SAN DIEGO (KGTV) - A new "meta prediction model" developed by researchers at Scripps Research in La Jolla combines artificial intelligence, genomics, and electronic health records to better predict a person's risk of Coronary Artery Disease.
The model is twice as effective as traditional screening methods, according to the study's author, Dr. Ali Torkamani.
"If you're an older male, you're essentially going to get a high-risk prediction, almost no matter what. So the personalization is sort of lacking," said Torkamani.
Torkamani is a computational biologist at Scripps Research. He led the team that built the model, which provides individualized predictions of heart attack risk over 10 years.
Heart disease remains the leading cause of death in the U.S., which motivated Torkamani's focus on improving prediction methods.
"It's a very complex disease in terms of the factors that can lead to an eventual heart attack," said Torkamani.
The new model incorporates traditional risk factors like cholesterol, age, and blood pressure. It also analyzes years of electronic health records and genomic information to create a more personalized assessment.
The research team used data from the UK Biobank and the United States "All of Us" data set to help test their predictions. The AI component of the model identified additional risk factors not typically included in heart assessments, including poor mental health and lack of sleep.
"I think those are the two that really stood out as really not being used routinely," Torkamani said.
Now that the model exists, the next challenge is implementation in clinical settings and patient adoption.
"We're at the cusp," Torkamani says. "We have the pieces there, the pieces of technology to enable this, but they haven't really been all put together in one place... You have to actually get people to respond to this. I believe, as we personalize the risk more and more, that will get people to engage in using that information."
Torkamani adds the model becomes more accurate with more electronic medical data, giving patients an incentive to get regular medical care.
This story was reported by a journalist and has been converted to this platform with the assistance of AI. Our editorial team verifies all reporting on all platforms for fairness and accuracy.
