We use cookies to ensure that our site works correctly and provides you with the best experience. If you continue using our site without changing your browser settings, we'll assume that you agree to our use of cookies. Find out more about the cookies we use and how to manage them by reading our cookies policy. Hide

AI used to create inexpensive early detector of heart disease

Heart-EKG Graphic

8 Jan 2019

Researchers have developed an inexpensive and widely available detector for the early diagnosis of heart disease using artificial intelligence (AI).
Cassie Sims

A new study by the Mayo Clinic has found that AI, alongside a widely available clinical screen – the electrocardiogram (EKG) – can be used as an early detection system for heart disease. It specifically detects asymptomatic left ventricular dysfunction, a precursor to heart failure. The new test has found to be as accurate as other commonly used screens, such as mammography for breast cancer and cervical cytology for cervical cancer.

Heart failure affects over 500,000 people in the UK, reduces quality of life, and can lead to further complications, such as stroke and other cardiovascular diseases. One risk of heart failure is left ventricular dysfunction, which is generally characterised by structural problems in the heart, specifically a weak pump. If identified, asymptomatic left ventricular dysfunction is treatable.

Left ventricular dysfunction cannot generally be diagnosed by a standard EKG; the best existing test is a screen that measures associated peptide levels. However, the screen requires a blood test and results have been disappointing. Other available screens include imaging, such as computerised topology (CT), magnetic-resonance imaging (MRI) or echocardiograms, are typically used, but they are less accessible and can be expensive.

In this study, digital data from 625,326 paired EKG and echocardiograms were analysed. This data analysis was then used to create, train, validate and an AI neural network. The new screen uses a combination of standard EKG and digital processing by the AI neural network.

‘The ability to acquire a ubiquitous, easily accessible, inexpensive recording in 10 seconds – the EKG – and to digitally process it with AI to extract new information about previously hidden heart disease holds great promise for saving lives and improving health,’ said Paul Friedman, MD, senior author and Chair of the Midwest Department of Cardiovascular Medicine at the Mayo Clinic.

The results showed that the new screen could not only detect symptomatic left ventricular dysfunction, but patients with an increased risk of developing ventricular dysfunction. ‘In other words, the test not only identified asymptomatic disease, but also predicted the risk of future disease, presumably by identifying very early, subtle EKG changes that occur before the heart muscle weakness,’ added Dr Friedman.

DOI: 10.1038/s41591-018-0240-2

Related links:

Share this article