CT平扫及增强影像组学在急性胰腺炎严重程度评估中的应用研究  

The application value of CT plain scan and enhanced imaging radiomics in the assessment of severity in acute pancreatitis

作　　者：李艳霞 蒋圣哲 陈强[1] 罗琳[1] LI Yanxia;JIANG Shengzhe;CHEN Qiang;LUO Lin(Department of Medical Imaging,the First Affiliated Hospital of Baotou Medical College,Inner Mongolia University of Science and Technology,Baotou 014010,China;Imaging Department,Baotou Sixth Hospital,Baotou 014010,China;Imaging Department,Hulunbuir People's Hospital,Hailar 021000,China)

机构地区：[1]内蒙古科技大学包头医学院第一附属医院医学影像科,内蒙古包头014010 [2]包头市第六医院影像科,内蒙古包头014010 [3]呼伦贝尔市人民医院影像科,内蒙古海拉尔021000

出　　处：《现代医学》2025年第3期356-361,共6页Modern Medical Journal

基　　金：内蒙古自治区自然科学基金项目(2021MS08003)。

摘　　要：目的:通过CT平扫及增强影像图像构建影像组学特征,评估急性胰腺炎(AP)患者病情严重性及其发展趋势,以预测AP的严重程度,为AP的临床管理提供一种新的工具。方法:回顾性收集2018年1月至2024年5月包头医学院第一附属医院CT平扫以及增强扫描AP患者159例。将患者按7∶3比例随机分配至两组,其中训练组为110例患者,验证组为49例患者。使用3D Slicer软件对胰腺整体区域进行逐层准确勾画,从中提取关键的纹理特征。通过数据集的标准化处理、单因素变量分析结合LASSO降维技术以及10折交叉验证法挑选出最佳特征,利用Logistic回归分析,独立构建不同阶段的影像组学模型,包括平扫期、动脉期、静脉期、延迟期、动静脉联合期、增强三期联合模型。通过受试者工作特征曲线下面积(AUC)值来评估其在判断AP严重程度的预测能力。结果:在训练组中,平扫期、动脉期、静脉期、延迟期、动静脉联合期、增强三期联合模型的AUC分别为0.898、0.867、0.847、0.892、0.890、0.893;在验证组中,平扫期、动脉期、静脉期、延迟期、动静脉联合期、增强三期联合模型的AUC分别为0.749、0.868、0.794、0.806、0.766、0.802。结论:增强三期联合模型的预测效能高于平扫期、动脉期、静脉期、延迟期及动静脉联合期模型,说明增强三期联合模型在AP患者的严重程度预测方面,具有潜在的临床应用价值。Objective:To construct imaging features through computed tomography CT plain scans and enhanced images to assess the severity and progression of acute pancreatitis(AP)patients,aiming to predict the severity of AP and provide a new tool for the clinical management of AP.Methods:159 AP patients were retrospectively collected from January 2018 to May 2024 at the First Affiliated Hospital of Baotou Medical College,including plain scans and enhanced CT scans.Patients were randomly assigned to two groups in a 7∶3 ratio,with the training group comprising 110 patients and the validation group comprising 49 patients.The entire pancreatic region was accurately delineated layer by layer using 3D Slicer software to extract key texture features.The best features were selected through standardization of the dataset,univariate analysis combined with LASSO dimensionality reduction,and 10-fold cross-validation.Logistic regression analysis was used to independently construct different stages of the radiomic model,including the plain scan phase,arterial phase,venous phase,delayed phase,combined arterial and venous phase,and the enhanced three-phase combined model.The predictive ability in determining the severity of AP was evaluated through the area under the curve(AUC)value.Results:In the training group,the AUC values for the plain scan phase,arterial phase,venous phase,delayed phase,combined arterial and venous phase,and enhanced three-phase combined model were 0.898,0.867,0.847,0.892,0.890,and 0.893,respectively;in the validation group,the AUC values were 0.749,0.868,0.794,0.806,0.766,and 0.802,respectively.Conclusion:The predict effectiveness of enhanced three-phase combined model outperformed the plain scan phase,aterial phase,venous phase,delayed phase,and combined arterial and venous phase models,indicating that the enhanced three-phase combined model has potential clinical application value in predicting the severity of AP patients.

关 键 词：急性胰腺炎 CT影像组学 纹理特征 LOGISTIC回归 

分 类 号：R576[医药卫生—消化系统] R445.3[医药卫生—内科学]