Heart Warning
AI Model Developed at the Technion's Faculty of Biomedical Engineering Predicts Heart Failure Years in Advance
Researchers at the Technion Faculty of Biomedical Engineering have achieved a breakthrough in the early detection of heart failure. They have developed DeepHHF, a pioneering artificial intelligence model that identifies patients at high risk of developing heart failure years before the onset of clinical disease, enabling preventive interventions that could spare patients significant suffering and potentially save lives.
The study, published in npj Digital Medicine, was led by Prof. Joachim Behar and Ph.D. student Eran Zvuloni of the Technion and collaborating researchers and physicians from Rambam Health Care Campus, Shaare Zedek Medical Center, the Hebrew University of Jerusalem, and Leumit Health Services.
Heart failure affects approximately 64 million people worldwide and is especially common among adults over the age of 65, affecting roughly 12% of this population in developed countries. The condition significantly reduces quality of life, causing symptoms such as fatigue, shortness of breath, edema, and exercise intolerance, and can ultimately lead to death. As a result, there is a growing emphasis on early diagnosis, which makes it possible to initiate preventive treatments that can improve outcomes and save lives.
The Technion-developed model, DeepHHF, was trained on approximately 70,000 Holter examinations performed at the Leumit Health Services. It analyzes standard 24-hour ambulatory electrocardiogram (Holter ECG) recordings obtained during routine home monitoring and identifies patients at elevated risk of heart failure. The model detects subtle abnormalities in ECG recordings that are often imperceptible to the human eye, generating an early warning that can precede the onset of heart failure by several years.
According to the study’s senior author, Prof. Joachim Behar, “To the best of our knowledge, no existing model can predict the risk of heart failure up to five years in advance using raw Holter ECG recordings. By relying on standard, non-invasive diagnostic tools, our model provides clinically valuable information that enables early identification of high-risk patients and timely preventive interventions, with the potential to reduce hospitalizations, suffering, and mortality.”
Collaborating researchers and physicians, including Dr. Ronit Almog from the Technion and Rambam Health Care Campus, Michael Glikson from Shaare Zedek Medical Center, Shany Brimer Biton from the Technion, Dr. Ilan Green and Izhar Laufer from Leumit Health Services, and Offer Amir from Hadassah Medical Center, also participated in the research.
To read the full paper, click here
