A Research On IMRT Radiation Treatment Plan Retrieval Technology Based On OVH Descriptor

As an accurate radiation therapy technology,Intensity-Modulated Radiation Therapy(IMRT)has become one of the widely used technique of radiotherapy.Plan design is an important part of IMRT technology,and its goal is to find the best balance between high dose coverage of the target and low dose distribution of normal tissue.At present,the IMRT plan is still a trial and error process,and its quality depends much on the physical experience,which leads to time-consuming and difficult to develop high-quality IMRT plans because of the numerous parameters involved in the process of planning.Therefore,how to ensure the quality of the plan and improve the efficiency of the IMRT planning has become a research focus in radiotherapy.The most recent research indicates that prior knowledge based radiation treatment planning can be very useful in ensuring the quality of the IMRT plan and achieving rapid planning.The IMRT radiation treatment plan retrieval technology based on OVH descriptor has been deeply studied in this paper.Firstly,the step of the IMRT plan design in clinical application and the IMRT plan retrieval technology was introduced,Secondly,the description of geometric relationship was studied deeply,and an effective calculation method for an OVH descriptor based on morphology was proposed,which can be used to calculate OVH descriptor accurately;Then,an improved method of similar case retrieval based on OVHp and OVHo descriptor was proposed,which can be used to retrieve similar cases with higher shape similarity;Finally,a clustering method of the radiation treatment knowledge base based on K-means method was proposed,which can make the cases in the same category w ith high similarity,and improve the retrieval efficiency by finding out the category of the new case firstly.To verify the reliability and practicability of the proposed method,the retrieval performance and quality control performance of the IMRT plan retrieval method based on OVH descriptor has been analyzed in detail by using clinical cases of nasopharyngeal carcinoma and prostate cancer.The experimental results show that: 1)OVH descriptors can describe the geometrical relationship between the target and OARs well,and retrieve the high similar ity cases;2)Using the planning parameter of the similar cases to re-optimize the new cases,the dose of their OARs are reduced,but without changing the dose coverage rate of target,and the re-optimized plan can be compared with the plan which is designed by repeated optimization of the Senior Physicist.In conclusion,the quality control of IMRT plan can be achieved;3)Because the planning is directly guided by the similar case,the process of optimizing the planning parameters is shortened,thus it lay a foundation for the efficiency of IMRT planning.