News

Researchers at Imperial develop AI tool designed to read ECGs for signs of heart block

A team of researchers at Imperial College Healthcare NHS Trust have created an AI tool designed to read ECGs and support doctors identify those at risk of developing the potentially fatal heart condition, heart block.

AIRE-CHB is an AI tool that has been developed with support from the NIHR Imperial Biomedical Research Centre and funding from the British Heart Foundation. It has been trained on over 1.1 million ECG heart recordings from 190,000 patients in Boston and tested on a further 50,000 people from the UK, said to allow it to detect the earliest signs of heart block and identify patients who will encounter problems later on in life if the condition is left untreated.

Explaining more on the condition, 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.”

According to the trust, AIRE-CHB “performed much better than existing methods for predicting heart block”, with a reported 89 percent success rate. This compares to the existing standard where doctors use clues from an ECG while also following international guidelines, with 59 percent of cases identified correctly. Those who have been identified as high-risk by the AI are “about 7-12 times more likely to develop complete heart block compared to the individuals identified as low-risk”, the trust shared.

“Complete heart block is a very serious condition affecting around 24,000 patients a year in the UK,” Dr Fu Siong Ng, the senior author and reader in cardiac electrophysiology at Imperial College London, commented. “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.”

This is just one of the many AI models being developed as part of the trust’s wider ECG project, with other tools being trained to analyse ECGs in relation to female heart disease risk, health risks including early death, high blood pressure, type 2 diabetes and heart valve disease.

Digital tech and heart health: the wider trend 

Barking, Havering and Redbridge University Hospitals NHS Trust are taking part in a wider NHS study, trialling the use of AI scanners to help reduce hospital readmissions for people with heart failure, while also allowing patients to manage their condition at home. Developed by the medical technology manufacturer, Heartfelt Technologies, the scanner is an at-home device that takes “thousands of detailed images” of patients as they get in and out of bed to help detect the build-up of fluid in their legs.

A PhD study supported by the Royal Papworth Charity’s Innovation Fund has been exploring the use of wearables in monitoring patients with congenital heart disease. They’re looking to assess accuracy in estimations of peak VO₂ levels during everyday activities, hoping to enable earlier intervention in cases of deterioration.

NHS England recently published data on the use of the AI-driven, 3D heart scan technology, Heartflow, demonstrating its impact across 56 different NHS hospitals in England when diagnosing and supporting patients with suspected heart disease. According to a recent study, over 24,300 patients have benefited from the use of this technology since it was introduced in 2021, with Nature Magazine reporting that it has led to a reduction of patients needing invasive angiogram tests by 16 percent.