Electroencephalography-based machine learning models for hyperacute detection of large vessel occlusion stroke
Source check: Source checked. Not independent scientific confirmation.
Finding: EEG symmetry features were used to train machine-learning models for hyperacute large-vessel-occlusion stroke detection. In nested cross-validation across 336 patients, including 16 positive cases, the best observed AUC was 0.85, sensitivity 69% and specificity 86%. The small positive sample and cross-validation design limit the result; improved acquisition is needed before clinical implementation.
Technology: Machine-learning models using EEG symmetry features
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Machine-learning models used EEG symmetry features to classify suspected large-vessel-occlusion stroke.
- Test and result
- Across two prospective diagnostic studies, nested cross-validation evaluated 336 included patients, of whom 16 had anterior-circulation large-vessel occlusion. Naïve Bayes had observed AUC 0.85 (95% CI 0.77–0.93), sensitivity 69% and specificity 86%.
- Limitation / still unknown
- Only 16 positive cases were included; nested cross-validation is not prospective evaluation of a locked model. Authors state improved EEG acquisition is needed before implementation.
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/PMC13742134/ · https://pmc.ncbi.nlm.nih.gov/articles/PMC13742134/#methods · https://pmc.ncbi.nlm.nih.gov/articles/PMC13742134/#results
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 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.