Artificial intelligence predicts outcomes for heart patients

12 February 2019        

Category: Research

For the first time researchers that we part-fund have used Artificial Intelligence to predict outcomes for heart patients using MRI scans, enabling doctors to find the best treatments for individual patients.

Artificial intelligence 3D heart

The computer program, called 4Dsurvival, takes each patient’s scan and tracks the motion of the heart at hundreds of points every second. From these 3D pictures of the moving heart the machine learns to predict the risk of dying from heart failure.

So far, the team at Imperial College London and the Medical Research Council have used the technology to predict the prognosis for 302 people with a heart condition called pulmonary hypertension (PH) a rare but serious condition, which damages the arteries in the lungs, and can be fatal.

The technology outperformed doctors, being able to correctly predict a patient’s prognosis 75% of the time. It is hoped that the new technology will help doctors to spot patients likely to deteriorate quickly enabling them to act immediately to halt the disease's progression.

The researchers are now looking to test the technology’s ability to help the estimated 920,000 people in the UK living with heart failure.

Our Medical Director, Professor Sir Nilesh Samani, described the enormous benefit that this research could result in for patients:

“This study shows the potential of machine learning and artificial intelligence to spot clinically useful information from heart scans - information that would be missed by the human eye. This could have enormous medical benefit by spotting problems earlier, at a stage when action can be taken. More research is necessary before this approach can become routine in the NHS, and there will always be a need for doctors to check that mistakes are not being made."

Read more about AI helping people with heart failure