摘要
目的 探究T1WI压脂增强静脉期轴位图像纹理分析来鉴别诊断肝内胆管细胞癌（ICC）与肝细胞癌（HCC）价值。方法 回顾性分析我院65例病例，病理确诊为肝内胆管细胞癌22例，为训练组，肝细胞癌43例，为测试组。上腹MRI平扫及增强扫描均在术前一周内完成。利用3d-Slicer软件对T1-Vibe压脂序列增强静脉期横断位图像勾画感兴趣区（ROI），提取各个病灶的3D纹理特征共851个，并筛选出有鉴别诊断意义的特征参数。结果 通过LASSO回归共筛选出9个有意义的纹理特征，采用logistic回归，进一步筛选出预测模型的3个影像学特征。该模型曲线下面积 （AUC） 为0.864，提示该模型有鉴别诊断价值。结论 基于T1WI压脂增强序列纹理分析来鉴别肝内胆管细胞癌与肝细胞癌，有较好的鉴别诊断价值。
关键词： T1WI；纹理分析；肝内胆管细胞癌；肝细胞癌
Abstract
Objective: To investigate the value of T1WI lipid compression enhanced axial image texture analysis in the differential diagnosis of intrahepatic cholangiocarcinoma (ICC) and hepatocellular carcinoma (HCC). Methods: Retrospective analysis was performed on 65 cases in our hospital. 22 cases were pathologically diagnosed as intrahepatic cholangiocarcinoma, which were in the training group and 43 cases were in the test group. Both plain and enhanced MRI scans of upper abdomen were completed within one week before surgery. 3D-Slicer software was used to sketch the ROI of t1-Vibe enhanced venous transect images, extract a total of 851 3D texture features of each lesion, and screen out characteristic parameters with differential diagnosis significance. Results: Nine significant texture features were screened out by LASSO regression, and three imaging features of the prediction model were further screened out by Logistic regression. The area under curve (AUC) of this model is 0.864, suggesting that this model has the value of differential diagnosis. Conclusion : T1WI lipid compression enhanced sequence texture analysis has good value in differentiating intrahepatic cholangiocarcinoma from hepatocellular carcinoma.
Key words： T1WI; Texture Analysis; Intrahepatic Cholangiocarcinoma; Hepatocellular Carcinoma
参考文献 References
[1] MA MT,FENG ZC,PENG T,et al.Radiomics and its advances in hep-atocellular carcinoma[J].Zhong Nan Da Xue Xue Bao Yi Xue Ban, 2019,44(3):225-232.
[2] SIA D,VILLANUEVA A,FRIEDMAN SL, et al.Liver cancer cell of origin, molecular class, and effects on patient prognosis[J].Gastroen-terology,2017,152(4):745- 761.
[3] WANG CH,WEY KC,MO LR,et al.Current trends and recent advances in diagnosis, therapy, and prevention of hepatocellular carci-noma[J].Asian Pac J Cancer Prev, 2015, 16(9): 3595-3604.
[4] 罗柳平,李建伟,曹利,等.腹腔镜肝切除术治疗肝内胆管癌的临床 疗效[J].中华消化外科杂志,2019,18(2): 169-175.
[5] YU JH,SHI ZF,LIAN YX,et al.Noninvasive IDH1 mutation estimation based on a quantitative radiomics approach for grade Ⅱ glioma[J].Eur Radiol,2017,27(8): 3509-3522.
[6] LAMBIN P,RINS VE, LEL JR,et al.Radiomics: extracting more information from medical images using advanced feature analysis[J].Eur J Cancer,2012,48(4):4414-4446．
[7] 黄冠,尹芳艳,李小雪,等.影像组学研究方法进展[J].实用放射学杂志,2019,35(2):308-311.
[8] KUMAR V,GU Y,BASU S,et a1.Radiomics: the process and the challenges[J].Magn Reson Imaging,2012,30(9): 1234-1248．
[9] PINHEIRO PS,CALLAHAN KE,JONES PD,et al.Liver cancer: a leading cause of cancer death in the United States and the role of the 1945-1965 birth cohort by ethnicity[J].IHEP Rep,2019,1(3): 162-169.
[10] 中华人民共和国卫生部.原发性肝癌诊疗规范(2011 年版)摘要[J].中华肝脏病杂志,2012,20(6):419-426.