SU‐E‐T‐504: Incident Fluence Reconstruction Based on Monte Carlo Finite‐Size Pencil Beam Model for Dose Guided Radiation Therapy

Purpose: In dose guided radiation therapy, incident fluence is one of the important parameters for both influence verification and dose reconstruction. The objective of this investigation was to develop a method to obtain the incident fluence for dose reconstruction in dose guided radiation therapy (DGRT). Methods: An incident fluence reconstruction model based on Monte Carlo Finite‐size pencil beam (MCFSPB) model and “Matthew Effect” Conjugate‐gradient (CG) algorithm was developed, where the MCFSPB model was developed by FDS Team (www.fds.org.cn) and the “Matthew Effect” was proposed to advance the accuracy of CG lgorithm. Then, by measuring dose with amorphous silicon flat panel (a‐Si flat panel), the incident fluence was unfolded. An artifical head phantom was used to test sample. By using the incident fluence to dose calculation with MCFSPB, the 2D and 3D dose was also reconstructed and compared with the measurement dose and the planing dose respectively. The Passing Rate of Dose Difference (PRDD) in 3% was used to learn the accuracy. Results: Two representative cases (6MV and 10MV) with both five beams were used for beam‐by‐beam verification, and the corresponding incident fluence was obtained. By comparing the measurement dose with the calculation dose on measurement plane, the results showed that the PRDD (3%) in field was 100%. And by comparing 3D plannning dose with 3D reconstructed dose, the results indicated that the PRDD (3%) in field was 95%∼98%. Conclusions: The incident fluence reconstruction model based on MCFSPB model and “Matthew Effect” Conjugate‐gradient algorithm was developed. The tested results indicated its accuracy. This study is expected to be applied to 3D dose reconstruction and 3D dosimetric verification in DGRT. Supported by the National Natural Science Foundation under grant No.30900386 and the National Science Foundation of Anhui Province (Nos.090413095 and 111040606Q55).