Deep Learning-Based Automatic Segmentation Combined with Radiomics to Predict Post-TACE Liver Failure in HCC Patients

Shuai Li,1,* Kaicai Liu,1,* Chang Rong,1,* Xiaoming Zheng,1 Bo Cao,2 Wei Guo,3 Xingwang Wu1 1Department of Radiology, the First Affiliated Hospital of AnHui Medical University, Hefei, Anhui Province, People’s Republic of China; 2Department of radiology, the Second affiliated hospital of NanJing Medical University, Nanjing, Jiangsu Province, People’s Republic of China; 3Department of Radiology, the Second Affiliated Hospital of ShanDong First Medical University, Taian, Shandong Province, People’s Republic of China*These authors contributed equally to this workCorrespondence: Xingwang Wu, Department of Radiology, The first affiliated hospital of AnHui medical university, No. 218 Jixi Road, Shushan District, Hefei, 230022, People’s Republic of China, Email [email protected]: To develop and validate a deep learning-based automatic segmentation model and combine with radiomics to predict post-TACE liver failure (PTLF) in hepatocellular carcinoma (HCC) patients.Methods: This was a retrospective study enrolled 210 TACE-trated HCC patients. Automatic segmentation model based on nnU-Net neural network was developed to segment medical images and assessed by the Dice similarity coefficient (DSC). The screened clinical and radiomics variables were separately used to developed clinical and radiomics predictive model, and were combined through multivariate logistic regression analysis to develop a combined predictive model. The area under the curve (AUC), calibration curve, and decision curve analysis (DCA) were applied to compare the performance of the three predictive models.Results: The automatic segmentation model showed satisfactory segmentation performance with an average DSC of 83.05% for tumor segmentation and 92.72% for non-tumoral liver parenchyma segmentation. The international normalized ratio (INR) and albumin (ALB) was identified as clinically independent predictors for PTLF and used to develop clinical predictive model. Ten most valuable radiomics features, including 8 from non-tumoral liver parenchyma and 2 from tumor, were selected to develop radiomics predictive model and to calculate Radscore. The combined predictive model achieved the best and significantly improved predictive performance (AUC: 0.878) compared to the clinical predictive model (AUC: 0.785) and the radiomics predictive model (AUC: 0.815).Conclusion: This reliable combined predictive model can accurately predict PTLF in HCC patients, which can be a valuable reference for doctors in making suitable treatment plan.Keywords: hepatocellular carcinoma, TACE, liver failure, deep learning, radiomics