Detecting Cardiovascular Disease with the CarDS Lab
Recent papers published in Nature Digital Medicine, JAMA Network Open, the Journal of the American College of Cardiology: Advances and Circulation have been featured in the news. Produced by the Cardiovascular Data Science (CarDS) Lab in the Department of Medicine, Section of Cardiovascular Medicine, the research broadly investigates wearable devices and AI as detection and prevention technologies for those at risk of cardiovascular disease. Study authors include Biomedical Informatics & Data Science faculty Cynthia Brandt and Rohan Khera, the latter of whom is also director of the CarDS Lab.
As covered in Stat News, CarDS Lab researchers developed a deep-learning algorithm to detect hidden cardiovascular disease based on electrocardiograms (ECGs). Researchers proposed a new training approach that accounts for noise and interference, which would otherwise degrade the quality of ECG readings taken in the real-world. In August, the Hartford Courant reported that CarDS researchers were also able to develop a deep learning approach that uses ECG images to automatically screen for Left ventricular systolic dysfunction. These results are promising for wearable health devices and offer “significant potential” for the detection of disease in under-resourced settings, where hospital-grade equipment might not be readily accessible.
Meanwhile, popular tech magazine The Verge highlighted CarDS findings on the use of smartphones and tablets to track health goals. Using data from the National Cancer Institute, investigators analyzed smart device usage among 16,092 participants. They found that although older patients and male patients were at higher risk of cardiovascular disease, they were also less likely to use a smart device to monitor aspects of their health. The research raises important reminders about access and equity. Although the development of new health technologies continues to advance, affordability and technical literacy pose significant barriers. Victoria Song for The Verge noted, “The big question is whether any of it matters if the people who would most benefit are unlikely to use them.”
For Khera, the research findings represent exciting possibilities. “This can transform how we identify people with diseases,” he wrote in an interview with the Yale Daily News.
News Clips
Faculty affiliated with Biomedical Informatics and Data Science were also featured in news publications including Slate magazine, MedPage Today and Yale School of Medicine news. Lee Schwamm, MD, Professor of Biomedical Informatics & Data Science; Dennis Shung, MD, MHS, PhD, Assistant Professor of Medicine (Digestive Diseases); and Janitza Montalvo-Ortiz, PhD, Assistant Professor of Psychiatry were interviewed for these articles.