Recently, American researchers detailed a new artificial intelligence of their creation. Its objective is to detect cardiovascular accidents five years in advance. This is an innovation at a time when cardiovascular diseases and cerebrovascular accidents (CVA) represent the leading cause of death in the world.
According to a WHO publication, nearly 18 million people die each year from cardiovascular disease (i.e. 31% of global mortality total), more than any other reason. Today, research is continuing to try to better predict the risks. In February 2022, we reported on research at the University of Leeds (UK) relating to an AI analyzing retinal scans to assess the risk of heart attack.
In a study to be published in the journal The Lancet Digital Health, scientists from the Cedars-Sinai Biomedical Imaging Research Institute (BIRI) in Los Angeles (USA) have also found point their own AI. However, it does not analyze retinal scans, but the results from a CT angiography. This device allows a medical imaging examination showing blood vessels that are not usually visible by means of standard x-rays.
According to researchers, their artificial intelligence is able to predict the risk of heart attack with a lead of five years and with a very strong correlation (r=0.922, p<0.0001). The training of the AI took place using a database containing CT angiography images of 921 patients. Thus, the AI has learned to recognize atheroma plaques , a deposit of fat, blood, fibrous or calcareous tissue on the inside of an artery or blood vessel. However, the internal diameter can be reduced and thus hinder blood circulation , which increases the risk of heart attack.
Next, the researchers defined a threshold of atherosclerotic plaques with the aim of separating the patients into two distinct groups. These groups contained either people with a low risk of heart attack in the next five years, or a high risk. These results have been validated on 1,611 patients who participated in the Scot-Heart study (2018). These scientists said that further research would be needed to confirm the viability of their innovation . On the other hand, they are convinced that their system coupled with an angioscanner already makes it possible to predict the risks of infarction with a minimal error rate.
