Artificial Intelligence helps predict risky patients after surgery

Inteligencia Artificial ajuda a prever doentes de risco após cirurgia

Researchers from the Faculty of Medicine of the University of Porto (FMUP) have used Artificial Intelligence to predict which cardiovascular patients are most at risk of developing neurological problems after carotid artery surgery, it was revealed today.

In a statement, FMUP explains that in the study, published in the Journal of Clinical Medicine, the researchers applied an algorithm based on Artificial Intelligence to "determine the characteristics of patients who may have intraoperative neurological problems" if they undergo surgery, clinically known as carotid endarterectomy.

This surgery aims to prevent patients with narrowed or blocked internal carotid arteries - which carry oxygen and blood to the brain - from having strokes.

"Although it is widely used, this treatment can have adverse neurological effects, such as confusion and changes in speech, motor function and consciousness," says the FMUP, noting that these effects can, in the short and long term, lead to risks of cardiovascular events.

In the course of the study, the researchers discovered that obesity is the "main independent risk factor" for the occurrence of persistent neurological deficits after this type of surgery, and developed a "therapeutic decision tree based on an artificial intelligence algorithm".

Quoted in the press release, the study's authors explain that the new algorithm "makes it possible to advise against surgery in some cases and to optimize a suitable alternative treatment plan".

"The clinical usefulness of this algorithm could be extensive, especially in planning and optimizing the quality of intervention in these patients, and it will be subject to international validation," they add.

Carotid endarterectomy is the recommended surgery for treating patients with and without symptoms.

It is estimated that 58 million people in the world, mainly men, have a carotid stenosis that requires this intervention.

The research was carried out by Juliana Pereira Macedo, Luís Duarte Gamas, António Pereira Neves, Joana Mourão, José Paulo Andrade and João Rocha Neves.

Share this article

Leave a Reply

Your email address will not be published.