AI Science Tracker
EN

Fujita Health University

1 discovery from 2021

Discoveries

Random forest evaluates two-year risks in a Japanese atrial-fibrillation registry

Source check: Source checked. Not independent scientific confirmation.

Finding: Random forest models evaluated two-year thromboembolism, major bleeding and mortality risks in 7,406 patients from the J-RHYTHM atrial-fibrillation registry. Thromboembolism discrimination exceeded logistic regression (C-statistic 0.66 versus 0.59), while mortality discrimination was comparable. This retrospective model evaluation used internal cross-validation; participants were Japanese, most received warfarin, and no direct oral anticoagulants were used. Clinical benefits from using the predictions were not tested.

Technology: Random forest

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

AI contribution
RF with 1,000 trees, training-only feature selection and hyperparameter tuning; 80:20 split and five-fold cross-validation. Prospective registry follow-up does not make this a prospective frozen-model evaluation.
Test and result
Retrospective 7,406-patient risk evaluation against logistic regression and established scores.
Limitation / still unknown
External transportability, clinical benefit and performance under current anticoagulant regimens 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/PMC8427975/

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 article Methods, Results and limitations plus Europe PMC bibliographic metadata; PMC full article..

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 the interactive profile for the timeline, or browse all organizations and model families.