Researchers have developed an AI tool that can help doctors predict who might develop a potentially fatal heart condition, just from an ECG.
The tool, developed by UK researchers, uses artificial intelligence to detect the earliest signs of a serious condition called 'complete heart block' – which can be fatal if left untreated.
Heart block is caused by problems with the electrical signals moving from the upper to lower chambers of the heart as it beats. The signal can only get through sometimes or not at all. (This makes the heart beat more slowly or skip beats. People with heart block may experience fainting, fatigue, injuries from falls, or even sudden death.
At present, doctors use clues from an ECG - which records the electrical activity of the heart and is one of the most common medical tests in the world – along with international guidelines to identify people who have or are likely to develop complete heart block. But these findings are non-specific and can miss patients who might benefit from earlier treatment – particularly as the abnormal heartbeat in the early stages of the condition can be intermittent.
In a study, published today in JAMA Cardiology, researchers at Imperial College London and Imperial College Healthcare NHS Trust show how they have trained their AI tool to read an ECG and identify patients who will have future heart block problems.
The AI tool, known as AIRE-CHB, performed much better than existing methods for predicting heart block. It was able to correctly identify the risk of the patient developing complete heart block, (from high to low), in 89% of cases. That compares to the existing standard of 59% correct identification.
Dr Arunashis Sau, Academic Clinical Lecturer at Imperial College London's National Heart and Lung Institute, and cardiology registrar at Imperial College Healthcare NHS Trust, said: "When complete heart block occurs it can initially be intermittent and therefore difficult to identify; yet by the time it becomes permanent it is far more dangerous for the patient. If we doctors can better identify patients early on with this condition, we'll be able to monitor them more closely or progress to treating them with a pacemaker - avoiding their serious injury, emergency hospital admission, or even their death."
The researchers believe their tool could be especially valuable for evaluating patients who have unexplained fainting episodes, which sometimes may be due to heart block. The cause of these episodes could be diagnosed earlier, using AIRE-CHB.
To develop their tool, the researchers analyzed over 1.1 million ECG heart recordings from nearly 190,000 patients at a Boston hospital. They used this data to train an AI system to recognize patterns that predict future complete heart block. They then tested their tool on a separate group of over 50,000 people from the UK Biobank.
People identified as high-risk by AIRE-CHB were about 7-12 times more likely to develop complete heart block compared to the individuals identified as low-risk. Overall, AIRE-CHB was 84%-94% (averaged at 89%) accurate in its predictions, depending on the population tested.
Dr Fu Siong Ng, the senior author, Reader in Cardiac Electrophysiology at the National Heart & Lung Institute at Imperial College London and a consultant cardiologist at Imperial College Healthcare NHS Trust and Chelsea and Westminster Hospital NHS Foundation Trust, said "Complete heart block is a very serious condition affecting around 24,000 patients a year in the UK. Our tool could provide reassurance for patients and their doctors that their condition could be diagnosed earlier and that we can tailor the frequency of monitoring and timing of pacemaker implants for those at high risk of developing this condition."
The research follows on from the team's development of the related AI-ECG risk estimation model, known as AIRE, which can predict patients' risk of developing and worsening disease from an ECG. Other AI models from this project have been trained to analyse ECGs to predict problems such as female heart disease risk, health risks including early death, high blood pressure, type 2 diabetes and most recently, heart valve disease.
Trials of AIRE in the NHS are planned for late 2025. These will evaluate the benefits of implementing the model with real patients from hospitals across Imperial College Healthcare NHS Trust and Chelsea and Westminster Hospital NHS Foundation Trust.
This research was funded by the British Heart Foundation via a BHF Clinical Research Training Fellowship to Dr Sau, a BHF Programme Grant to Dr Fu Siong Ng, and the BHF Centre of Research Excellence at Imperial College London. It was also supported by the NIHR Imperial Biomedical Research Centre, a translational research partnership between Imperial College Healthcare NHS Trust and Imperial College London.
-
'Artificial intelligence–enhanced electrocardiography for complete heart block risk stratification' by Arunashis Sau, Fu Siong Ng et al, is published in JAMA Cardiology. DOI: 10.1001/jamacardio.2025.2522
