AI Tools May Detect Heart Valve Disease, Predict Cardiovascular Risk

November 06, 2023 – Artificial intelligence (AI) and deep learning tools may be capable of accurately detecting heart valve disease and predicting the risk of cardiovascular disease events, according to two preliminary research studies to be presented at the American Heart Association (AHA) Scientific Sessions 2023.

The first study, “Real World Evaluation of an Artificial Intelligence Enabled Digital Stethoscope for Detecting Undiagnosed Valvular Heart Disease in Primary Care,” sought to compare the ability of a primary care professional to detect potential heart valve disease using a standard stethoscope versus that of an AI program using sound data taken from a digital stethoscope.

The study cohort was comprised of 369 adults without a prior diagnosis of heart valve disease or a history of heart murmurs who received primary care at clinics in Queens, New York, and Lawrence and Haverhill, Massachusetts.

Each participant received a physical exam from either a physician or a nurse practitioner, during which the healthcare provider listened to the patient’s heart and lungs with a standard stethoscope for unusual sounds or murmurs. Each patient in the cohort also had an exam in which their heart sounds were recorded using a digital stethoscope.

The digital stethoscope data were then fed to an AI tool for assessment.

READ MORE: American Heart Association Releases Guidance for Patient Data Sharing

At a follow-up appointment one to two weeks later, all patients also received an echocardiogram to determine if heart valve disease was present.

The analysis found that the AI was able to detect 94.1 percent of cases of valvular heart disease compared to the primary care professionals, who detected 41.2 percent of cases.

The AI tool also flagged 22 people with previously undiagnosed moderate-or-greater heart valve disease, whereas the clinicians identified eight previously undiagnosed patients.

“The implications of undiagnosed or late diagnosis of valvular heart disease are dire and pose a significant cost to our health care system,” said lead author Moshe Rancier, MD, senior medical director of Mass General Brigham Community Physicians in Lawrence, Massachusetts, in the press release. “This study demonstrates that health care professionals can screen patients for valvular heart disease more effectively and quickly using a digital stethoscope paired with high-performing AI that could detect cardiac murmurs associated with significant valvular heart disease.”

The second study, “Deep Learning-Based Retinal Imaging for Predicting Cardiovascular Disease Events in Prediabetic and Diabetic Patients: A Study Using the UK Biobank,” tasked a deep learning model with determining patients’ risk of cardiovascular disease events by evaluating eye images of people with prediabetes and Type 2 diabetes.

READ MORE: Deep Learning Model Automates Severe Heart Valve Disease Detection

The researchers defined cardiovascular disease events as heart attack, ischemic stroke, transient ischemic attack, or death due to heart attack or stroke.

The research team provided the deep learning tool with retinal images of 1,101 people with prediabetes or Type 2 diabetes and asked the model to categorize each individual into low-risk, moderate-risk, and high-risk groups based on likelihood of cardiovascular disease.

Then, the number of cardiovascular disease events among participants was tracked for approximately 11 years.

At the end of the study period, 12.5 percent of the cohort had experienced at least one cardiovascular event.

Of those who experienced a cardiovascular event, 8.2 percent were categorized into the low-risk group, 15.2 percent into the moderate-risk group, and 18.5 percent into the high-risk group.

READ MORE: Deep Learning-Based Electrocardiogram Screening Detects Heart Defects

After the researchers accounted for potential cardiovascular disease risk factors, such as age, gender, smoking history, and medication use for cholesterol and high blood pressure, they found that people in the moderate-risk group were 57 percent more likely than people in the low-risk group to experience a cardiovascular event. Likewise, people in the high-risk group were 88 percent more likely to experience a cardiovascular event than their peers in the low-risk group.

“These results show the potential of using AI analysis of retinal imaging as an early detection tool for heart disease in high-risk groups such as people who have prediabetes and Type 2 diabetes,” said study lead author Chan Joo Lee, MD, PhD, an associate professor at Yonsei University in Seoul, Korea. “This could lead to early interventions and better management of these patient groups, ultimately reducing the incidence of heart disease-related complications.”

These studies demonstrate the potential for AI tools in cardiovascular medicine.

“Computational methods to develop novel predictors of health and disease — ‘artificial intelligence” — are becoming increasingly sophisticated,” said Dan Roden, MD, FAHA, professor of medicine, pharmacology and biomedical informatics and senior vice-president for personalized medicine at Vanderbilt University Medical Center, as well as chair of the AHA’s Council on Genomic and Precision Medicine. “Both of these studies take a measurement that is easy to understand and easy to acquire and ask what that measurement predicts in the wider world.”