期刊文献+

宫颈癌危及器官MIM-Maestro和RayStation软件自动勾画优势对比研究 被引量:10

Comparative study of two software tools on the atlas-based auto-segmentation of organs-at-risk in cervical cancer
原文传递
导出
摘要 目的放射治疗过程中危及器官的勾画繁琐耗时,自动勾画软件的出现在一定程度上提高了勾画效率。本研究探讨基于Atlas实施宫颈癌危及器官自动勾画时不同软件对自动勾画结果的影响。方法分别基于MIM-Maestro6.6.5和RayStation 6.0软件建立入库病例数目为60例的宫颈癌Atlas模板库,该60例患者由中国人民解放军总医院放射治疗科2014-07-01-2016-12-31收治,随机选择Atlas库外10例宫颈癌患者,由中国人民解放军总医院放射治疗科2017-01-01-2017-04-30收治,临床医生手动勾画危及器官(膀胱、直肠和双侧股骨头),并定义为参考勾画(Vref)。分别运用2款软件对10例患者的CT图像进行自动勾画。采用勾画时间(T)、相似性系数(dice similarity coefficient,DSC)、敏感性指数(sensitivity index,SI)、包容性系数(inclusiveness index,IncI)、质心偏差(deviation of centroid,DC)、Jaccard系数(jaccard index,JAC)、Hausdorff距离(hausdorff distance,HD)评价勾画结果,并进行配对样本t检验。结果MIM软件的勾画时间为(1.89±0.09)min,小于RayStation软件的勾画时间(8.17±0.63)min,Z=-2.81,P=0.01。膀胱的勾画结果显示,除HD〔(2.92±1.17)cmvs(3.68±1.51)cm,Z=-1.07,P=0.29〕外均差异有统计学意义,MIM软件勾画结果优于RayStation软件。对于直肠,2款软件勾画结果的各评价参数差异均无统计学意义,均P>0.05;从均值分析,MIM软件的结果在体积(DSC:0.53±0.08;JAC:0.37±0.08)上优于RayStation软件(DSC:0.48±0.18;JAC:0.34±0.16),但在位置上MIM软件〔DC:(1.58±0.75)cm;HD:(3.54±0.89)cm〕略差于RayStation软件〔DC:(1.56±0.52)cm;HD:(3.40±1.0)cm〕。对于双侧股骨头,2款软件的勾画结果差异均无统计学意义,勾画结果均较为理想(DSC左侧股骨头0.86±0.04 vs 0.86±0.05,Z=-0.12,P>0.05;DSC右侧股骨头0.85±0.03 vs 0.85±0.04,Z=0.36,P>0.05)。结论基于Atlas实施危及器官自动勾画时,MIM软件勾画效率高于RayStation软件。2款软件对于双侧股骨头的勾画结果处于较好水平,但对于直肠的勾画结果都不理想,在膀胱的勾画中MIM软件优于RayStation软件。 OBJECTIVE The contouring of organs during radiotherapy is time-consuming,the emergence of autosegmentation software improves the efficiency of delineation to some extent.The aim of this study was to evaluate the effect of two software tools on the atlas-based auto-segmentation of organs-at-risk in cervical cancer.METHODS The atlas of cervical cancer(the cases were 60)was established in MIM-Maestro 6.6.5 and Ray Station 6.0 respectively,these cases of cervical cancer were collected in department of radiotherapy of PLA General Hospital from 2014-07-01 to2016-12-31.Another 10 patients of cervical cancer were randomly selected out of the atlas which were also collected in department of radiotherapy of PLA General Hospital from 2017-01-01 to 2017-4-30.An experienced radiation oncologist contoured 4 OARs(bladder,rectum and femoral heads)based on the published guidelines which were defined as the reference volumes(Vref).The OARs were auto-contoured by two sets of software respectively.The similarity of the ABAS and reference contours were assessed using dice similarity coefficient(DSC),sensitivity index(SI),inclusiveness index(IncI),deviation of centroid(DC),jaccard coefficient(JAC)and hausdorff distance(HD).The time for manual and auto-contouring were also recorded.The results were statistically evaluated using paired t-test.RESULTS The delineated time of MIM was significantly less than that of RayStation[(1.89±0.09)min vs(8.17±0.63)min,Z=-2.81,P<0.001].The results of bladder contouring in MIM showed better than those in RayStation except the value of HD[(2.92±1.17)cmvs(3.68±1.51)cm,Z=-1.07,P=0.29]significantly(P<0.05).For the rectum,there were no significant differences in the analysis parameters of two tools,but MIM had the advantage in volume[VMIM vs VRS:0.53±0.08 vs 0.48±0.18(DSC);0.37±0.08 vs 0.34±0.16(JAC)]instead of in position[VMIMvs VRS:(1.58±0.75)cmvs(1.56±0.52)cm(DC);(3.54±0.89)cmvs(3.40±1.00)cm(HD)]for the analysis of mean values.For the bilateral femoral heads,the contouring results were both well and there was no significant difference between two sets of software[DSC:0.86±0.04 vs 0.86±0.05,Z=-0.12(left femoral head);0.85±0.03 vs 0.85±0.04,Z=0.36(right femoral head);both P>0.05].CONCLUSIONS MIM is higher in efficiency than RayStation for processing atlas-based auto-segmentation.The auto-contouring in two sets of software for bilateral femoral heads are kept at the good level,except for the rectum.In addition,MIM is superior to RayStation in the delineation of bladder.
作者 王金媛 郑庆增 张慧娟 杨微 曲宝林 徐寿平 WANG Jin-yuan;ZHENG Qing-zeng;ZHANG Hui-juan;YANG Wei;QU Bao-lin;XU Shou-ping(Department of Radiotherapy,PLA General Hospital,Beijing 100853,P.R.China;Department of Radiotherapy,Beijing Geriatric Hospital,Beijing 100095,P.R.China;Beijing Advanced Innovation Center for Big Date-based Precision Medicine,Beihang University,Beijing 100083,P.R.China)
出处 《中华肿瘤防治杂志》 CAS 北大核心 2019年第24期1889-1894,共6页 Chinese Journal of Cancer Prevention and Treatment
基金 解放军总医院临床科研扶持基金(2017FC-TSYS-3005) 国家重点研发计划(2017YFC0112100)
关键词 宫颈癌 危及器官 ABAS MIM RayStation cervical cancer organs at risk ABAS MIM RayStation
  • 相关文献

