Machine Learning for the ECG Diagnosis and Risk Stratification of Occlusion Myocardial Infarction at First Medical Contact

Abstract

Patients with occlusion myocardial infarction (OMI) and no ST-elevation on presenting ECG are increasing in numbers. These patients have a poor prognosis and would benefit from immediate reperfusion therapy, but we currently have no accurate tools to identify them during initial triage. Herein, we report 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, significantly boosting both precision and sensitivity. Our derived OMI risk score provided superior rule-in and rule-out accuracy compared to routine care, and when combined with the clinical judgment of trained emergency personnel, this score 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.

Estimating Ejection Fraction from the 12 Lead ECG among Patients with Acute Heart Failure

“…17 Each 12-lead ECG captured data for 10 seconds, digitalized at 500 samples per second, with a recording speed of 25 millimeters per second (mm/s) and a standard calibration of 10 mm/millivolts(mV). We electronically stored these ECGs in the ItelliVue ECG management database (Philips Healthcare) and linked them to patient records via medical record numbers and hospitalization dates 18,19 . We obtained PDF copies of the morphology and analysis performed by the Philips DXL advanced algorithm and extracted measurements.…”

Use of Artificial Intelligence in Improving Outcomes in Heart Disease: A Scientific Statement From the American Heart Association

“…26 Second, AI/ML algorithms can identify subtle and interrelated nonlinear patterns in the ECG often not recognizable to experts, enhancing disease phenotyping. 30 Third, because cardiac electrical activity may be affected before mechanical or structural abnormalities are evident on imaging, such algorithms may enable the identification of occult disease and prediction of impending disease. By segregating subtypes of similar conditions, AI/ML of the ECG may reveal novel phenotypes.…”

“A Nova Eletricidade”: Aplicações, Riscos e Tendências da IA Moderna

“…Sinais biomédicos (e.g., eletrocardiograma e eletroencefalograma), exames laboratoriais, dados genéticos (e.g., mutações no DNA e expressão gênica) e informações de prontuários médicos eletrônicos também foram utilizados com sucesso nas mais diversas especialidades médicas, e a evolução da IA vem possibilitando que os médicos façam diagnósticos mais rápidos e precisos. A IA foi empregada, por exemplo, para estratificação de risco em pacientes com infarto do miocárdio por oclusão [Al-Zaiti et al 2023], detecção precoce da doença de Alzheimer [Mahendran and PM 2022] e estimativa de risco de câncer de pulmão em 3 anos a partir de tomografia computadorizada e outras informações clínicas .…”