New
HealthMore in Health →
AI Predicts Liver Cancer Risk from Routine Blood Tests
A random forest model trained on 500,000 UK Biobank participants achieved 0.88 AUROC predicting liver cancer risk using only routine clinical data — and found 69% of cases had no prior liver disease diagnosis.
Key Takeaways
- A random forest model achieved 0.88 AUROC predicting liver cancer risk using only routine blood tests and EHR data from 500,000+ patients
- Crucially, 69% of liver cancer cases in the study had no prior liver disease diagnosis — a gap current screening protocols miss entirely
- A simplified 15-feature version still outperformed all existing clinical risk tools and validated well across diverse US populations
DE
DT Editorial AI··via medicalxpress.com