The AI tool which detects heart disease in 20 seconds

10 March 2022

New artificial intelligence (AI) technology is detecting heart disease in record speed and vastly improving care for heart patients at the Royal Free Hospital.

The first-of-its-kind AI tool analyses heart MRI scans in just 20 seconds whilst the patient is in the scanner. This compares to the 13 minutes or more it would take for a doctor to manually analyse the images after the MRI scan has been performed. It also detects changes to the heart structure and function, with 40 per cent greater precision and extracts more information than a human can.

The results of the research, funded by the British Heart Foundation, have been published in the Journal of Cardiovascular Magnetic Resonance today.

The technology is making a huge difference to patients at the Royal Free Hospital, University College Hospital, and St Bartholomew’s Hospital, where it is now being used to treat a combined total of over 140 patients a week.

Each year, around 120,000 heart MRI scans are performed in the UK. The researchers say that the AI will free-up valuable time of healthcare professionals – saving around 3,000 clinician days every year – so attention can be directed to seeing more patients waiting for treatment. The AI will also give patients and doctors more confidence in the results so that they can make better decisions about a patient’s treatment and possible surgeries.

The scientists and cardiologists who have developed the AI plan to extend the roll-out later this year to 40 other locations across the UK and globally.

Professor Marianna Fontana, professor of cardiology at UCL and director of the cardiovascular magnetic resonance unit at the Royal Free Hospital said: “This ultra-fast AI analyses complex heart scans in record speed, detecting the structure and function of a patient’s heart with more precision than ever before. This will have significant implication in clinical practice as it will unveil changes in structure and function that were not visible before, with enhanced accuracy and precision. This could also be very useful in assessing treatment response to new drugs at different stages of drug development."

The technology will improve diagnosis and treatment for a multitude of heart conditions. It is designed to diagnose a new heart condition when someone is first assessed for heart disease. It can spot early signs of heart disease - such as after having chemotherapy - which in some cases can cause damage to the heart, and it has the potential to screen for heart conditions in people with a family history of heart disease. The tool also helps doctors to see how patients with heart conditions are responding to their treatment, so they can then make any necessary adjustments.

The team trained the AI to measure the size of the left ventricle (the main pumping chamber of the heart), the thickness of the heart muscle and how well the left ventricle was able to pump blood around the body. They did this using heart MRI scans from 1923 people – including people with seven different heart conditions at 13 different hospitals and using 10 different models of MRI scanner. The AI was then validated on a further 109 patients who were scanned twice. They found that the AI technology analysed heart MRI scans more precisely than three doctors, removing the issue of subjectivity with human analysis. They found that the AI technology analysed heart MRI scans more precisely than three doctors, removing the issue of subjectivity with human analysis.

Dr Sonya Babu-Narayan, associate medical director at the British Heart Foundation, said: “This is a huge advance for doctors and patients, which is revolutionising the way we can analyse a person’s heart MRI images to determine if they have heart disease at greater speed.

“The pandemic has resulted in a backlog of hundreds of thousands of people waiting for vital heart scans, treatment and care. Despite the delay in cardiac care, whilst people remain on waiting lists, they risk avoidable disability and death. That’s why it’s heartening to see innovations like this, which together could help fast-track heart diagnoses and ease workload so that in future we can give more NHS heart patients the best possible care much sooner.”

The study was a collaboration between researchers at the Royal Free London NHS Foundation Trust, UCL, Barts Heart Centre at St Bartholomew’s Hospital, Queen Mary University of London, and the National Institutes of Health (NIH) in the USA.