show image

Sooraj Shah

Contributing Editor

Sooraj Shah is Contributing Editor of New Statesman Tech with a focus on C-level IT leader interviews. He is also a freelance technology journalist.

How the NHS is using deep learning to combat coronary heart disease

Coronary heart disease (CHD) is one of the leading causes of death in the UK; it’s responsible for more than 66,000 deaths each year, and it’s estimated that 2.3 million people in the UK are currently living with the disease.

Imaging is an important part of diagnosing people and managing treatment with a range of cardiovascular diseases, and the NHS uses Echo, MRI and CT scans to do this. A CT scan is used to do a coronary CT angiography, which involves taking images of the heart blood vessels and using the CT scan with some dye contrast into the blood vessels. This enables doctors to see if blood vessels are narrowing, causing patients to have chest pain or angina. The test has been around for the last 10 or 15 years but has increased in usage as radiation doses have decreased dramatically.

“The main advantage of the CT is if there is any disease there they should pick it up – so it’s great if there’s no disease, which is what most patients will see, because you can reassure them and tell them it’s not your heart that’s causing chest pain,” explains Dr Timothy Fairbairn, the clinical lead for cardiac diagnostics at Liverpool Heart and Chest Hospital.

However, the majority fall within the intermediate to severe range where the narrowing of their blood vessels can be anything from 30% to 80 per cent, and this is where it is very difficult for doctors to judge whether that narrowing is causing a functional problem. This then requires a further test, such as a stress MRI or a stress Echo test. Now, however, there is a test that can use the CT scan, which tells doctors whether there is a problem with blood flow.

“That’s where HeartFlow analysis comes in because it tells us that narrowing is causing limitation of blood flow and that helps us manage the patient and give appropriate diagnosis,” says Fairbairn.

Data from a coronary CTA is uploaded from the hospital’s system to HeartFlow’s software application running in the AWS cloud. HeartFlow uses deep learning and processes a huge data set in order to create a detailed digital 3D model of the patient’s coronary arteries following the scan. It then applies algorithms to solve millions of complex equations to assess the impact of any blockages on blood flow to the heart.

Liverpool Heart and Chest Hospital was contacted by medical tech company HeartFlow back in 2015-16 to take part in research. The hospital agreed, and it used the technology on patients who consented to being part of the research, and who were being investigated for suspected angina and chest pain. It is now one of 30 hospitals in the UK to use the tool.

The hospital found that the technology would take between four and five hours to come back with a diagnosis for each patient – and that it had a high level of accuracy.

“From a patient’s perspective, they want to make sure they get the right diagnosis as quickly as possible – with a CT scan, there is an area of uncertainty which would require another test

“These are good tests – such as stress MRIs or Echos but it could take another six weeks for that test, another day off work, and then another wait for results, and the main advantage apart from the high accuracy is that it can give us an answer without additional testing,” says Fairbairn.

Despite Fairbairn and his peers wanting to introduce the technology for general NHS patients, there were several obstacles in the way. Firstly, the National Institute for Health and Care Excellence (NICE) had to check whether the tool is beneficial for patients and cost-effective.

“On the HeartFlow analysis, it said the evidence supported its use and it was cost-effective and it could save the NHS £9m each year by avoiding inappropriate tests,” he says.

“We had the research element, clinical experience, an independent NICE review, and gauged with commissioners in getting it funded but it was difficult in terms of financial constraints for the NHS, because even though you can present data to them saying it was cost effective, they are very hesitant to make long-term decisions about financing these things,” he adds.

The company then had to submit an application to be part of the Innovation and Technology Payment (ITP) programme in order to receive funding for one year from NHS England. This comes to an end in April 2019, and HeartFlow is still seeking a one-year extension.

Fairbairn believes that the most likely way for HeartFlow to get funding is through local CCGs rather than a central NHS England pot. For him, it’s imperative that the NHS does keep using the technology.

“I think the key point is that as a doctor you want to offer your patient the best available treatments or tests to be able to manage them in the best way. Evidence from the data suggests it is one of the best tests available so I want to give the opportunity to give my patients that test,” he says.