Computational intelligent image analysis for assisting radiation oncologists' decision making in radiation treatment planning

Abstract

This chapter describes the computational image analysis for assisting radiation oncologists' decision making in radiation treatment planning for high precision radiation therapy. The radiation therapy consists of five steps, i.e., diagnosis, treatment planning, patient setup, treatment, and follow-up, in which computational intelligent image analysis and pattern recognition methods play important roles in improving the accuracy of radiation therapy and assisting radiation oncologists' or medical physicists' decision making. In particular, the treatment planning step is substantially important and indispensable, because the subsequent steps must be performed according to the treatment plan. This chapter introduces a number of studies on computational intelligent image analysis used for the computer-Aided decision making in radiation treatment planning. Moreover, the authors also explore computer-Aided treatment planning methods including automated beam arrangement based on similar cases, computerized contouring of lung tumor regions using a support vector machine (SVM) classifier, and a computerized method for determination of robust beam directions against patient setup errors in particle therapy.