基于三维CT影像的计算机辅助肺结节检测原型系统的临床应用
The clinical application of computer aided pulmonary nodule detecting system on account of three-dimentional CT.
阳世雄 1曾建业 1金人超 2黄启良 1许向阳 2韦涌初 1黄思光 1卢苇 1王斌强1
作者信息
- 1. 530031广西南宁,广西医科大学第三附属医院心胸外科
- 2. 华中科技大学计算机科学与技术学院
- 折叠
摘要
目的 探讨基于三维CT影像的计算机辅助肺结节检测原型系统在临床应用的效果.方法 分析2009年6月至2011年6月68例肺结节病例,分别采用单纯螺旋CT检查和应用基于三维CT影像的计算机辅助肺结节检测原型系统检测,以术后病理学检查结果 为"金标准",计算两种检查方法 判断肺结节良恶性的准确度.结果 单纯CT检查判断肺结节良恶性的准确度为75%,Kappa值0.504,术后病理检查结果 一致性一般;基于三维CT影像的计算机辅助肺结节检测原型系统检测,判断肺结节良恶性的准确度为91.6%,Kappa值0.805,术后病理检查结果 一致性较好.结论 基于三维CT影像的计算机辅助肺结节检测原型系统具有较好的临床应用价值,可以用于辅助医师进行CT图像中肺癌等肺部疾病的早期诊断.
Abstract
Objective To investigate the clinical application effect of computer aided pulmonary nodule(PN)detecting system on account of three-dimentional CT(3DCT)imaging.Methods 68 cases with PN wereincluded in this study in our hospital from Jun.2009 to Jun.201 1.All cases underwent simple spiral CT scan andcomputer aided PN detecting system on account of 3DCT scan.Postoperative pathological resuh wa.s considered as the“gold standard”.The two checking methods were calculated to determine the diagnostic accuracy of benign andmalignant PN.Results The accuracy of distinguishing benign fl'om maligant PN through simple spiral CT scan was0.75,the kappa value was 0.504,pathological resuh after operation had conlnlon con'espondence,the accuracy ofdistinguishing benign from maligant PN through computer aided PN detecting system on account of 3DCT imaging was0.916,the kappa value was 0.805,pathological result after operation had better con'espondence.Conclusions Computer aided PN detecting system on account of 3 DCT imaging has better clinical applicationvalue,and can help doctor carry out early diagnosis of lung disease(such cancer etc.)through CT image.
关键词
硬币病变,肺/诊断,计算机辅助/成像,三维Key words
Coin lesion,pulmonary/Diagnosis,computer-assisted/Imaging,three-dimentional引用本文复制引用
基金项目
广西科技开发项目(桂科攻0816004-18)
出版年
2012
NETL
NSTL
维普
万方数据