期刊文献+

一种基于模型分割的自动勾画算法在临床中的应用 被引量:1

Clinical application of an automatic delineation algorithm with model based segmentation(MBS)
下载PDF
导出
摘要 目的:对基于模型分割(model based segmentation,MBS)算法的自动勾画模式在临床中的应用进行评价和分析。方法:选取本院头颈部、胸腹部、盆腔进行放疗的患者共计90例,分别由计划系统自带的MBS和医生勾画左右肺、左右肾、脊髓、肝脏、膀胱、左右股骨等器官轮廓。以医生所勾画轮廓为参考,使用戴斯系数、敏感度指数等指标来评估MBS勾画器官的优劣。结果:由MBS勾画器官的戴斯系数、体积差(%)、敏感度指数、包容性指数分别为左肺(0.930±0.013、-9.5±2.1、0.886±0.021、0.979±0.010)、右肺(0.946±0.006、-8.4±1.6、0.906±0.013、0.989±0.002)、脊髓(0.877±0.041、-17.7±2.3、0.799±0.046、0.971±0.037)、肝脏(0.886±0.055、23.0±16.5、0.985±0.011、0.809±0.097)、左肾(0.817±0.224、6.5±3.6、0.842±0.224、0.794±0.224)、右肾(0.856±0.104、14.4±13.2、0.913±0.062、0.809±0.143)、左侧股骨(0.931±0.038、-1.2±6.9、0.880±0.067、0.991±0.003)、右侧股骨(0.920±0.015、-9.4±9.0、0.877±0.055、0.970±0.034)、直肠(0.577±0.093、-12.9±59.5、0.523±0.077、0.736±0.301)、膀胱(0.966±0.025、-2.4±3.3、0.955±0.040、0.978±0.014)、前列腺(0.940±0.078、-1.4±3.6、0.934±0.092、0.946±0.063)、脑干(0.672±0.106、2.9±19.9、0.688±0.167、0.664±0.050)、右眼球(0.961±0.017、-2.6±1.9、0.948±0.025、0.973±0.011)、左眼球(0.823±0.050、-5.2±4.9、0.929±0.046、0.980±0.021)、下颌骨(0.699±0.191、15.2±54.5、0.719±0.070、0.716±0.294)、左侧腮腺(0.585±0.060、-48.1±3.2、0.445±0.056、0.854±0.053)、右侧腮腺(0.591±0.041、-47.3±2.9、0.451±0.024、0.859±0.087)。结论:左右肺、脊髓、肝脏、左右肾、左右股骨、膀胱等器官的自动勾画可以基本满足临床要求或需很少的修改,而直肠、腮腺、下颌骨等器官需更多的调整。基于模型分割算法在自动勾画过程中可以人为调整和修改,临床医生经过适当培训和使用后,最终可以获得较理想的结果。 Objective:To evaluate and analyze the automatic delineation algorithm with model based segmentation(MBS)in clinical application.Methods:A total of 90 patients underwent radiotherapy in head and neck,chest and abdomen,pelvic were selected.The doctors and MBS that was included in the planning system delineated the outlines of the left and right lung,the left and right kidney,the spinal cord,the liver,the bladder,and the left and right femurs and so on.Taking the contour delineated by the doctor as reference,the parameters such as the Dice coefficient and the sensitivity index were used to evaluate the advantages and disadvantages of the MBS delineation organ.Results:The Dice coefficient,volume difference(%),sensitivity index and inclusiveness index of the organs delineated by MBS were left lung(0.930±0.013,-9.5±2.1,0.886±0.021,0.979±0.010),right lung(0.946±0.006,-8.4±1.6,0.906±0.013,0.989±0.002),spinal cord(0.877±0.041,-17.7±2.3,0.799±0.046,0.971±0.037),liver(0.886±0.055,23.0±16.5,0.985±0.011,0.809±0.097),left kidney(0.817±0.224,6.5±3.6,0.842±0.224,0.794±0.224),right kidney(0.856±0.104,14.4±13.2,0.913±0.062,0.809±0.143),left femur(0.931±0.038,-1.2±6.9,0.880±0.067,0.991±0.003),right femur(0.920±0.015,-9.4±9.0,0.877±0.055,0.970±0.034),rectum(0.577±0.093,-12.9±59.5,0.523±0.077,0.736±0.301),bladder(0.966±0.025,-2.4±3.3,0.955±0.040,0.978±0.014),prostate(0.940±0.078,-1.4±3.6,0.934±0.092,0.946±0.063),brainstem(0.672±0.106,2.9±19.9,0.688±0.167,0.664±0.050),right eye(0.961±0.017,-2.6±1.9,0.948±0.025,0.973±0.011),left eye(0.823±0.050,-5.2±4.9,0.929±0.046,0.980±0.021),mandible(0.699±0.191,15.2±54.5,0.719±0.070,0.716±0.294),left parotid(0.585±0.060,-48.1±3.2,0.445±0.056,0.854±0.053),right parotid(0.591±0.041,-47.3±2.9,0.451±0.024,0.859±0.087).Conclusion:The automatic delineation of the left and right lung,spinal cord,liver,left and right kidney,left and right femur,bladder,and other organs can basically meet the clinical requirements,or require few modifications,while the rectum,parotid,mandible and other organs need more adjustments.The feature of the MBS algorithm is that it can be artificially adjusted and modified during the automatic delineation process.After proper training and use,the clinician can finally obtain better results.
作者 纪天龙 吴珊 李光 JI Tianlong;WU Shan;LI Guang(Department of Radiation Oncology,the First Hospital of China Medical University,Liaoning Shenyang 110001,China.)
出处 《现代肿瘤医学》 CAS 北大核心 2021年第14期2521-2524,共4页 Journal of Modern Oncology
关键词 自动勾画 基于模型分割算法 放射治疗 automatic delineation modeled based segmentation algorithm radiation therapy
  • 相关文献

