Deep residual fully connected neural network classification of Compton camera based prompt gamma imaging for proton radiotherapy

Proton beam radiotherapy is a method of cancer treatment that uses proton beams to irradiate cancerous tissue, while minimizing doses healthy tissue. In order guarantee the prescribed radiation dose delivered tumor and ensure tissue spared, many researchers have suggested verifying delivery through use real-time imaging using methods which can image prompt gamma rays are emitted along beam’s path patient such as Compton cameras (CC). However, because limitations CC, their images noisy unusable for delivery. We provide detailed description deep residual fully connected neural network capable classifying improving measured CC data with an increase in fraction usable by up 72% allows improved reconstruction across full range clinical conditions.