AI improves therapy for women with heart attacks


Women die more often from heart attacks than men. Reasons are differences in age and in concomitant diseases, which complicate risk assessment. The symptoms also differ and are often misinterpreted in women. Using artificial intelligence (AI), researchers in Zurich, London and Graz have developed a new risk assessment that improves care for women with heart attacks, the University of Zurich announced.

Unlike men, who usually feel a painful pressure in the chest extending to the left arm, a heart attack in women often causes abdominal pain and radiating to the back, or nausea and vomiting. Misjudging these symptoms can have disastrous consequences. In their scientific study, the researchers analyzed data from more than 420,000 patients from across Europe with the most common type of heart attack.

New risk assessment takes gender into account
“The study shows, among other things, that established risk models that control current patient management are less accurate in women and encourage undertreatment of female patients,” emphasized Austrian physician Florian Wenzl, lead author, who works at the Center for Molecular Medicine at the University. from Zurich. “Using machine learning and the largest datasets in Europe, we have developed a new risk score that takes into account gender differences in the risk profile and improves the prediction of mortality in women and men,” he reported.

What role do statistical differences play?
Female patients have a higher mortality rate than male patients, ignoring differences in age at admission and existing risk factors such as hypertension and diabetes. “However, if these differences are statistically taken into account, women and men have similar mortality rates,” explains Wenzl’s institute colleague Thomas Lüscher.

“Our study ushers in the era of artificial intelligence in treating heart attack patients,” Wenzl says. Modern computer algorithms can learn from large data sets and make accurate predictions about the prognosis of individual patients. And these, in turn, are key to individualized treatments. Sereina Herzog of the Institute of Medical Informatics, Statistics and Documentation of the Medical University of Graz was also involved in the study, which was published in the renowned journal “The Lancet”.

Source: Krone


Please enter your comment!
Please enter your name here

Share post:



More like this

Motive still unclear – murder in Vienna hotel: suspect (33) arrested

After the horrific murder in a Viennese hotel last...

Trapped in a shaft – elevator rope torn: man falls into the depths

On Wednesday afternoon, a man (56) in Gumpoldskirchen (Lower...

Horror accident in Germany – father and son burned during illegal car race

Two occupants of a sports car were killed in...