Radiomics-based Non-invasive Classifier Research for HCC-related Liver Transplantation

Radiomics-based Non-invasive Classifier Research for HCC-related Liver Transplantation

The purpose of this study is to establish a non-invasive radiomics method to filter high recurrent-risk liver transplantation recipient population

Hepatocellular carcinoma is one of the most common malignant tumors in the world with high incidence and mortality. Liver transplantation is the most effective treatment for HCC and is in high demand in China. However, recurrence of HCC among liver transplantation recipients is still a great challenge and threat to the survival of recipients. So it is of great significance to establish a non-invasive way to filter high liver transplantation recipient population before transplantation. Based on large scales of pre-treatment MR images, the investigator's study aims to establish a non-invasive radiomics method to filter high recurrent-risk liver transplantation recipient population. This study will contribute to the precise selection of HCC-related liver transplantation recipients.

Inclusion Criteria:

• All patients had postoperative pathological diagnosis of HCC
• All patients had underwent contrast-enhanced MRI scan within 2 weeks before liver transplantation

Exclusion Criteria:

• patients whose pathological diagnosis is not HCC
• patients whose CT images were affected by strong imaging artifacts, i.e. artifacts obscuring more than 10% of whole volume of interest
• patients whose clinical data or CT images were missing