Doctors have developed an artificial intelligence tool that can identify signs of heart failure and valve disease from routine electrocardiograms (ECGs) in less than two seconds. The technology, trained on millions of patient ECGs, analyzes the electrical activity of the heart to extract more detailed information than the human eye can typically discern. This advancement could lead to faster diagnoses for patients with common forms of heart disease.
The newly developed AI tool analyzes electrocardiograms (ECGs) to detect signs of heart failure and heart valve disease. This technology has been been trained on data from millions of patients, allowing it to identify subtle patterns within ECG readings that are not readily apparent to human clinicians. Traditional ECGs, used for over a century to record the heart's electrical activity, rate, and rhythm, are primarily used to diagnose heart attacks and abnormal rhythms. However, they have not been capable of detecting broader heart disease, which typically requires an echocardiogram, an ultrasound scan that can involve significant waiting times for patients.
The AI's ability to process ECGs rapidly could significantly improve the early detection of heart conditions. In a trial involving 67,000 patients in the United States, the AI tool successfully identified up to 81% of individuals with heart failure and up to 90% with heart valve disease. Dr. Sonya Babu-Narayan, a consultant cardiologist and clinical director of the British Heart Foundation (BHF), which funded the trial, stated that the AI can deliver an ECG read-out "in what feels like the blink of an eye." While the AI cannot definitively diagnose or rule out these conditions on its own, it provides a strong indication that further investigation is warranted.
The widespread use of ECGs, with approximately one billion performed globally each year, makes this AI development particularly significant. Dr. Ahmed El-Medany, a BHF clinical research fellow who led the analysis at Imperial College London, described the tool as a "superhuman AI." The next steps for this research include designing handheld AI-led ECG readers for healthcare professionals.
This advancement aligns with a growing trend of AI integration in cardiovascular diagnostics. For instance, research presented at the American Heart Association's Scientific Sessions in November 2025 demonstrated an AI algorithm paired with smartwatch ECG sensors that could accurately diagnose structural heart diseases. That study, which analyzed ECGs from 600 adults, showed the AI algorithm could identify issues such as weakened pumping ability, damaged valves, or thickened heart muscle with high accuracy.
Furthermore, the U.S. Food and Drug Administration (FDA) cleared an AI tool named EchoNext in June 2026, designed to detect hidden structural heart disease from routine ECGs before symptoms appear. Developed by researchers at NewYork-Presbyterian and Columbia University, EchoNext analyzes ECG waveforms to flag individuals at high risk for structural heart disease, prompting further echocardiogram testing. In a 2025 study published in Nature, EchoNext correctly identified 77% of structural heart problems from ECGs, outperforming cardiologists who achieved 64% accuracy on the same readings. Pathway Labs, the maker of EchoNext, has raised $8.5 million to expand its adoption.
Another study published in NEJM AI in December 2025 detailed an AI model developed at the University of Michigan that can diagnose coronary microvascular dysfunction, a condition often missed in emergency settings, using a standard 10-second ECG strip. This model significantly outperformed previous AI attempts in predicting myocardial flow reserve, a key diagnostic measure for this condition.
Researchers at UC Berkeley also reported in June 2026 the discovery of a previously unrecognized signal within ECGs, identified by a custom AI tool, that can better detect patients at high risk for sudden cardiac death. This discovery was made using over 440,000 ECGs from Sweden.
Additionally, a study led by UT Southwestern Medical Center researchers in Kenya, published in JAMA Cardiology in May 2026, explored the use of AI-augmented ECG analysis to screen for heart failure precursors in resource-limited settings. The AI-ECG approach demonstrated a high negative predictive value, suggesting it could be a practical and scalable method for identifying at-risk individuals where access to advanced diagnostics like echocardiography is constrained.
These developments collectively indicate a significant shift towards AI-powered diagnostic tools in cardiology, aiming to enhance early detection, improve accuracy, and broaden access to cardiac care globally.