基于图谱库的自动轮廓勾画软件在宫颈癌自适应放疗中的应用

目的 评估基于图谱库的自动轮廓勾画软件(ABAS)在宫颈癌自适应放疗中的应用.方法 选取2014年1月至3月收治的21例已行第1程调强放疗的宫颈癌患者,将其已勾画器官的CT图像及第2程未勾画器官的定位CT图像传输至ABAS软件系统,以第1程图像为模板图像,第2程定位图像作为目标图像.在第2程定位图像上手工勾画出靶区和危及器官,将ABAS软件自动勾画图像和医师手工勾画的靶区、危及器官图像传输至飞利浦Pinnacle计划系统,对两组结果进行评估.比较相似性指数(DSC)和勾画体积.结果 ABAS自动勾画与医师手工勾画的DSC平均值均大于0.7,其中靶区的DSC最大为肿瘤临床靶区(CTV, 0.89±0.08),最低为肿瘤区(GTV, 0.72±0.16).对于危及器官,DSC最高的为右股骨头(0.88±0.05),最低为直肠(0.73±0.07).左右髂骨自动勾画体积较手工勾画小,且差异具有统计学意义(t=3.37、2.74, P<0.05),其他轮廓体积差异无统计学意义.结论 宫颈癌放疗过程中,基于图谱库的ABAS勾画软件,节省了临床器官勾画工作时间,加强自动勾画后的轮廓修改,并建立患者模板数据库,得到满意度更高的重合结果,也为开展自适应放疗提供了强有力的支持.

Systematic evaluation of atlas-based autosegmentation (ABAS) software for adaptive radiation therapy in cervical cancer

Objective To evaluate the application of atlas-based autosegmentation (ABAS) software for adaptive radiation therapy in cervical cancer.Methods Twenty-one patients were enrolled from January to March in 2014. In addition to the planning CT (pCT) images, first re-planning CT (rCT) image set was acquired during the radiotherapy course. The targets and regions of interested (ROI) were outlined on the pCT and rCT by experienced physicians. All contours were transmitted to Pinnacle treatment planning system in DICOM format, which contained both automatic and manual contouring. The dice similarity coefficient (DSC) and delineation volume were used to evaluate the quality of contours which obtained automatically from the software and manual contouring.Results The average values of DSC for automatic and manual contours were both larger than 0.7. The maximum (Max) DSC value was 0.89±0.08 in clinical target volume (CTV), and the minimum (Min) was 0.72±0.16 in gross tumor volume (GTV) for the target. For the ROIs, the Max DSC value was 0.88±0.05 in right femoral head, and the Min was 0.73±0.07 in rectum. ABAS contouring volume was less than manual contouring on iliac bone, and there was statistical difference for the contour volume (t=3.37, 2.74, P<0.05). There was no statistically significant difference for other contours volume.Conclusions ABAS software might shorten the time of clinical organ delineation work in cervical cancer radiotherapy, and have shown the advantage in adaptive radiation therapy technology.