Heart Disease and AI: An artificial intelligence tool can identify signs of heart failure and heart valve disease from a routine electrocardiogram (ECG) in less than two seconds, according to research presented at the European Society of Cardiology Congress 2026 in Munich.
The model was tested in around 67,000 patients in the US and identified up to 81% of people with heart failure and up to 90% of those with heart valve disease.
The research was funded by the British Heart Foundation (BHF) and involved researchers from Imperial College London.
AI finds patterns in routine ECGs
An ECG records the electrical activity of the heart and is routinely used to assess heart rate, rhythm and conditions such as abnormal heart rhythms.
The new AI model analyses the same ECG data but looks for subtle patterns associated with heart failure and valve disease that may not be apparent from a standard ECG interpretation.
Researchers said the technology had been trained using data from millions of patients.
The tool is not intended to diagnose heart failure or valve disease on its own. Patients flagged as being at higher risk would still need further assessment, usually including an echocardiogram, or heart ultrasound.
Also Read | AI just designed new viruses, and biology may never be the same
Could help prioritise patients for heart scans
Patients referred for an echocardiogram can sometimes wait weeks or months for the test. Researchers believe the AI system could help identify people who are more likely to have an underlying heart problem and move them forward for further testing.
Dr Sonya Babu-Narayan, consultant cardiologist and clinical director at the British Heart Foundation, said the technology could help identify high-risk patients earlier, although it would not detect every person with heart disease.
Earlier diagnosis of heart failure and valve disease can allow treatment to begin before the condition becomes more advanced.
Prof Fu Siong Ng of Imperial College London said the technology could be used to identify patients most at risk and prioritise them for heart ultrasound scans.
AI could also flag unsuspected heart disease
Researchers are also studying whether the model could be applied to ECGs performed for unrelated reasons.
An ECG taken during a hospital visit could potentially flag a person as being at high risk of heart failure or valve disease even when neither condition was initially suspected.
This could provide another route to earlier diagnosis, particularly in people who have not yet developed obvious symptoms.
Dr Ahmed El-Medany, a BHF clinical research fellow who led the Imperial College London analysis, said researchers are now interested in developing handheld ECG devices that incorporate the AI technology.
ECGs are among the most commonly performed medical tests worldwide, with an estimated one billion carried out each year. If the model performs well in wider clinical use, researchers believe it could help doctors decide which patients need more urgent cardiac imaging.
The findings were presented at the European Society of Cardiology Congress 2026 in Munich.