Classification of Compton Camera Based Prompt Gamma Imaging for Proton Radiotherapy by Random Forests

Abstract

Proton beam radiotherapy is a method of cancer treatment that uses proton beams to irradiate cancerous tissue, while simultaneously sparing healthy tissue. One promising method of real-time imaging during treatment is the use of a Compton camera, which can image prompt gamma rays that are emitted along the beam’s path through the patient. However, because of limitations in the Compton camera’s ability to detect prompt gammas, the reconstructed images are often noisy and unusable for verifying proton treatment delivery. Machine learning ensemble methods like random forests are able to automatically learn patterns that exist in numerical data, making them a promising method to analyze Compton camera data for the purpose of reducing noise in the reconstructed images. We conduct a hyperparameter search to find an optimal random forest model. We then present the results of the best performing random forest model, which demonstrate that this ensemble method is less effective than competing machine learning techniques for this application.