Values Of Enhanced CT-based Radiomics For Evaluating Preoperative Liver Function Of Patient With Hepatic Alveolar Echinococcosis

Objective: To evaluate the value of preoperative hepatic reserve function in patients with alveolar hydatid by constructing an imaging model based on enhanced CT portal phase image features.Methods: A retrospective analysis was performed on 110 patients with hepatic alveolar echinococcosis confirmed by postoperative pathology in our hospital,According to the ALBI scoring standard,the patients were divided into ALBI grade 1(n =46)and ALBI grade 2(n = 64).Non-lesion liver regions were manually delineated as regions of interest(ROI)on contrast-enhanced portal phase CT images.were extracted:The optimal radiomics features were selected by Pearson correlation method,univariate selection method and Lasso CV.Results: 851 relevant features were extracted from the enhanced portal phase CT images.After dimension reduction,8 optimal features were selected to construct the random forest(RF)machine learning model.The AUC of the RF model in the training set and test set were 0.9764 and 0.7970,The sensitivity were 0.9111 and 0.7368,and the specificity were 0.9688 and 0.7857.Conclusion: The RF model established based on the radiomics features of preoperative contrast-enhanced portal venous phase CT images of patients with HAE can be used as an auxiliary test tool for preoperative liver reserve function in patients with HAE.