Women fight against heart attacks with AI

Women fight against heart attacks with AI


Women are more likely to die from a heart attack than men and its symptoms are often misinterpreted. Researchers from Zurich, London and Graz have developed a new risk assessment with the help of artificial intelligence (AI).

Unlike men, who usually feel chest pain that radiates to the left arm, a heart attack in women often causes pain in the abdomen and radiates to the back, or nausea and vomiting. Mistaking these traits can have disastrous consequences.

Data from over 420,000 patients

In their scientific study, researchers from Zurich, London and Graz analyzed data from more than 420,000 patients with the most common type of heart attack from across Europe.

“The study shows, among other things, that the established risk models that govern current patient management are less accurate in women and encourage the undertaking of female patients,” said Austrian physician Florian Wenzel, first author, who works at the Center for Molecular Medicine. University of Zurich. “Using machine learning and the largest data set in Europe, we have developed a new risk score that accounts for gender differences in risk profile and improves the prediction of mortality in women and men,” he explained.

beginning of new era

Female patients have a higher mortality rate than male patients, ignoring the age difference at admission and existing risk factors such as hypertension and diabetes. “If these differences are taken into account statistically, however, women and men have similar mortality rates,” explained Thomas Luscher, a colleague of Wenzel’s institute.

See also  Google: No more passwords for activity page

“Our study marks the beginning of an era of artificial intelligence in the treatment of heart attack patients,” Wenzel said. Modern computer algorithms can learn from large data sets and make accurate predictions about the prognosis of individual patients. And these, in turn, are the key to personalized healing. Serena Herzog from the Institute for Medical Informatics, Statistics and Documentation at the Medical University of Graz was also involved in the study, which was published in the renowned specialist journal “The Lancet.”


Please enter your comment!
Please enter your name here