Deep Learning-Based Electrocardiogram Model (EIANet) to Predict Emergency Department Cardiac Arrest: Development and External Validation Study.
Source check: Source checked. Not independent scientific confirmation.
Finding: EIANet predicts emergency-department cardiac arrest from triage ECG images; full text reports development plus independent two-hospital validation, with performance lower externally.
Technology: EIANet applies a spatial-attention network and recall-balanced loss to image-based triage ECGs to predict ED cardiac arrest.
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- EIANet applies a spatial-attention network and recall-balanced loss to image-based triage ECGs to predict ED cardiac arrest.
- Test and result
- The paper reports EIANet performance on 571 cases/826 controls at NTUH and an independent FEMH cohort of 378 cases/713 controls; external AUROC is 0.803.
- Limitation / still unknown
- Independent second-hospital retrospective evaluation is reported; the authors state prospective real-world testing remains future work.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://europepmc.org/article/MED/40053733 · https://www.jmir.org/2025/1/e67576 · https://preprints.jmir.org/preprint/67576 · https://pubmed.ncbi.nlm.nih.gov/40053733/
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: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..
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 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.