AI is Better at Foretelling Risk of Heart Attack

AI is Better at Foretelling Risk of Heart Attack

Healthcare Tech Outlook | Monday, September 23, 2019

The new biomarker, 'fingerprint' will help the healthcare industry to save more lives by detecting early signs of a heart attack.

FREMONT, CA: Harnessing cutting-edge artificial intelligence technologies could bring new hope for people with heart diseases. The researchers of Oxford University have developed a new biomarker, or 'fingerprint', called the fat radiomic profile (FRP). By using machine learning, the fingerprint can detect biological red flags in the perivascular space lining blood vessels, which supply blood to the heart. It can find all indicators to a future heart attack such as inflammation and scarring or changes to these blood vessels.

The new finding related to heart ailment treatment is published in the European Heart Journal. These findings were also presented at the European Society of Cardiology (ESC).

The team of researchers surveyed data from fat biopsies of around 167 patients who were undergoing cardiac surgery. The team also investigated CT angiography (CCTA) images of those patients as a reference to see which features suggested changes to an individual's perivascular fat. The FRP biomarker was developed by using comparisons and machine learning. To carry out a comparison, the research group compared scans of two sets of people. The first set of patients includes scans of 101 patients who suffered a heart attack or cardiovascular death within five years. The second set consists of the people who did not experience a heart attack or cardiovascular death.

As compared to current methods, the new process will bring striking improvements in predicting cardiac risk. To prove it, the team tested the performance of this perivascular fingerprint in 1,575 people in the SCOT-HEART trial. It shows that the FRP had an outstanding value in predicting heart attacks, clearly better than other tools circulating in the market for clinical practices.

Generally, when a person goes to a hospital with chest pain, a coronary CT angiogram (CCTA) is given. This scan can check for any narrowed or blocked segments in the coronary arteries. If there is no significant narrowing of the artery, the patient is not diagnosed with any heart ailment. Still, they have chances of a heart attack in the future, which goes unpredicted by doctors, as they cannot mark all of the underlying red flags for a future heart attack.  The artificial system is better than human physicians at predicting the heart ailment.

The advanced technology will reduce heart attack largely, and most probably, it will roll out to health care professionals in the next year. In the next two years, it will be included in routine NHS practice alongside CCTA scans.

Professor of Cardiovascular Medicine and BHF Senior Clinical Fellow at the University of Oxford, Charalambos Antoniades says that a scan showing zero narrowings of the coronary artery does not mean that patient is safe from a heart attack. Now the AI-enabled method 'Fingerprint' has the vast potential to detect early signs of disease. Patients can take precaution before a heart attack strikes, and this will save many lives. He believes that this technology will prove to be a lifesaver for many in the coming years.

According to reports, every five minutes, a new person is admitted to a UK hospital due to heart ailment. It is a significant advance, which will revolutionize the healthcare industry.

Weekly Brief