Optimizing Detector Location in Prompt Gamma Timing for Stopping Power and Range Estimations in Proton Therapy.

Abstract

Proton therapy for tumour treatment exploits the characteristic energy-deposition pattern of protons in tissue to spare surrounding healthy structures; however, uncertainties in particle range arising from imperfect stopping -power estimation and patient positioning require safety margins. Prompt Gamma Timing (PGT) addresses this gap by measuring the arrival times of prompt gammas in external detectors and correlating them with the passage of primary protons. In this way, PGT enables in vivo range verification. In this work, a Python framework that integrates an analytical PGT model was developed to optimize detector locations for the MERLINO detection concept. The algorithm iteratively identifies the least useful detectors, proposes new positions and evaluates reconstruction metrics over many configurations, yielding layouts with reduced mean relative error from iteration to iteration. This optimization framework provides the first integration of an analytical PGT model into an automated detector-layout optimization tool within the PROSIT project, enabling systematic studies of MERLINO-like detector configurations.