Collision-Free Path Planning and Delivery Sequence Optimization in Noncoplanar Radiation Therapy
针对非共面放射治疗中的碰撞风险和耗时问题，提出结合改进概率路线图法和遗传算法的方案，实现无碰撞路径规划和递送序列优化，仿真实验验证了有效性。
Radiation therapy is among the top three cancer treatments in current medical services. The novel noncoplanar radiation therapy which claimed the best characteristics in almost all dosimetric properties encountered the challenges of the potential collision and the long time delivering. In this paper, we proposed a brand new scheme which uses a combined method of the collision avoidance path planning based on an improved probability roadmap method (PRM) and the delivery sequence optimization based on a modified genetic algorithm (GA) to solve the problems in noncoplanar radiation therapy. A uniform sampling strategy, an improved connection strategy, and an efficient local planner are introduced to optimize the roadmap result and accelerate the roadmap construction. The GA is improved by the elitist selection, the local search strategy, and the similar substitution strategy to achieve a better performance both in convergence rate and optimal solution. Experiments are carried out on the simulation platform with typical therapy system models. The results show that our proposed methods work well with the radiation therapy system in a compact working area. Collision is avoided and time consumption is reduced. We believe that our proposed algorithms could solve the problems in current radiation therapy and promote their clinic applications.