AI Tool Helps Identify Heart Failure Risk in Diabetes Patients

Diabetic cardiomyopathy is a heart condition marked by abnormal changes in the structure and function of the heart, which increases the risk of heart failure in patients. Defining this condition has been challenging due to its asymptomatic early stages and the varied effects it can have on the heart. Machine learning has emerged as a tool to identify high-risk patients, potentially providing a more nuanced approach compared to traditional diagnostic methods. Researchers have now created a machine learning model capable of identifying patients with diabetic cardiomyopathy. The findings, published in the European Journal of Heart Failure, present a data-driven strategy to detect a high-risk diabetic cardiomyopathy phenotype, facilitating early interventions that could help prevent heart failure in this vulnerable group.

Phenotypes refer to the observable physical characteristics of individuals that confer specific biological traits. Researchers at UT Southwestern Medical Center (Dallas, TX, USA) analyzed data from the Atherosclerosis Risk in Communities cohort, which consisted of over 1,000 participants with diabetes but no prior history of cardiovascular disease. By examining a set of 25 echocardiographic parameters and cardiac biomarkers, the team identified three patient subgroups. One of these subgroups, comprising 27% of the cohort, was classified as the high-risk phenotype. Patients in this group showed significantly elevated levels of NT-proBNP, a biomarker associated with heart stress, along with abnormal heart remodeling features such as increased left ventricular mass and impaired diastolic function. Notably, the five-year incidence of heart failure in this subgroup was 12.1%, which was considerably higher than that in the other groups.

Following these findings, the researchers developed a deep neural network classifier to identify diabetic cardiomyopathy. When validated on additional cohorts, the model detected between 16% and 29% of diabetic patients as having the high-risk phenotype. These patients consistently displayed a higher incidence of heart failure. Clinically, this model could assist in targeting intensive preventive therapies, such as SGLT2 inhibitors, which are medications used to manage Type 2 diabetes, to those patients who are most likely to benefit. It may also enhance clinical trials focused on heart failure prevention strategies in diabetic patients. By offering a new method to identify individuals at risk for heart failure, the model could enable earlier and more proactive interventions, thereby improving patient outcomes and influencing future research in cardiovascular care.

“This research is noteworthy because it uses machine learning to provide a comprehensive characterization of diabetic cardiomyopathy – a condition that has lacked a consensus definition – and identifies a high-risk phenotype that could guide more targeted heart failure prevention strategies in patients with diabetes,” said senior author Ambarish Pandey, M.D., Associate Professor of Internal Medicine in the Division of Cardiology at UT Southwestern. “This builds on our previous work that evaluated the prevalence and prognostic implications of diabetic cardiomyopathy in community-dwelling adults. It extends those efforts by using machine learning to identify a more specific high-risk cardiomyopathy phenotype.”