According to the latest research, people at high risk of a heart attack can be identified at least five years before it strikes. Researchers from the University of Oxford have developed a new bio-marker, which uses artificial intelligence (AI) for the detection.
The team funded by the British Heart Foundation (BHF) have developed the bio-marker, or ‘fingerprint’ called the fat radiomic profile (FRP). This bio-marker can be used to reveal biological red flags in the perivascular space lining of the blood vessels, supplying to the heart. Furthermore, the tool identifies the chances of a heart attack in the future. This is done by detecting the factors such the inflammation, scarring, and changes to these blood vessels.
In the past, as a standard component of care, coronary CT angiogram (CCTA) was used. This can scan the coronary arteries to check for any narrowed or blocked segments. When no significant narrowing of the artery was present, people are sent home. This accounts for about 75 per cent of scans. However, people who have previously gone for check-ups, still have a heart attack in the future. Also, there are currently no methods used routinely by doctors that can spot all underlying red flags for a future heart attack.
It was widely accepted that if patient’s scan of their coronary artery shows no narrowing, then they were meant safe from a potential heart attack. However, Professor Charalambos Antoniades, professor of cardiovascular medicine and BHF senior clinical fellow at the University of Oxford and his team decided to challenge this issue. They used fat biopsies from 167 people undergoing cardiac surgery. After analysing the expression of genes from the sample collected, they then matched them to the CCTA scan images. This was done in order to determine which features best indicate changes to perivascular fat. The changes indicated include inflammation, scarring, and new blood vessel formation.
Next, the team compared the CCTA scans of the 101 people out of 5,487 individuals. These people went on to have a heart attack or cardiovascular death within five years. Their CCTA were then matched with controls who did not. This helped them to understand the changes in the perivascular space, indicating that patient is at higher risk of a heart attack.
Using machine learning, the researchers developed the FRP fingerprint that captures the level of risk. According to the research team, the more heart scans that are added, the more accurate the predictions will become. Thus, the more information embedded in the tool will become ‘core knowledge’.
They tested the performance of this perivascular fingerprint in 1,575 people in the SCOT-HEART trial, showing that the FRP had a striking value in predicting heart attacks, above what can be achieved with any of the tools currently used in clinical practice. By harnessing the power of AI, they have developed a fingerprint to find potential harmful characteristics around patients’ arteries. This has a huge potential to detect the early signs of disease. Doctors can then be able to take all preventative steps before a heart attack strikes, ultimately to save lives.
The team hopes the technology will enable a greater number of people to avoid a heart attack. They also plan to roll it out to healthcare professionals in the next year, with the hope that it will be included in routine medical practice alongside CCTA scans in the next two years.
The findings are being presented at the European Society of Cardiology (ESC) Congress in Paris and published in the European Heart Journal.
–originally published in New Scientist