Optimization in radiosurgery treatment planning

Abstract

Gamma knife radiation treatment is an alternative to treating brain tumors with surgery. Currently, treatment planning is a manual and time-consuming task that involves an iterative process of shot selection and placement. In this study Genetic Algorithm (GA) is introduced to automatically select shot location and shot size for radiosurgery treatment planning, which will make planning process simpler, less time-consuming and more effective. First, a distance transformation/coding method is used to generate the skeleton of the target. Then, along the skeleton, the GA-based shot placement algorithm is applied to find the best location to place a shot. By continuously iterating the algorithm, the number, size and the location of all the shots are generated. The proposed GA-based model has been tested using numerous simulated targets with different shapes and sizes. Results demonstrate that the GA-based shot location and size determination can speed up the process of shot placement.