Machine learning for ECG diagnosis and risk stratification of occlusion myocardial infarction

• Published in: Nature medicine Vol. 29; no. 7; pp. 1804 - 1813
• Main Authors: Al-Zaiti, Salah S., Martin-Gill, Christian, Zègre-Hemsey, Jessica K., Bouzid, Zeineb, Faramand, Ziad, Alrawashdeh, Mohammad O., Gregg, Richard E., Helman, Stephanie, Riek, Nathan T., Kraevsky-Phillips, Karina, Clermont, Gilles, Akcakaya, Murat, Sereika, Susan M., Van Dam, Peter, Smith, Stephen W., Birnbaum, Yochai, Saba, Samir, Sejdic, Ervin, Callaway, Clifton W.
• Format: Journal Article
• Language: English
• Published: New York Nature Publishing Group US 01.07.2023; Nature Publishing Group
• Subjects: 631/114/1305; 692/308/575; 692/699/75/2/1674; Algorithms; Biomedical and Life Sciences; Biomedicine; Cancer Research; Diagnosis; EKG; Electrocardiography; Emergency Service, Hospital; Heart attacks; Humans; Infectious Diseases; Learning algorithms; Machine learning; Metabolic Diseases; Molecular Medicine; Myocardial infarction; Myocardial Infarction - diagnosis; Neurosciences; Observational studies; Occlusion; Reperfusion; Risk Assessment; Time Factors
• ISSN: 1078-8956; 1546-170X
• DOI: 10.1038/s41591-023-02396-3

Patients with occlusion myocardial infarction (OMI) and no ST-elevation on presenting electrocardiogram (ECG) are increasing in numbers. These patients have a poor prognosis and would benefit from immediate reperfusion therapy, but, currently, there are no accurate tools to identify them during initial triage. Here we report, to our knowledge, the first observational cohort study to develop machine learning models for the ECG diagnosis of OMI. Using 7,313 consecutive patients from multiple clinical sites, we derived and externally validated an intelligent model that outperformed practicing clinicians and other widely used commercial interpretation systems, substantially boosting both precision and sensitivity. Our derived OMI risk score provided enhanced rule-in and rule-out accuracy relevant to routine care, and, when combined with the clinical judgment of trained emergency personnel, it helped correctly reclassify one in three patients with chest pain. ECG features driving our models were validated by clinical experts, providing plausible mechanistic links to myocardial injury. A machine learning algorithm, developed to detect occlusion myocardial infarction with no-ST elevation from electrocardiogram, outperforms clinicians in diagnostic assessments.