Artificial intelligence in prehospital assessment of acute coronary syndrome: a scoping review
AI analyzing heart rhythm tracings shows strong promise for recognizing heart attacks in ambulances, though real-world testing is still limited.
This scoping review of 19 studies (n=319,709) finds that AI — particularly ECG-based deep learning — demonstrates strong diagnostic performance (AUC up to 0.99) for prehospital ACS assessment, with emerging applications in risk stratification and clinical decision support beyond initial diagnosis. Evidence remains limited by methodological heterogeneity and lack of prospective multicenter validation studies.
What the study was
- Study design
- PRISMA-ScR scoping review; 19 studies (n=319,709 patients); AI-based prehospital ACS diagnosis, prediction, risk stratification
- Population
- Patients with suspected acute coronary syndrome in prehospital/emergency settings (aggregated from 19 included studies)
- Sample size
- 319709
- Category
- Diagnostics
- Maturity
- Exploratory
- Journal
- BMC Emergency Medicine
Why it surfaced
PRISMA-compliant scoping review with large combined dataset (n=319,709). Confirms maturation of AI-based prehospital ECG diagnosis. Score 6 reflects review nature and wide AUC variability indicating methodological heterogeneity.
A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.