New research finds AI-supported test identifies cardiovascular disease before symptoms shown

New research evaluating a test which utilises AI algorithms to determine if patients have cardiovascular disease before they show symptoms could help bolster community testing for high-risk individuals and reduce hospital waiting times.

The study, carried out by University of Suffolk Visiting Professor, Dr Simon Rudland, alongside Consultant Cardiologist Dr Nisar Shah from Sandwell and West Birmingham NHS Trust and the University of Wolverhampton’s Professor Alan Nevill, evaluated Cardisio™ – a test developed in Germany which interprets heart activity using cloud-based artificial intelligence algorithms.

The 10-minute test utilises five electrodes (four on the chest and one on the back) to monitor heart activity, with the Cardisio™ test returning a green, amber or red score. 

The technology measures the heart’s electrical activity in three dimensions rather than the traditional two-dimensional ECG, and uses AI to accurately interpret the data. That includes measuring the rhythm, structure and perfusion of the heart muscle.

Cardiovascular diseases – conditions which affect the heart or blood vessels causing heart attacks, abnormal heart rhythms and heart failure – disproportionately affect hard-to-reach individuals, and can cause long term health conditions or fatalities.

The £340,000 study, funded by the SBRI (Small Business Research Initiative) and Cardisio™, aimed to assess if the test could identify asymptomatic cardiovascular disease, and determine if testing in a community rather than hospital setting could reduce health inequalities.

The test was used in three settings in ethnically-diverse areas of the West Midlands between August 2023 and February 2024 – a general practice (GP) surgery in Handsworth, a pharmacy near Dudley and a community health centre in Quinton, with asymptomatic individuals aged 18 and above considered to be at risk of cardiovascular disease. 

The study included pre- and post-test questionnaires for patients to feed back on their experience, while results were also reviewed by an independent consultant cardiologist.

The study featured 628 individual tests, with the consultant cardiologist reporting a strong association between red Cardisio™ test results and a referral to a cardiology clinic being indicated.

The data found a positive predictive accuracy of 80 percent and a negative predictive accuracy of 90.4 percent in the opinion of the consultant cardiologist. Fewer than 2 percent of the tests failed, according to the study results. 

In addition, 87.5 percent of participants recommended the test. 

The research authors 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”.

The research paper was published in the BJGP Open Journal.

Dr Rudland, lead author of the study who is on the board at Suffolk GP Federation, said that the AI can analyse an enormous amount of data that a human being could not interpret.

Currently, health professionals have to interpret the results of ECG tests, which can be a tricky process, while the Cardisio™ test provides a clearer answer with a much larger amount of data to draw on with a greater degree of accuracy.

The test could provide a means of carrying out more accurate testing in community healthcare settings, ensuring the right patients are referred to hospital-based cardiologists, which could help to reduce overall waiting times. It also has the potential to improve access for those who can be difficult to reach, such as those in rural communities and ethnically diverse communities.

Dr Rudland said: “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.

“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.”

Conversations are now underway in the hope that a pilot can be carried out in Suffolk or north Essex for women – another group which traditionally do not access early diagnosis and may present differently to men.

Professor Alan Nevill from the University of Wolverhampton, said: “We believe the implications of this research are huge. It means serious illness can be detected quickly and it relieves some of the burden of work on overworked doctors all over the world.

“It has been a great privilege to play a role in this research over two years.”

To read the full results of the study, visit the BJGP Open website here.

Find out more about the University of Suffolk's Institute of Health and Wellbeing here.