Abstract
Knowledge-based planning (KBP) is an automated approach to radiation therapytreatment planning that involves predicting desirable treatment plans beforethey are then corrected to deliverable ones. We propose a generativeadversarial network (GAN) approach for predicting desirable 3D dosedistributions that eschews the previous paradigms of site-specific featureengineering and predicting low-dimensional representations of the plan.Experiments on a dataset of oropharyngeal cancer patients show that ourapproach significantly outperforms previous methods on several clinicalsatisfaction criteria and similarity metrics.
Quick Read (beta)
loading the full paper ...