A new artificial intelligence-powered method shows promise in identifying patients with advanced heart failure with high accuracy, potentially expanding access to critical care for a condition that often goes undiagnosed. The research, a collaboration between Weill Cornell Medicine, Cornell Tech, Cornell Ann S. Bowers College of Computing and Information Science, Columbia University Vagelos College of Physicians and Surgeons and NewYork-Presbyterian, was published March 3 in npj Digital Medicine.
Currently, advanced heart failure is typically detected using cardiopulmonary exercise testing (CPET), a specialized procedure requiring dedicated equipment and trained personnel. This limits its availability, meaning an estimated 190,000 Americans with advanced heart failure do not receive appropriate care, according to the National Heart, Lung, and Blood Institute. The new AI method aims to overcome this diagnostic bottleneck by predicting peak oxygen consumption (peak VO2) – a key CPET measure – using cardiac ultrasound images and patient electronic health records.
“This opens up a promising pathway for more efficient assessment of patients with advanced heart failure using data sources that are already embedded in routine care,” said Dr. Fei Wang, the associate dean for AI and data science and the Frances and John L. Loeb Professor of Medical Informatics at Weill Cornell Medicine.
The project emerged from the Cardiovascular AI Initiative, a joint effort by Cornell, Columbia, and NewYork-Presbyterian to explore the application of AI in improving heart failure diagnosis and management. Dr. Nir Uriel, director of advanced heart failure and cardiac transplantation at NewYork-Presbyterian, explained that a group of over 40 heart failure specialists were consulted to identify areas where AI could have the greatest impact. Using AI to analyze cardiac ultrasound data to identify advanced heart failure patients emerged as a leading possibility.
Dr. Uriel, who is also the Seymour, Paul and Gloria Milstein Professor of Cardiology in the Department of Medicine at Columbia University Vagelos College of Physicians and Surgeons and an adjunct professor of medicine in the Greenberg Division of Cardiology at Weill Cornell Medicine, emphasized the collaborative nature of the research. “The close interaction between clinicians and AI researchers on this project ended up driving the development of new AI techniques that would not have been explored otherwise,” said Dr. Deborah Estrin, the Robert V. Tishman ’37 Professor of Computer Science at Cornell Tech, a professor in Cornell Bowers and a professor of population health sciences at Weill Cornell Medicine.
The machine learning model, developed by Dr. Wang’s team and led by Dr. Zhe Huang and Dr. Weishen Pan, processes data from multiple sources, including moving ultrasound images of the heart, related waveform imagery displaying heart valve dynamics and blood flow, and information from electronic health records. The model was initially trained on deidentified data from 1,000 heart failure patients treated at NewYork-Presbyterian/Columbia University Irving Medical Center.
Subsequently, the model was tested on a new set of 127 heart failure patients from three other NewYork-Presbyterian campuses. The results demonstrated an accuracy of roughly 85% in distinguishing high-risk patients, surpassing previous AI-based peak VO2 prediction tools. Researchers used a measure relating to the probability that a randomly chosen high-risk patient in the sample has a higher predicted risk than a randomly chosen lower-risk patient to assess performance.
NewYork-Presbyterian has recently expanded its heart transplant and heart failure services, making exceptional heart transplant care available at the NewYork-Presbyterian/Weill Cornell Medical Center campus on the Upper East Side, and expanding heart failure services throughout the New York region, particularly in Brooklyn and Queens.
The research team is now planning clinical studies to evaluate the new approach, a necessary step for U.S. Food and Drug Administration approval and widespread clinical implementation. Dr. Uriel stated, “If we can employ this approach to identify many advanced heart failure patients who would not be identified otherwise, then this will change our clinical practice and significantly improve patient outcomes and quality of life.”
Leave a Reply