What Imaging Modality Is More Effective in Predicting Early Recurrence of Hepatocellular Carcinoma after Hepatectomy...

Area of Science:

• Radiology
• Oncology
• Medical Imaging Analysis

Background:

• Early recurrence (ER) of hepatocellular carcinoma (HCC) after hepatectomy is a critical concern.
• Preoperative imaging modalities like CT and MRI are used for prediction, but comparative studies on their effectiveness with radiomics are lacking.

Purpose of the Study:

• To compare the efficacy of CT and MRI radiomics in predicting early recurrence of HCC post-hepatectomy.
• To evaluate the performance of combined CT and MRI radiomics models for ER prediction in HCC.

Main Methods:

• Retrospective analysis of 119 HCC patients who underwent preoperative CT and MRI.
• Extraction and selection of radiomic features using MRMR and LASSO algorithms.
• Construction of predictive models using Support Vector Machines (SVM) and logistic regression, including clinical-radiological-pathological (CRP) traits.

Main Results:

• CT and MRI radiomics models showed no significant difference in predicting HCC recurrence individually (p=0.911).
• Combined CT and MRI radiomics demonstrated superior predictive performance (p=0.032, 0.039).
• The combined model achieved high AUC values (0.951 training, 0.955 test) for stratifying high-risk patients, outperforming individual models.

Conclusions:

• Individual CT and MRI radiomics models do not significantly differ in predicting HCC early recurrence.
• A multimodal radiomics model integrating both CT and MRI data significantly enhances the prediction of early recurrence in HCC patients post-resection.