MR影像组学诺模图预测子宫内膜癌宫颈间质浸润的价值  

Radiomics Nomogram for Prediction of Cervical Stromal Invasion in Endometrial Cancer

作　　者：方如旗[1] 陈霞平[1] 章国伟 薛晓铃 翁淑萍[1] 杨青霞 周作福[1] FANG Ruqi;CHEN Xiaping;ZHANG Guowei(Department of Radiology,Fujian Maternity and Child Health Hospital,Fuzhou,Fujian Province 350000,P.R.China)

机构地区：[1]福建省妇幼保健院影像科,福州350000

出　　处：《临床放射学杂志》2022年第6期1105-1109,共5页Journal of Clinical Radiology

基　　金：福建省卫生健康科研人才培养项目(中青年骨干人才培养项目)(编号:2019-ZQN-26)。

摘　　要：目的探讨基于MR影像组学的诺模图术前预测子宫内膜癌(EC)宫颈间质浸润(CSI)的价值。方法回顾性分析156例EC患者术前临床及MR资料。利用MaZda软件获取影像组学特征,采用LASSO筛选参数并建立影像组学得分。利用多因素Logistic回归方法建立预测模型,并以诺模图呈现,采用受试者工作特征曲线(ROC)评价模型的预测效能。结果CSI+56例,CSI-100例,常规MR模型、影像组学模型、诺模图模型诊断CSI的曲线下面积(AUC)为分别为0.80、0.90、0.91,影像组学模型及诺模图模型预测效能优于常规MR模型(P<0.05)。结论基于MR影像组学的诺模图对EC CSI术前个体化预测具有较高的价值。Objectives To evaluate the value of radiomics nomogram for preoperative prediction of cervical stromal invasion(CSI)in patients with endometrial cancer(EC).Methods The clinical and MR data of 156 consecutive EC patients were retrospectively analyzed.The radiomic features were extracted by MaZda software,LASSO analysis was used for feature selection and radiomics signature building.Multivariable Logistic regression analysis was used to develop the predicting model and a nomogram was presented.The performance of the models to predict CSI was assessed by ROC analysis.Results 56 cases were CSI+and 100 patients were CSI-in this study.The conventional MRI,radiomics and nomogram created AUC of 0.80,0.90,0.91 to predict CSI.Both nomogram and radiomics models shown superior dignostical performance than conventional MRI(P<0.05).Conclusion A nomogram base on MRI radiomics is useful for preoperative individualized prediction of CSI in patients with EC.

关 键 词：子宫内膜癌 宫颈间质浸润 影像组学 诺模图 

分 类 号：R737.33[医药卫生—肿瘤] R445.2[医药卫生—临床医学]