Explainable machine learning models predict liver fibrosis risk and outcome in the general population: Development and multi-cohort external validation.
A machine-learning model using routine blood tests accurately predicts liver scarring risk across different patient populations, enabling non-invasive screening.
An explainable XGBoost model using only routine clinical variables accurately predicts liver fibrosis risk and outcomes across multiple external validation cohorts, enabling non-invasive population screening. Multi-cohort validation strengthens generalizability over single-site ML models.
What the study was
- Study design
- Retrospective cohort with multi-cohort external validation
- Population
- General population cohorts for liver fibrosis prediction; multi-cohort validation design
- Category
- Diagnostics
- Maturity
- Validated
- Journal
- Computer methods and programs in biomedicine
Why it surfaced
Explainable ML with multi-cohort validation for liver fibrosis; practical model using routine variables; rigorous validation design.
A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.