Genetic algorithm-based dynamic intraoperative treatment planning for prostate brachytherapy

Abstract

This paper reports on the use of a genetic algorithm (GA) for dynamic intraoperative treatment planning (DITP) with stepwise replanning for the prostate implant. The study applied a simple needle and seed reconstruction model to simulate the seed displacement. The simulation of DITP consists of an initial plan and four re-optimized plans. The results show that, on average, the total number of seeds required by each re-plan increased. The average maximum urethral dose and maximum rectal dose showed a slight increase during the first two replanning steps, but increased substantially during the latter two replanning steps. Several issues are discussed regarding the limitation and future improvement of GA-based DITP. © 2005 by International Federation for Information Processing.