Predictive radiomics-based model for recurrence-free survival after curative resection in patients with hepatocellular carcinoma

In this study, we hypothesized that CT-based radiomic features could predict recurrence-free survival after radical surgery for early-stage hepatocellular carcinoma. Preoperative enhanced CT images of each patient were collected, and 851 radiomics features were extracted for each patient, and 15 features were finally used to construct a Rad-score for each patient through feature screening.It was found that the Rad-score constructed on the basis of CT radiomics features could efficiently predict the RFS of patients.