The value of CT radiomics in predicting severe hyperlipidemic acute pancreatitis
10.3969/j.issn.1002-1671.2023.12.016
- VernacularTitle:CT影像组学预测重症高脂血症性急性胰腺炎的价值
- Author:
Qing JIA
1
;
Xiaohua HUANG
;
Shize QIN
;
Fang WANG
Author Information
1. 川北医学院附属医院放射科,四川 南充 637000
- Keywords:
hyperlipidemic acute pancreatitis;
radiomics;
prediction;
severity
- From:
Journal of Practical Radiology
2023;39(12):1980-1984
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore the value of CT enhanced radiomics model in predicting severe hyperlipidemic acute pancreatitis(HLAP).Methods The data of 117 HLAP patients were analyzed retrospectively and the patients were randomly divided into a training set(93 cases)and a test set(24 cases)in the ratio of 8∶2.CT enhanced images of arterial phase and venous phase were collected,and the optimal radiomics features were extracted and screened.The arterial phase model and venous phase model were established by support vector machine(SVM).Meanwhile,the bedside index for severity in acute pancreatitis(BISAP)was scored and a BISAP model was developed for the patients based on clinical information.The area under the curve(AUC)under the receiver operating characteristic(ROC)curve was used as the evaluation criterion for the model.DeLong test was used to compare the prediction efficiency of each model.Results The training set AUC of the arterial phase model,venous phase model and BISAP model were 0.777,0.788 and 0.732,respectively.And the AUC of the test set were 0.836,0.734 and 0.695,respectively.DeLong test results showed that the training set AUC of the arterial phase model and the venous phase model was better than that of BISAP model(P＜0.05),but there was no significant difference between the AUC of the test set AUC(P＞0.05).There was no significant difference in AUC between arterial phase model and venous phase model(P＞0.05).Conclusion The radiomics model based on CT enhancement can predict the severity of HLAP at the early stage,which helps to target the treatment of HLAP patients in the early clinical stage.
