This AI could predict a patient’s risk of having a heart attack within five years

American scientists have developed a new artificial intelligence that would be able to predict a patient’s risk of having a heart attack over the next five years. To achieve this feat, they trained an algorithm capable of measuring coronary artery deposits using CTA (computed tomography angiography, also called CT angiography) data from 921 patients, explains Slate​ relaying an article from New Atlas.

“When the analysis of CTA images takes a specialist thirty minutes, this program could do it in five or six seconds,” explained Damini Dey, author of this new study. Artificial intelligence could classify patients into two categories: on the one hand those with a high risk of having a heart attack in the next five years and on the other those with a low risk.

Several hundred tests carried out

For the moment, the algorithm has been tested on several hundred patients and has given results almost similar to the reports of specialists. However, for this innovation to be ready, a larger study is needed and more resources are needed because CTA is a very expensive method.

Despite these complications to come, Damini Dey assures that with this new artificial intelligence, scientists will “soon be able to predict if and when a person is likely to have a heart attack”.