参考文献6

二级参考文献45

  • 1International Commission on Radiation Units and Measurements. Re- port 83 :Prescrilaing, recording and reporting intensity modulated plao- ton beam therapy (IMRT) (ICRU report 83) [M]. Bethesda, MD: ICRU, 2011 : 1-106.
  • 2Sun Y,X/u XL, Luo W, et al. Recommendation {or a contouring method and atlas of organs at risk in nasopharyngeal carcinoma patients receiving intensity-modulated radiotherapy [J]. Radiotla- er Oncol,2014,110(3) :390-397.
  • 3Nelms BE, Tom6 WA, Robinson G, et al. Variations in the con touring o{ organs at risk:test case from a patient with oroplaa- ryngeal cancer [J]. Int J Radiat Oncol Biol Plays, 2012,82 (1) : 368-378.
  • 4Han X, Hoogeman MS, Levendag PC,et aL Atlas based auto-segmen- tation of head and neck CT images [J]. Med Image Comput Assist Inter, 2008,11 (2) : 434-441.
  • 5Kouwenlaoven E, Giezen M, Struikmans H. Measuring similarity of target volume delineations independent of the number of ob- servers [J]. Phys Med Biol, 2009,54 (9) : 2863-2873.
  • 6Qazi AA, Pekar V, Kim J, et al. Auto-segmentation of normal and target structures in head and neck CT images:a feature-driv- en model-based approach [J]. Med Phys, 2011,38(11) : 160-170.
  • 7La Macchia M, Fellin F, Amiehetti M, et al. Systematic evalua- tion of three different commercial software solutions for auto- matic segmentation for adaptive therapy in head-and-neck, pros- tate and pleural cancer [J]. Radiat Orlcol,2012,7:160.
  • 8Sims R, Isambert A, Gregoire V, et al. A pre-clinical assessment of an atlas-based automatic segmentation tool for the head and neck [J]. Radiother Oueol, 2009,93 (3) : 474-478.
  • 9Teguh 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.
  • 10Simmat I,Georg P,Georg D, et al. Assessment of accuracy and efficiency o{ atlas-based autosegmentation for prostate radiotb.er- apy in a variety of clinical conditions [J], Strahlenther Onkol, 2012,188(9) : 807-815.

共引文献50

同被引文献87

引证文献10

二级引证文献19

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