差异进化
粒子群优化
数学优化
元启发式
多群优化
计算机科学
光圈(计算机存储器)
最优化问题
算法
差速器(机械装置)
数学
物理
声学
热力学
作者
Ali Fallahi,Mehdi Mahnam,Seyed Taghi Akhavan Niaki
标识
DOI:10.1016/j.asoc.2022.109798
摘要
Intensity-modulated radiation therapy is a well-known technique for treating cancer patients worldwide. A treatment plan in this technique requires decision-making for three main problems: selection of beam angles, intensity map calculation, and leaf sequencing. Previous works investigated these problems sequentially. We present a new integrated framework for simultaneous decision-making of directions, intensities, and aperture shape, called direct angle and aperture optimization, and develop a mixed-integer nonlinear mathematical model for the problem. Due to the nonlinearity and the dimension of the problem, three efficient metaheuristics based on differential evolution (DE) called classic differential evolution (cDE), discrete differential evolution (dDE), and adaptive hybrid discrete differential evolution-particle swarm optimization (ahdDE-PSO) algorithms are designed to solve the problem. Parameters calibration is performed using the Taguchi design of experiments. The performance of the algorithms is evaluated by solving the problem for ten real cases of liver cancer disease from the TROTS data set. The performed ablation study and statistical analysis of computational results demonstrate that ahdDE-PSO is capable of finding high-quality treatment plans.
科研通智能强力驱动
Strongly Powered by AbleSci AI