AI Cardiology Advances | Latest News
- An artificial intelligence algorithm, when used with standard electrocardiograms (ECGs), significantly improves the detection of low ejection fraction (EF), a key indicator of asymptomatic left ventricular systolic dysfunction...
- ALVSD,often arduous to detect,is characterized by low EF,which measures the heart's efficiency in pumping blood.
- The algorithm, developed jointly by Mayo Clinic departments and Mayo Clinic Platform, was evaluated in the EAGLE trial.
Artificial intelligence is revolutionizing cardiology. This news reveals how an AI-enhanced algorithm paired with routine ECGs radically improves the detection of low ejection fraction (EF), a critical marker of asymptomatic left ventricular systolic dysfunction (ALVSD). The implications are considerable: earlier diagnoses could dramatically reduce the risk of heart failure and mortality.This innovative AI tool, tested in a large-scale trial, allows for more accessible and cost-effective screening, potentially transforming standard clinical practices. Explore how this technology, developed by Mayo Clinic departments, improved diagnoses by 32% compared to standard care. news Directory 3 stays on the cutting edge of these advances.Discover what’s next in AI cardiology.
AI-Enhanced ECGs Improve Heart Failure diagnosis via Low Ejection Fraction Detection
Updated June 09, 2025
An artificial intelligence algorithm, when used with standard electrocardiograms (ECGs), significantly improves the detection of low ejection fraction (EF), a key indicator of asymptomatic left ventricular systolic dysfunction (ALVSD).This growth, according to Mayo Clinic Platform, could lead to earlier diagnoses and reduced risks of heart failure and death.
ALVSD,often arduous to detect,is characterized by low EF,which measures the heart’s efficiency in pumping blood. While echocardiograms can diagnose it, their cost prohibits routine screening. The new AI tool offers a more accessible method for identifying at-risk individuals.
The algorithm, developed jointly by Mayo Clinic departments and Mayo Clinic Platform, was evaluated in the EAGLE trial. The results, published in Nature Medicine, showed that primary care physicians using the AI tool increased low EF diagnoses by 32% compared to standard care. For every 1,000 patients screened, the AI system generated five new diagnoses of low EF.
The EAGLE trial involved over 22,000 patients managed by 358 clinicians across 45 clinics and hospitals. Clinicians in the intervention arm had access to the AI results when deciding whether to order an echocardiogram. Ultimately, nearly 50% of patients in this group underwent echocardiography, compared to 38.1% in the control group.
Xiaoxi Yao, with the Kern Center for the Science of Health Care Delivery, Mayo Clinic, reported that the intervention increased the diagnosis of low EF in the overall cohort.
The AI tool’s neural network has undergone rigorous testing. Initial research involved training the network on over 44,000 Mayo Clinic patients and then testing it on nearly 53,000 independent patients. Subsequent prospective studies have further validated the algorithm’s value in clinical practice.
What’s next
The AI-enhanced ECG represents a significant step toward integrating machine learning into routine clinical practice, potentially transforming how clinicians identify and manage heart conditions like ALVSD. Further research will likely focus on refining the algorithm and expanding its application to other cardiovascular conditions.