Fox News reports that AI can help doctors find heart problems recently overlooked because the heart looks normal in common tests. An EKG measures the electrical activity of a person’s heart.
Dr. Brick Lantz, vice president of advocacy and bioethics, of Christian Medical and Dental Association says researchers at University of California, Berkeley, let AI read hundreds of thousands of EKGs from Sweden.
“They had artificial intelligence examine these over 400,000 EKGs and see if there was any anomaly or difference from the general population that would be an identifying marker of someone that died from sudden cardiac death,” Lantz says.
The AI model was trained by looking at the Sweden EKGs then tested on patient data from the U.S. and Taiwan. AI pointed at one of the readings of the heart's electrical activity and found a link between one possible reading and a statistically significant number of cardiac deaths.
Lantz says doctors had never made the connection before.
“This is what's interesting. It's a new finding,” Lantz continues.
By comparing low-risk and high-risk EKGs, a visible feature was found in an area called aVL, which was a strong signal that predicted cardiac death.
Further testing will be needed, but it's likely that people with the marker will be fitted with pacemakers, thereby, perhaps, saving their lives. Lantz says that kind of analysis is the strong suit of artificial intelligence.
“Taking large volumes of data — and we're talking very large volumes of data — process that very rapidly and give us at least some information,” Lantz states.
AI has already made similar strides in cancer research, he says, and has the potential to revolutionize medicine as it’s practiced.
“I think that's going to be a useful tool for AI in the future, particularly in the field of data collection,” Lantz says.
