A DNA genetic algorithm for beam angle selection in radiotherapy planning

There are few of evolutionary algorithms (EAs) considering the influence of bit positions when mutation operation is implemented. A DNA genetic algorithm (DNA-GA) with a novel bit mutation strategy is presented in this paper. ldquoHot spotsrdquo and ldquocold spotsrdquo are set in DNA individuals with different mutation probabilities, and three structure mutation operations are designed to replace those bad individuals with better ones. DNA-GA uses DNA encoding method which borrows the intelligence from the biological DNA encoding mechanism to encode the solutions. The presented DNA-GA is applied to automatically select the beam angles for intensity-modulated radiotherapy (IMRT) planning, which uses a triplet code to represent a beam angle. The preliminary results show that DNA-GA is feasible and effective for the beam angle optimization (BAO) problem in IMRT planning and faster to obtain the optimal plan than GA.