although it results in a 12.5% reduction of the contrast noise ratio (CNR) of the EPID image compared to the 5 mm.

Conclusion The dual detector is capable of measuring transmission dose which matches the TPS calculations, without requiring a complicated beam model or multiple correction factors necessary for EPID dosimetry. The system correctly measured the reduction in an absolute dose deposition due to the presence of a higher density target in the phantom as predicted by TPS. The array detector was sensitive to a dose error of less than 8%, due to the target shift, confirmed by EPID image analysis. The system is therefore feasible for performing simultaneous in-vivo dosimetry and imaging of the patient without modifying the EPID functionality.