For the first time artificial intelligence (AI) has been used to instantly and accurately measure blood flow and the information obtained can predict the chances of death, heart attack and stroke, so it could be used to recommend treatments that could improve the patient’s blood flow and thus prevent these diseases.
This is explained by a study by the University College of London-UCL and the Barts Health NHS Trust (United Kingdom) which is published in “Circulation”.
Heart disease is the world’s leading cause of death and disease in the world. One of its symptoms is a reduced blood flow, which is often treatable, due to an obstruction of the arteries. Therefore, international guidelines recommend a series of evaluations to measure a patient’s blood flow, but many are invasive and with an added risk.
One of these techniques is cardiovascular magnetic resonance imaging (MRI), but so far, the images obtained have been incredibly difficult to analyze in a precise enough way to offer a prognosis or recommend a treatment.
With the clinical data of each patient it was discovered that those with reduced blood flow were more likely to have complications, including death, heart attack, stroke and heart failure
In this work the researchers worked with routine RMC images of more than 1,000 patients and used a new automated artificial intelligence technique to analyze the images. In this way, they were able to quantify precisely and instantaneously the blood flow.
When comparing the results of blood flow generated by AI with the clinical data of each patient, it was found that those with reduced blood flow were more likely to have complications, including death, heart attack, stroke and heart failure.
Professor James Moon (UCL Institute of Cardiovascular Sciences and Barts Health NHS Trust) said: «Artificial intelligence is moving from computer labs to the real world of health, performing some tasks better than doctors could do alone We’ve tried before to measure blood flow manually, but it’s tedious and time-consuming, since it takes doctors away from where they are most needed, with their patients. ”
“The predictive power and reliability of AI were impressive. The calculations were produced at the same time the patient was scanned and the results were immediately delivered to the doctors, ”said Kristopher Knott, author of the study. And, because poor blood flow is treatable, he adds, “these more accurate predictions lead to better patient care, as well as providing information on how the heart works.”