GAN augmentation improves retrospective metabolomic prediction of incident myocardial infarction
Source check: Source checked. Not independent scientific confirmation.
Finding: The study used KORA S4 and F4 data from 1,454 participants without prior myocardial infarction, including 78 incident cases, to model seven-year risk from clinical and metabolomic variables. A GAN generated minority-class samples for an MLP predictor, and the combined approach improved reported accuracy by about two percentage points over comparison models. This is retrospective prediction in an imbalanced observational cohort, not evidence of causation or clinical utility.
Technology: Generative adversarial network data augmentation and multilayer perceptron prediction
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- A GAN synthesizes minority-class examples for training a metabolomics-plus-clinical risk prediction model.
- Test and result
- The GAN-augmented MLP was evaluated on KORA cohort data for seven-year incident myocardial infarction prediction; the study reports 78 cases among 1,454 participants.
- Limitation / still unknown
- The retrospective cohort is imbalanced and the review found no independent prospective validation or evidence of clinical utility.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://www.mdpi.com/2218-1989/14/5/258
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: 22 Nov 2026. Review scope: Full primary publisher article, including Methods, Results, author affiliations and publication history.
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 3/5 · 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.