Machine learning positioning algorithms for long semi-monolithic scintillator PET detectors

作者：Samuel Mungai Kinyanjui, Zhonghua Kuang, Zheng Liu, Ning Ren, Yongfeng Yang · 发表于：Physics in Medicine and Biology · 年份：2025 · DOI：10.1088/1361-6560/addbbe · 被引用次数：3 · 研究领域：Radiation Detection and Scintillator Technologies、Medical Imaging Techniques and Applications、Particle Detector Development and Performance

Abstract Objective. In this work, machine learning positioning algorithms are developed to improve the spatial resolutions of the semi-monolithic scintillator detectors in both monolithic ( y ) and depth of interaction ( z ) directions. Approach. Two long semi-monolithic scintillator detectors consisting of 12 lutetium yttrium oxyorthosilicate (LYSO) slabs of 0.96 × 56 × 10 mm 3 and 14 LYSO slabs of 0.81 × 56 × 10 mm 3 were manufactured. The scintillator arrays were read out by a 4 × 16 silicon photomultiplier array. 27 × 5 ( y, z ) positions of each detector were irradiated via a collimated 22 Na pencil beam. Extreme gradient boosting (XGBoost) machine learning model was used to predict the interaction positions for y and z . The genetic algorithm (GA) or particle swarm optimization (PSO) algorithm was used to optimize hyperparameters for the XGBoost model. The results of the machine learning positioning algorithms were compared to analytical positioning methods. Main results. The GA and PSO algorithms provided similar results. Compared to the analytical methods, the machine learning positioning methods improved both y and z spatial resolutions especially at both ends of the detectors. The average y spatial resolutions using the machine learning positioning methods were 0.92 ± 0.41 mm and 0.94 ± 0.44 mm as compared to those obtained with the squared center of gravity method of 1.38 ± 0.23 mm and 1.39 ± 0.25 mm for the two detectors, respectively. The average z spatial resolu...