The abstract is read at the publisher; the summary is JClub's.
A learned hepatic phenotype model for MASLD in HALO-MASLD predicts future liver status with predictive correlation r=0.860 and R^2=0.671, outperforming LSM and CAP benchmarks.
- Why it matters: Existing prognostic tools for MASLD often fix endpoints in advance, which may cap achievable accuracy and limit clinically interpretable risk tracking.
- What they did: Using Sparse Canonical Outcome REgression with broken adaptive ridge regression, the model learned a weighted phenotype from eight hepatic measures and 99 baseline predictors, then was nested five-fold cross-validated and externally validated in NIDDK NAFLD Adult Database 2.
- The result: In NIDDK, predictive correlation was r=0.836 with predictive R^2=0.680, and the learned phenotype’s links to fibrosis and portal hypertension replicated, with bootstrap relearning showing mean correlation 0.947.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.