Decision support for ischemia imaging based on AI analysis of low-cost/no-cost screening CT and ECG data summary atherosclerotic cardiovascular disease is the leading cause of global morbidity and…

The National Heart Lung and Blood Institute awarded Case Western Reserve University $1,573,126 on July 15, 2026, under the Cardiovascular Diseases Research program (CFDA 93.837) to develop an artificial intelligence-based decision-support tool that identifies patients at risk of myocardial ischemia using low-cost or no-cost screening data.

The tool will analyze CT calcium scores, electrocardiogram data, and clinical risk factors to serve as a gatekeeper strategy for determining which patients with elevated coronary artery calcium would benefit from advanced diagnostic imaging such as stress positron emission tomography, magnetic resonance imaging, single-photon emission computed tomography, echocardiography, or coronary CT angiography with fractional flow reserve. Current practice lacks standardized guidelines for ordering these resource-intensive advanced tests, which identify ischemia in only 3–6 percent of cases, leading to unnecessary testing and inconsistent clinical practices. The funded work opportunistically leverages readily available data to improve the precision and efficiency of ischemia risk stratification and physician-patient shared decision-making for preventive strategies.

Performance occurs in Cleveland, Ohio, with the period of performance extending through March 31, 2028. Case Western Reserve University's Office of Research Administration operates the award in affiliation with University Hospitals of Cleveland.

Decision support for ischemia imaging based on AI analysis of low-cost/no-cost screening CT and ECG data summary atherosclerotic cardiovascular disease is the leading cause of global morbidity and mortality, responsible for over 17 million deaths annually. The CT calcium score (CTCS) exam is a widely used, guideline-recommended, low-cost, low-radiation tool for early atherosclerosis detection. The coronary artery calcium agatston score SUP- ports physician-patient shared decision-making for precise preventive strategies, both pharmacologic and non- pharmacologic. While the agatston score serves as a surrogate for atherosclerosis burden, IT does not explicitly provide information on coronary stenoses that cause myocardial ischemia-a modifiable risk factor for mortality and morbidity, treatable via revascularization (e.g., stents or surgery). In practice, determining which patients with elevated agatston scores need further ischemia imaging evaluation (stress pet/spect/echo/MRI or ccta-derived ffrct) is unclear. These advanced tests are costly, resource-intensive, and low yield, with only ~3-6% identifying ischemia. Although physicians use agatston scores to guide additional testing decisions, there are no standard guidelines, leading to inconsistent practices and unnecessary testing. We will opportunistically use AI on readily available low-cost (arguably no additional cost) data (CTCS, ECG, and clinical risk factors (CRF)) as a gatekeeper strategy to identify patients at risk of ischemia who would, therefore, benefit from addi- tional advanced diagnostic imaging. We will create a decision-support tool for ordering advanced imaging exams such as stress pet/MRI/spect/echo or CTA with ffrct. The scientific premise is robust, supported by EVI- dence linking atherosclerosis burden to ischemia and our preliminary data showing ischemia prediction using both CTCS and ECG features independently. We will leverage the clarify program, the world's largest free CTCS initiative (>130K scans, >25K annually), and houston heartscan (>60K scans, >12K annually), providing many paired CTCS and ischemia evaluation exams for the training of our classification models. This innovative approach builds on preliminary results and a strong foundation, promising to improve diagnostic precision and reduce healthcare costs. We will create a decision support APP and determine its utility and usability. The goal of our entrepreneurial group is to translate the developed tool into clinical practice.