参考文献6

二级参考文献36

  • 1Young A, Wortham A, Wemick I, et al. Atlas-based segmentation improves consistency and decreases time required for contouring postoperative endometrial cancer nodal volumes[J]. Int J Radiat Oncol Biol Phys, 2011, 79(3): 943-947.
  • 2Langmack K, Perry C, Sinstead C, et al. The utility of arias-assisted segmentation in the male pelvis is dependent on the interobserver agreement of the structures segmented[J]. Br J Radiol, 2014, 87 (1043): 20140299.
  • 3Teguh DN, Levendag PC, Voet PW, et al. Clinical validation of atlas- based auto- segmentation of multiple target volumes and normal tissue (swallowing/mastication) structures in the head and neck[J]. Int J Radiat Oncol Biol Phys, 2011, 81(4): 950-957.
  • 4Chao KS, Bhide S, Chen H, et al. Reduce in variation and improve efficiency of target volume delineation by a computer-assisted sys- tem using a deformable image registration approach[J]. Int J Radiat Oncol Biol Phys, 2007, 68(5): 1512-1521.
  • 5Gambacorta MA, Valentini C, Dinapoli N, et al. Clinical validation of atlas-based auto-segmentation of pelvic volumes and normal ti- ssue in rectal tumors using auto-segmentation computed system[J]. Acta Oncol, 2013, 52(8): 1676-1681.
  • 6Han X, Hoogeman MS, Levendag PC, et al. Atlas-based autosegm- entation of head and neck CT images[M]. Med Image Comput Com- put Assist Interv-MICCAI 2008. Berlin: Springer, 2008: 434-441.
  • 7Stapleford L J, Lawson JD, Perkins C, et al. Evaluation of automatic atlas-based lymph node segmentation for head-and-neck cancer[J]. Int J Radiat Oncol Biol Phys, 2010, 77(3): 959-966.
  • 8Hwee J, Louie AV, Gaede S, et al. Technology assessment of automated atlas based segmentation in prostate bed contouring[J]. Radiat Oncol, 2011, 6(110): 1-9.
  • 9Anders LC, Stieler F, Siebenlist K, et al. Performance of an atlas- based autosegmentation software for delineation of target volumes for radiotherapy of breast and anorectal cancer[J]. Radiat Oncol, 2012, 102(1): 68-73.
  • 10Sims R, Isambert A, Gr6goire V, et al. A pre-clinical assessment of an atlas-based automatic segmentation tool for the head and neck [J]. Radiat Oncol, 2009, 93(3): 474-478.

共引文献74

同被引文献13

引证文献1

二级引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部