LCC-Demons形变配准算法在自适应放疗中的应用研究  

Research on LCC-Demons Deformation Registration Algorithm in Adaptive Radiotherapy

作　　者：李汉 王玉[1] 王明泉[1] 张鹏程[1] 阙禄松 LI Han;WANG Yu;WANG Mingquan;ZHANG Pengcheng;QUE Lusong(Shanxi Provincial Key Laboratory for Biomedical Imaging and Big Data, North University of China, Taiyuan 030051, China)

机构地区：[1]中北大学生物医学成像与影像大数据山西省重点实验室,山西太原030051

出　　处：《测试技术学报》2020年第3期215-219,共5页Journal of Test and Measurement Technology

基　　金：国家重大科学仪器设备开发专项资助项目(2013YQ240803);山西省重点研发计划资助项目(201803D121069);国家自然基金资助项目(61171177);山西省科技攻关资助项目(20140321010-02);山西省应用基础研究项目面上自然基金资助项目(201801D121162)。

摘　　要：癌症患者在接受放疗过程中,体重以及肿瘤体积的变化等因素很可能导致靶区、危及器官OARs的解剖结构发生变化.传统的形变配准算法对于这种复杂大形变鲁棒性较弱,难以完成配准.针对上述情况,本文采用LCC-Demons形变配准算法对病人放疗计划中的CT图像进行配准,利用得到的形变场完成医生勾画信息映射.通过比较配准前后靶区与危及器官的DSC相似性系数(重叠度)与Hausdoff距离来评判配准算法精度,从而对配准质量作出评价.实验结果表明,LCC-Demons算法具有很好的棒性,对复杂的头颈部配准有很好的效果,对于自适应放疗具有重要意义.During radiotherapy,changes in body weight and tumor volume of cancer patients are likely to lead to changes in the anatomical structure of target area and OARs..The traditional deformation registration algorithm is weaker in robustness for this complex large deformation and it is difficult to complete the registration.In view of the above situation,this paper used the LCC-Demons deformation registration algorithm to register the CT images in the patients'radiotherapy plan,and used the deformation field to complete the doctor's sketch information mapping.The accuracy of the registration algorithm is evaluated by comparing the DSC similarity coefficient(overlap)and Hausdoff distance between the target area and the crisis organ before and after registration,so as to evaluate the registration quality.The experimental results show that the LCC-Demons algorithm has good robustness and has a good effect on complex head and neck registration,which is of great significance for adaptive radiotherapy.

关 键 词：医学图像 刚性配准 LCC-Demons 形变配准 自适应放疗 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]