Machine learning in predicting hepatocellular carcinoma in patients with chronic viral hepatitis in Hong Kong: abridged secondary publication.

1. Novel machine-learning models generate accurate risk scores for hepatocellular carcinoma (HCC) in patients with chronic viral hepatitis. 2. HCC ridge score is consistently more accurate than existing HCC risk scores. 3. These machine-learning models may be incorporated into electronic health systems to guide cancer surveillance strategies and reduce cancer death.