中国癌症杂志 ›› 2019, Vol. 29 ›› Issue (4): 294-299.doi: 10.19401/j.cnki.1007-3639.2019.04.009

• 论著 • 上一篇    下一篇

直肠癌放射治疗中快速自动计划设计的可行性研究

张海英,陈元华,朱 骥,李桂超,胡伟刚,彭佳元   

  1. 复旦大学附属肿瘤医院放射治疗中心,复旦大学上海医学院肿瘤学系,上海 200032
  • 出版日期:2019-04-30 发布日期:2019-05-17
  • 通信作者: 彭佳元 E-mail: ppjerry@163.com
  • 基金资助:
    国家自然科学基金(11805038)。

A feasibility study of quick auto-planning for rectal cancer radiotherapy

ZHANG Haiying, CHEN Yuanhua, ZHU Ji, LI Guichao, HU Weigang, PENG Jiayuan   

  1. Department of Radiation Oncology, Fudan University Shanghai Cancer Center; Department of Oncology, Shanghai Medical College, Fudan University, Shanghai 200032, China
  • Published:2019-04-30 Online:2019-05-17
  • Contact: PENG Jiayuan E-mail: ppjerry@163.com

摘要: 背景与目的:传统调强适形放疗(intensity-modulated radiotherapy,IMRT)计划设计是一个反复优化且耗时的过程。该研究通过使用射野形变技术,研究在直肠癌的放射治疗中进行快速自动计划设计(quick auto-planning,QAP)的可行性(计划设计时间小于10 s),以提高直肠癌放射治疗过程的效率。方法:QAP方法通过分析源计划(模板计划)的计划靶区体积(planning target volume,PTV)轮廓和新患者待设计计划(即目标计划)的PTV轮廓,得到轮廓边界形变函数。通过此边界形变函数将模板计划中每个射野形状变成目标计划射野的形状。然后,重新计算目标计划的射野权重。用9例直肠癌患者(1例患者为模板计划,8例患者分别采用QAP设计目标计划和人工设计目标计划)对QAP方法进行可行性验证,评估QAP计划设计的时间和质量(PTV剂量指标和正常组织剂量指标)。结果:8例QAP设计的目标计划的用时均小于6 s。模板计划、QAP计划(x±s)和人工计划(x±s)的PTV平均剂量(Dmean)分别为51.70、(52.90±0.40)和(52.00±0.30)Gy;适形指数(conformity index,CI)分别为1.12、1.13±0.05和1.05±0.06;膀胱Dmean分别为35.00、(35.80±5.00)和(36.70±5.10)Gy,左股骨头Dmean分别为27.40、(29.40±3.70)和(24.70±2.70)Gy,右股骨头Dmean分别为29.20、(28.40±2.20)和(24.20±2.80)Gy。结论:QAP技术可以应用于直肠癌IMRT计划设计中。使用本技术设计的目标计划质量接近于模板计划和人工计划质量,计划效率相比传统的逆向优化技术有显著性提高,其可行性得到验证。

关键词: 调强放疗, 快速自动计划设计, 射野形变技术, 直肠癌, 放射治疗

Abstract: Background and purpose: The treatment planning of conventional intensity-modulated radiotherapy (IMRT) is a time-consuming optimization process. The aim of this work was to study the feasibility of quick auto-planning (QAP) technique using a new beam morphing method to improve the treatment planning efficiency (planning time <10 s) in rectal cancer radiotherapy. Methods: The QAP method was used to analyze the boundary deformation from the planning target volume (PTV) of source plan (template plan) to the PTV of the treatment plan to be designed for a new patient (target plan) in order to derive the source-totarget shape mapping function. The derived shape mapping function was used to morph the source beam aperture to the target beam aperture over all segments in each beam direction. Secondly, the beam weights of the target plan were re-calculated. The approach was tested on nine rectal cancer patients (one patient with template plan, eight target patients for QAP and manual planning) for proof-of-principle purpose. The planning time and the plan quality (dose endpoints of PTV and organs at risk) of the QAP plans were evaluated. Results: The QAP planning time of eight patients was less than 6 s each. For the template plan, QAP plans (x±s) and manual plans (x±s), the PTV Dmean was 51.70, (52.90±0.40) and (52.00±0.30) Gy. The conformity index (CI) was 1.12, 1.13±0.05 and 1.05±0.06. The bladder Dmean was 35.00, (35.80±5.00) and (36.70±5.10) Gy. The left femoral head Dmean was 27.40, (29.40±3.70) and (24.70±2.70) Gy. The right femoral head Dmean was 29.20, (28.40±2.20) and (24.20±2.80) Gy. Conclusion: The QAP with a new developed beam morphing method can be applied to the IMRT planning for rectal cancer radiotherapy. The plan quality by QAP is close to the plan quality of the template plan and manual plans. Compared with the conventional optimization method, the planning efficiency of the QAP is significantly improved, and its feasibility is verified.

Key words: Intensity-modulated radiotherapy, Quick auto-planning, Beam morphing method, Rectal cancer, Radiotherapy