AI 'provides early diagnosis' of heart problems

A senior doctor is hopeful that an "exciting" breakthrough has been made in the early diagnosis of heart problems using AI technology.

Dr Simon Rudland, visiting professor at the University of Suffolk, was the lead author of a study evaluating a test which uses AI (artificial intelligence) to determine if patients have cardiovascular disease before they show symptoms.

It uses five electrodes - four on the chest and one on the back – and returns a green, amber or red score.

This means the technology measures the heart's electrical activity in three dimensions, unlike the traditional two-dimensional ECG (electrocardiogram).

AI can then interpret the data, including measuring the rhythm, structure and perfusion of the heart muscle.

It is hoped the research could help with testing of people deemed to be in high-risk categories and could also help reduce hospital waiting times.

Dr Rudland, who is a board member of the Suffolk GP Federation, was part of the study assessing Cardisio, a 10-minute test developed in Germany.

"These are early days, and we need to do more work with more patients to go through the algorithm, but this is an exciting test," he said.

"Using digital technology to support patient diagnosis has the capacity to really change care pathways, helping to make referrals that are more appropriate or more specific to a patient's problem - as well as initiating treatment in a primary care setting rather than placing someone on a waiting list, and establish which patients need to be referred to hospital."

Dr Rudland added that a pilot may now be carried out in Suffolk or north Essex for women.

Cardiovascular diseases are conditions which affect the heart or blood vessels and can cause heart attacks, abnormal heart rhythms and heart failures.

The study focused on asymptomatic adults considered to be at risk of cardiovascular disease.

It featured 628 individual tests, with an independent consultant cardiologist reporting a strong association between red results and a referral to a cardiology clinic.

The data found a positive predictive accuracy of 80% and a negative predictive accuracy of 90.4% - with fewer than 2% of the tests failing.

The research paper, published in the BJGP Open Journal, concluded that the test "afforded high-risk, hard-to-reach individuals access to a test more effective at identifying underlining cardiovascular disease than a traditional 12-lead ECG".

It was considered that AI was able to analyse the huge amount of data in a way that a human could not.

Earlier this year, Health Secretary Wes Streeting said the NHS was "harnessing AI to deliver faster and smarter care" as part of the government's push to bring "our analogue NHS into the digital age".

Follow Suffolk news on BBC Sounds, Facebook, Instagram and X.