AI tool able to find people with heart condition before they even have symptoms

A revolutionary new artificial intelligence tool is identifying individuals with a heart condition before they even exhibit symptoms.

The innovative tool searches GP records for "red flags" that may indicate a patient's risk of developing atrial fibrillation (AF). John Pengelly, a former Army captain who participated in the trial, expressed gratitude for the detection of his AF, which now requires him to take only a "couple of pills a day" to mitigate his heightened stroke risk.

AF is a heart condition characterized by an irregular and often abnormally fast heart rate, significantly increasing the risk of stroke. While some experience heart palpitations, dizziness, shortness of breath, and tiredness, others remain asymptomatic and unaware of their irregular heart rate.

Approximately 1.6 million people in the UK have been diagnosed with AF, but the British Heart Foundation (BHF) estimates that thousands more remain undiagnosed and unaware of their condition. Early identification and treatment of AF can effectively manage the condition and reduce stroke risk. The new AI tool is being evaluated in the Find-AF trial, funded by BHF and Leeds Hospitals Charity.

The algorithm was developed by University of Leeds and Leeds Teaching Hospitals NHS Trust scientists and clinicians, with BHF funding. The tool was developed using anonymised electronic health records from over 2.1 million individuals, training the algorithm to detect warning signs that could suggest a person is at risk of developing AF.

The tool was then validated with medical records from an additional 10 million people. Experts are currently evaluating its effectiveness in identifying individuals who may be at risk of developing AF within the next six months, with those identified being offered further testing.

The algorithm is currently analysing GP records at several surgeries in West Yorkshire. It calculates someone’s risk based on various factors including age, sex, ethnicity and whether they have other medical conditions such as heart failure, high blood pressure, diabetes, ischaemic heart disease and chronic obstructive pulmonary disease.

If individuals involved in the study are identified as high risk, they will be offered a handheld electrocardiography (ECG) machine to monitor their heart rhythm twice daily for four weeks, as well as any time they experience heart palpitations. If the ECG machine readings indicate that a patient has AF, their GP is informed and they can discuss treatment options.

Mr Pengelly, a retired grandfather, was diagnosed with AF earlier this year after participating in the Find-AF trial. The 74 year old, from Apperley Bridge, Bradford in West Yorkshire, spent 29 years in the Army Catering Corps before he retired.

"I got a letter inviting me to take part in the study and I thought if it benefits somebody then great, I want to help," he said. After being identified by the algorithm as high-risk, Mr Pengelly was provided with an at-home ECG.

"They sent me a little digital monitor and a few times a day I had to put my thumbs on it so it could take a reading, which took about two minutes. Then I pressed send and the reading went to the trial team. I did that for a few weeks, and I sent the kit back – it was really straightforward."

"I was diagnosed with AF a few weeks after that. I’d heard of it, but you never think that these things will happen to you. I didn’t have any symptoms. I’d occasionally get a bit breathless when I’m out and about, but that’s because there are so many hills around us and some of them are really steep.

"I’m really grateful it has been picked up. I now take a couple of pills every day to reduce my risk of having a stroke."

Experts believe that the West Yorkshire study could lead to a nationwide trial, potentially preventing numerous avoidable strokes. It's estimated that AF contributes to approximately 20,000 strokes annually in the UK.