Particularly hopeful are the findings of new preliminary research to be presented at the American Heart Association’s Resuscitation Science Symposium 2023. Artificial intelligence (AI) could help us predict sudden cardiac death and perhaps even reduce a person’s risk of future death , “cutting edge” development for global health.
“Sudden cardiac death accounts for 10 to 20% of all deaths. It is difficult to predict and conventional approaches fail to identify high-risk individuals, particularly at the individual level,” says Dr. Xavier Jouven, lead author of the study and professor of cardiology and epidemiology at the Paris Cardiovascular Research Center.
“We proposed a new approach that is not limited to common cardiovascular risk factors, but includes all medical information available in electronic health records.”
What is the contribution of artificial intelligence?
The research team analyzed AI-assisted medical information from registries and databases in Paris, France and Seattle for 25,000 people who had died of sudden cardiac arrest and 70,000 people from the general population, with data from the two groups matched with age, gender and region of residence.
The data, which represented more than 1 million hospital diagnoses and 10 million drug prescriptions, was gathered from medical records up to ten years before each death.
Using artificial intelligence to analyze the data, the researchers created nearly 25,000 equations with personalized health factors that were used to identify people who were at very high risk of sudden cardiac death.
In addition, they developed a personalized risk profile for each of the subjects in the study.
What was the level of accuracy in the prediction?
Individualized risk equations included a person’s medical factors, such as treatment for high blood pressure and a history of heart disease, as well as mental or behavioral disorders, including alcohol abuse.
Through the analysis, the researchers identified those factors that are most likely to reduce or increase the risk of sudden cardiac death in a certain percentage and time frame, for example, an 89% risk of sudden cardiac death within three months.
Through AI analysis they were able to identify people who had a greater than 90% risk of dying suddenly and accounted for more than ¼ of all sudden cardiac death cases.
What are the risk factors?
“We have been working for almost 30 years in the field of predicting sudden cardiac death, however, we did not expect to reach such a high level of accuracy.
We also discovered that individualized risk factors are very different between participants and often come from different medical fields (a combination of neurological, psychiatric, metabolic and cardiovascular data), a picture that is difficult to synthesize or harness without the input of artificial intelligence.” says the head of the study.
“While doctors have effective approaches, such as reducing risk factors or specific drugs, the use of artificial intelligence can make a difference in early detection of risk,” he adds.
“We hope that with a personalized list of risk factors, patients in the future will be able to better work with their doctor to reduce these risk factors and ultimately avoid sudden cardiac death,” the researchers conclude.