A feasibility study on a machine-learning-based quality assurance tool for spot-scanning proton therapy using delivery log files and treatment plans

Abstract

Background and purpose: Although pencil beam scanning provides superior dose conformity, delivery uncertainties remain in beam monitoring and steering systems. This study evaluates a machine learning model that predicts delivered spot positions from plan parameters as a complementary quality assurance support tool. Materials and methods: The dataset consisted of a single routine quality assurance treatment plan and 64 corresponding delivery log files from a Hitachi proton scanning beam system (32 files each from gantry 1 and 2), collected over three months across various proton energies. Statistical analyses, including Levene&apos;s test, Welch&apos;s analysis of variance, and Games-Howell post hoc tests, assessed the effects of beam energy, treatment day, and room on spot position errors. Machine learning regression models were trained using 48 delivery log files to predict delivered spot positions from treatment planning data. Model performance was assessed using mean squared errors, R-2 score, and Euclidean distance. Results: Machine learning models demonstrated feasibility of spot position prediction with high accuracy (R-2 = 0.999; mean squared errors of 0.021 mm(2) and 0.003 mm(2)), with Euclidean errors <0.13 mm. Predicted dose distributions closely matched planned distributions, with an average absolute mean dose difference of 2.34 & times; 10(-5) Gy, and a maximum absolute voxel-wise dose difference of 1.796 Gy, defined as the largest absolute difference between predicted and planned dose values at corresponding voxels. Conclusions: This feasibility study demonstrates that machine-learning-based prediction of delivered spot positions can achieve sub-millimeter accuracy, potentially enhancing the precision and reliability of quality assurance processes in proton therapy.