AI Science Tracker
EN

Machine-learning variable selection supports 90-day heart-failure risk prediction

University of Florida · 23 Dec 2020 · medicine

Discovery

Machine-learning variable selection supports 90-day heart-failure risk prediction

Source check: Source checked. Not independent scientific confirmation.

Finding: In 3,189 patients hospitalized for heart failure, machine-learning variable selection supported a logistic model for 90-day acute-heart-failure readmission or death. LASSO selection and logistic regression yielded a reported AUC of 0.760, compared with 0.744 for logistic regression alone. This single-centre retrospective analysis used internal cross-validation; external performance and benefits from model-guided care were not tested.

Technology: LASSO; gradient boosting; random forest; support vector machine

Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.

AI contribution
Four-fold ML selection compared LASSO, GBM, RF, SVM.
Test and result
Retrospective single-centre 3,189-patient analysis with four-fold variable-selection cross-validation and LR comparator.
Limitation / still unknown
External validation, prospective performance and benefits of model-guided care remain unestablished.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://pmc.ncbi.nlm.nih.gov/articles/PMC7852168/

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review date: 21 Nov 2026. Review scope: Primary full paper Methods, Results and limitations plus Europe PMC bibliographic metadata;reviewed 2026-11-21..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 2/5 · Validation 2/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open in the full catalog for filters, charts and related discoveries. Browse medicine discoveries.

Related catalog entries

Links use documented relationships where available, plus shared catalog context. A shared topic or model family does not establish a scientific dependency.