Abstract

Abstract: The goal of our research is to automatically score radiosurgery plans. Radiosurgery is a technique for treating brain lesions with high dose radiation. When selecting a plan, the clinician must tradeoff between the complexity of executing the plan and the conformity in delivering the radiation. This decision depends on the treatment preferences of the clinician. Our research studies whether it is possible to learn to score radiosurgery plans in the same way the clinician does. We use regression to learn a preference function that maps plan properties such as complexity and conformity to the level of satisfaction that the plan holds for the clinician. The preference function is then used to predict the level of satisfaction that the clinician will have for unseen plans. The preference function makes it possible to either automatically select a radiosurgery plan for the clinician or make his task less time-consuming by sorting all plans according to their scores. We performed experiments with four different classes of preference functions and were able to identify the best plan with high reliability. We conclude that using regression to learn the treatment preferences of clinicians is a promising approach to automate or support the selection of radiosurgery plans in a clinician-specific manner.

Download the paper in pdf.

Download the paper in gzipped postscript (large download time).

Many publishers do not want authors to make their papers available electronically after the papers have been published. Please use the electronic versions provided here only if hardcopies are not yet available. If you have comments on any of these papers, please send me an email! Also, please send me your papers if we have common interests.

This page was automatically created by a bibliography maintenance system that was developed as part of an undergraduate research project, advised by Sven Koenig.