Acute Coronary Syndrome Risk Prediction Using Portable Cable-Free ECG Device Combined With Clinical Risk Assessment.
JACC. Advances · 2026-05-21 · Method & validation
Abstract
BACKGROUND: Patient uncertainty in symptom interpretation during acute coronary syndrome (ACS) contributes to patient-related presentation delays, increasing morbidity and mortality. Integrating clinical context with electrocardiogram (ECG) analysis in a portable patient-operated ACS risk assessment device may facilitate early self-triage. OBJECTIVES: The objective of the study was to evaluate the diagnostic performance of an algorithm combining pre-existing atherosclerotic cardiovascular risk (PER), symptom risk (SR), and portable ECG recorded using a hand-held cable-free device for ACS risk estimation. METHODS: This prospective single-center study enrolled 212 consecutive emergency department patients with chest pain; 184 patients with complete data were analyzed (96 learning set and 88 test set). Three independent components: PER, SR, and ST-vector loop parameters, derived from a three-lead cable-free credit card-size ECG device were integrated into a hierarchical fusion ACS risk assessment model. In 98 patients, asymptomatic postevent reference portable ECG was obtained 9 to 12 months later. Algorithm performance was evaluated using receiver operating characteristic curve analysis and compared with human consensus interpretation of 12-lead ECG plus PER and SR information. RESULTS: Algorithm ACS prediction using a single portable ECG achieved area under the curve of 0.865, increasing to 0.929 with postevent reference comparison (P = 0.036). At matched sensitivity, the model specificity was comparable to the human consensus (false-positive rate 0.527 vs 0.458; P = 0.709) and improved using reference comparison (model false-positive rate 0.198 vs human 0.556; P = 0.004). CONCLUSIONS: An algorithm combining clinical risk factors, symptoms, and ECG recording using hand-held cable-free device provides clinically meaningful ACS risk stratification. If fully integrated into a personal device, it may help support early self-triage decisions and reduce patient-related prehospital delays.