基于急诊CT平扫影像组学在预测腹主动脉综合征中的价值  

The value of emergency CT-based radiomics in predicting abdominal aortic syndrome

作　　者：林文潇 苏洁惠 郭艺帆 陈少卿 吴爱琴[1] 程建敏[1] LIN Wenxiao;SU Jiehui;GUO Yifan;CHEN Shaoqing;WU Aiqin;CHENG Jianmin(Department of Radiography,the Second Affiliated Hospital of Wenzhou Medical University,Wenzhou 325027,China;Department of Radiography,the First Affiliated Hospital of Zhejiang Chinese Medical University,Hangzhou 310006,China)

机构地区：[1]温州医科大学附属第二医院放射影像科,浙江温州325027 [2]浙江中医药大学附属第一医院放射影像科,浙江杭州310006

出　　处：《温州医科大学学报》2022年第4期283-287,293,共6页Journal of Wenzhou Medical University

摘　　要：目的:探讨基于急诊CT平扫影像组学在腹主动脉综合征(AS)中的诊断价值。方法:收集2012年8月至2020年10月于温州医科大学附属第二医院行腹部CT平扫和增强的急诊患者145例。根据是否患有腹部AS,将所有患者按分层抽样法以7:3分为训练集和验证集。逐层手动勾画CT平扫图像上的腹主动脉并从中提取影像组学特征。经特征筛选后构建影像组学标签,并以此计算每位患者的影像组学评分(Radscore)。采用单因素和多因素Logistic回归分析寻找预测腹部AS的临床危险因素。最后使用多因素Logistic回归构建基于临床危险因素和Rad-score的联合模型,绘制列线图。使用受试者工作特征(ROC)曲线评估模型的诊断性能。结果:单因素和多因素Logistic回归分析显示,腹痛(OR=0.48,95%CI=0.19~1.22,P=0.12)、钙化斑内移(OR=8.76,95%CI=3.27~23.45,P<0.001)是腹部AS的危险因素,其构建的临床模型在验证集的ROC曲线下面积(AUC)为0.79。8个影像组学特征被选择用于构建影像组学标签,其在验证集的AUC为0.88。将该标签与临床模型结合获得联合模型,其在验证集的AUC为0.89。临床决策曲线显示,联合模型的临床实用性最优。结论:基于腹部CT平扫影像组学和临床危险因素构建的列线图能较好地预测急诊患者是否患有腹部AS。Objective:To evaluate the value of radiomics based on emergency CT plain scan in the diagnosis of abdominal aortic syndrome(AS).Methods:A total of 145 emergency patients who underwent plain and enhanced abdominal CT scan in the Second Affiliated Hospital of Wenzhou Medical University from August 2012 to October 2020 were collected.According to abdominal AS,all subjects were stratified randomly into the training set and the validation set in a ratio of 7:3.The abdominal aorta on CT plain scan images was manually segmented layer by layer,from which the radiomic features were extracted.The radiomic signature was constructed based on the selected features,and the radiomics score(Rad-score)of each patient was calculated.Univariate and multivariate logistic regression analyses were used to identify the risk factors for predicting abdominal AS.Finally,a nomogram based on clinical risk factors and Rad-score was constructed by multivariate logistic regression.Receiver operating characteristic(ROC)curve was used to evaluate the predictive performance of the model.Results:Univariate and multivariate Logistic regression showed that abdominal pain(OR=0.48,95%CI=0.19-1.22,P=0.12)and displaced calcification(OR=8.76,95%CI=3.27-23.45,P<0.001)were the clinical risk factors of abdominal AS.The area under the curve(AUC)of ROC curve of the clinical model in the test set was 0.79.Eight radiomic features were selected to construct radiomic signature with an AUC of 0.88 in the test set.The nomogram was obtained by combining the radiomic signature with the clinical model,and its AUC in the test set was 0.89.The clinical decision curve showed that the nomogram had the best clinical practicability.Conclusion:The nomogram based on radiomics of CT plain scan and clinical factors can well predict abdominal AS in emergency patients.

关 键 词：主动脉综合征 腹部 影像组学 体层摄影术 X线计算机 

分 类 号：R445.3[医药卫生—影像医学与核医学]