UC Davis Medical Center

STEMI heart attacks require treatment within 90 minutes — delays triple mortality risk. Standard emergency department triage produced false positive rates of nearly 42%, overwhelming clinical teams with unnecessary interventions while also missing true STEMIs. Accurate, rapid triage at first medical contact was a critical unmet need.

UC Davis Medical Center participated in a multi-site U.S. study evaluating the Queen of Hearts AI-based ECG platform (pmcardio-stemi by Powerful Medical) for STEMI triage. The AI model analyzed initial ECGs from 1,000+ patients suspected of STEMI across three geographically diverse hospitals between January 2020 and May 2024, automatically distinguishing true STEMI from false positives at first medical contact.

The AI model correctly identified 553 confirmed STEMI cases on the initial ECG compared to 427 detected by standard triage methods. False positive rates dropped dramatically from ~42% with standard triage to ~8% with the AI model. Findings were published in JACC: Cardiovascular Interventions and presented at the 2025 TCT conference, demonstrating potential to shorten time-to-treatment and reduce unnecessary cath lab activations.

• AI-based ECG analysis at first medical contact can meaningfully outperform standard triage in both sensitivity and specificity for STEMI detection.
• Reducing false positive activations from 42% to 8% has major implications for resource utilization and clinical team burden in emergency settings.
• AI should be used as a decision-support tool alongside clinical judgment, not as a standalone diagnostic replacement.