Heart failure is a major health problem that affects millions of people around the world. It happens when the heart cannot pump blood as well as it should, making it harder for the body to receive enough oxygen and nutrients.
Some people develop a weakened pumping heart, while others have a heart that pumps normally but has become too stiff to fill properly. Both forms can cause tiredness, shortness of breath, and swelling, but they are often difficult to detect in the early stages.
Doctors usually diagnose these conditions using ultrasound scans of the heart, also called echocardiograms. Although these tests are very useful, they require special equipment and trained staff. In many places, patients may wait weeks or months before they can have the test, making early diagnosis more difficult.
A team of researchers from Wake Forest University explored whether artificial intelligence could help solve this problem. Their study was published in the Journal of the American Heart Association. The researchers wanted to know if AI could find hidden signs of heart disease using a simple electrocardiogram, or ECG, which records the heart’s electrical activity.
They developed two different AI systems. One analysed the standard 12-lead ECG used in hospitals, while the other used information from only a single ECG lead. A single-lead ECG could one day be collected by wearable devices or portable monitors, making heart screening easier and less expensive.
The study was extremely large. The researchers trained and tested the AI using nearly 1.08 million digital ECG recordings collected from 165,243 patients. They also checked the results using another large group of children and teenagers, helping determine whether the technology could work across different age groups.
The AI successfully identified several important forms of heart disease, including reduced left ventricular ejection fraction, mid-range ejection fraction, and heart failure with preserved ejection fraction. These conditions are often challenging to recognise because symptoms can be similar to many other illnesses.
The results showed that the AI models performed very well, especially for detecting reduced pumping function. Even the single-lead system achieved strong results, suggesting that simple portable devices may eventually help identify people who need further heart testing before serious symptoms develop.
The researchers also compared the AI with machine-learning systems based only on ordinary clinical information such as age and medical history. Those models were less reliable, while adding clinical information to the ECG-AI produced little extra benefit. This showed that the ECG itself contains a surprising amount of useful information when analysed by AI.
If future studies confirm these findings, doctors could use AI-powered ECG screening in family clinics, pharmacies, ambulances, and even at home. Earlier diagnosis could allow treatment to begin sooner, helping reduce hospital admissions and improving quality of life for many patients.
The findings are encouraging because they show that a simple, low-cost heart test may become much more powerful when combined with artificial intelligence.
However, the technology should be viewed as a screening tool rather than a replacement for specialist examinations. Larger real-world studies are still needed before widespread use, but this research represents an important step toward faster, cheaper, and more accessible heart disease detection.
If you care about health, please read studies about the benefits of low-dose lithium supplements, and what we know about egg intake and heart disease.
For more health information, please see recent studies about potatoes and high blood pressure, and results showing 6 best breads for people with heart disease.
Source: Wake Forest University.
