Electrocardiographic Detection of Non-ST Elevation Myocardial Events for Accelerated Classification of Chest Pain Encounters (ECG-SMART) ABSTRACT There is a clear need to develop improved tools to stratify risk in patients who seek emergency care for chest pain, one of the most common and potentially deadliest conditions encountered in acute care settings. The ECG has been the mainstay of initial evaluation of chest pain patients, yet is currently only diagnostic for a small subset of patients with ST-elevation myocardial infarction. Our well-rounded and collaborative team has been working together for the past four years to develop novel ECG interpretation algorithms that can greatly enhance the utility of the ECG in the early evaluation of acute coronary disease. This line of work is already substantiated by pilot studies that have identified candidate markers of ECG characteristics and preliminary algorithms that can identify patients with non-ST elevation myocardial infarction as well as those with very low risk of coronary artery disease. With these improved tools, emergency providers (physicians, nurses, and paramedics) will be able to streamline the care provided to these patients beyond the costly and time- consuming overnight observation for serial cardiac enzymes and provocative testing. This grant will provide needed resources for our team to enroll a cohort of over 1,800 patients needed to confirm the accuracy of these ECG markers and determine their maximal clinical utility as part of a risk stratification tool. It builds directly on our existing work in this area, which has already yielded necessary preliminary data and multiple joint publications. Validating this technology will set the groundwork for the future development and testing of targeted interventions to improve patient outcomes in non-ST elevation myocardial infarction.

Public Health Relevance

SUMMARY/NARRATIVE Current guidelines for diagnosing and treating myocardial infarction (MI) are based on identifying ST elevation (STE) on the 12-lead ECG, yet only one third of MI patients exhibit such elevation. We propose to (1) extend our prior research of novel repolarization indices of non-STE MI available on the initial ECG and (2) incorporate this information into real-time machine-learning algorithms for clinical decision support, which will improve our sensitivity in detecting and treating MI very early during urgent care. Such early and non-invasively acquired diagnostics can improve outcomes and save costs for millions of patients with chest pain annually.

National Institute of Health (NIH)
National Heart, Lung, and Blood Institute (NHLBI)
Research Project (R01)
Project #
Application #
Study Section
Biomedical Computing and Health Informatics Study Section (BCHI)
Program Officer
Sopko, George
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Pittsburgh
Other Health Professions
Schools of Nursing
United States
Zip Code