A deep learning radiomics model based on CT images for predicting the biological activity grading of hepatic cystic echinococcosis

Abstract

Abstract Objective This work aims to explore the potential applications of a deep learning radiomics (DLR) model, which is based on CT images, in predicting the biological activity grading of hepatic cystic echinococcosis. Methods A retrospective analysis of 160 patients with hepatic echinococcosis (109 cases of CE1 and 51 cases of CE2) was performed. A training set of 127 cases and a validation set of 33 cases were randomly divided from the data. Volume of interests (VOIs) were drawn from each patient’s CT image, and radiomics features and deep neural network features were extracted. Feature selection was performed on the training set, and radiomics score (Rad Score) and deep learning score (Deep Score) were calculated. Seven diagnostics models (based on logistic regression algorithm) for the biological activity grading were constructed using the selected radiomics features and two deep model features respectively. All models were evaluated using the receiver operating characteristic curve, and the area under the curve (AUC) was calculated. A nomogram was constructed using the combined model, and its calibration, discriminatory ability, and clinical utility were assessed. Results A total of 12 optimal features were selected from the radiomics features, and 6 and 10 optimal features were selected from two deep learning network (DLN) features (3D-ResNet-34, 3D-ResNet-50), respectively. For biological activity grading of hepatic cystic echinococcosis, the combined model demonstrated strong diagnostic performance, with an AUC value of 0.888 (95%CI:0.837-0.936) in the training set and 0.876 (0.761-0.964) in the validation set. The clinical decision analysis curve indicated that the DLR model had a greater clinical benefit than the single radiomics model and deep feature model, while the calibration curve revealed that the nomogram's prediction result was highly compatible with the actual result. Conclusion The DLR model can be used for predicting the biological activity grading of hepatic echinococcosis.