IMRT逆向计划中多目标优化算法及目标函数研究

项目名称： IMRT逆向计划中多目标优化算法及目标函数研究

项目编号： No.60872112

项目类型： 面上项目

立项/批准年度： 2009

项目学科： 金属学与金属工艺

项目作者： 李国丽

作者单位： 浙江工业大学

项目金额： 30万元

中文摘要： 三维适形调强放射治疗（IMRT）需要对成千上万的射束数量、方向、权重等进行设计，逆向计划过程必不可少，其涉及的关键问题有对射束各参数进行最优化设计的优化算法、目标函数的设定等。要达到最大限度地杀灭肿瘤细胞并使周围正常组织和重要器官得到有效保护的目的，放射治疗的逆向计划过程就必须对靶区、危及器官、正常组织分别进行目标设置，所以，逆向计划的优化问题属于多目标优化。本课题针对IMRT逆向计划的多目标特性，结合基于剂量分布和基于剂量- - 体积约束的物理目标函数的各自优缺点，发展符合临床实际的IMRT混合目标函数，并研究搜索和决策交替进行的、收敛速度快、鲁棒性好的多目标优化算法，为解决IMRT放射治疗逆向计划的关键技术问题提供研究基础。

中文关键词： 调强适形放射治疗；逆向计划；目标函数；多目标优化算法

英文摘要： The inverse planning process of intensity modulated radiation therapy (IMRT) is necessary because the number, direction and weight of thousands beamlet must be designed perfectly. The key problems of inverse planning are optimization algorithm for the design of beam parameters, objective function, etc. For the purposes of delivering a sufficient high dose in the cancerous tissue and protecting the surrounding normal tissue (NT) and the organs at risk (OAR), the inverse planning process of radiation therapy must set objective for planning target volume(PTV), NT and OAR respectively. Therefore, the optimization question of inverse planning belongs to the category of multi-optimization. This project arms at the multi-objective characteristics of inverse planning for IMRT, combines the advantages and shortcomings of average dose distribution based and the dose-volume constraints based objective functions, develops bybrid objective functions for IMRT which can meet the needs of clinical, and researches multi-optimization algorithm which can significantly enhance the speed of calculation and improve the ability of global optimization by the strategy of decision making and optimization alternatively. The purpose of the project is to offer the study foundation for solving the key technology problems of inverse planning on IMRT.

英文关键词： Intensity Modulated Radiation Therapy; Inverse Planning; Objective Function; Multi-objective Optimization Algorithm