Measuring variability in radiotherapy Organs at Risk (OAR) contours

One key step of radiotherapy treatment planning is contouring the OARs. Historically, clinicians have manually delineated these contours. Studies have shown significant differences in contours among various clinicians (inter-observer variability). Recently, AI solutions have offered a promising alternative to the traditional manual delineation approach. These allow for significant time savings while producing more homogenous delineations.

This project aims to model the variability of OAR contours in a population. This information will enable us to make predictions about whether a new contour lies within the clinically accepted variability or needs to be corrected. We will also be able to compare the variability between different populations of patients.