A Monte Carlo study on the feasibility of real-time in vivo source tracking during ultrasound based HDR prostate brachytherapy treatments

Purpose: This study aims to assess the accuracy of in-vivo source tracking during real-time trans-rectal ultrasound (TRUS) based high dose rate (HDR) prostate brachytherapy (pBT) through Monte Carlo simulations of multiple HDR pBT treatments with a two-dimensional (2D) diode array, the Magic Plate 900 (MP900), embedded below the patient in a carbon-fibre couch. Method: Monte Carlo simulations of source positions representing three separate real-time TRUS based HDR pBT treatments were performed using the Geant4 toolkit. For each source position, an Ir-192 source was simulated inside a voxelized patient geometry. Dose deposited from each source position to the MP900 diodes was used to perform source tracking, and the MP900 calculated position compared to known source positions from the treatment plan. Thresholding techniques were implemented to improve source tracking accuracy with the TRUS probe present. Results: The average three-dimensional source position reconstruction discrepancy was 11.9 +/- 2.4 mm and 1.5 +/- 0.3 mm with and without the TRUS probe, respectively. Thresholding techniques improved the source position reconstruction discrepancy in the presence of the TRUS probe to 1.8 +/- 0.4 mm. Conclusion: Inclusion of the TRUS probe inside the patient negatively affects source tracking accuracy when using the MP900 diode array for HDR pBT verification. Modification of the source tracking algorithm using thresholding techniques improves source tracking in the presence of the TRUS probe, achieving similar accuracy as when the TRUS probe is not present. This study demonstrates that accurate in-vivo source tracking during real-time TRUS based HDR pBT is feasible using the Magic Plate system.