How does it work? Through the eyes.
Scientists studied data from 284,335 patients and found the "deep-learning" AI algorithm could identify risk factors based on age, blood pressure, gender, smoking status and others just by scanning the individuals' retinas.
Scroll down to continue reading
More news from KIRO 7
- Dogs separated at adoption randomly find each other at Pike Place Market
- Hero dog takes bullet for teenager in Des Moines home invasion
- One Washington school district already has staff armed with guns
- Deputies: Woman lured from Arkansas to Washington located safe; 2 suspects in custody
- Washington lawmakers exempt themselves from Public Records Act
"The rear interior wall of the eye (the fundus) is chock-full of blood vessels that reflect the body's overall health," the Verge reported. "By studying their appearance with camera and microscope, doctors can infer things like an individual's blood pressure, age, and whether or not they smoke, which are all important predictors of cardiovascular health."
Google's AI was able to differentiate patients who suffered a major cardiac event in the following five years and those who didn't with a 70 percent accuracy.
"While doctors can typically distinguish between the retinal images of patients with severe high blood pressure and normal patients, our algorithm could go further to predict the systolic blood pressure within 11 mmHg on average for patients overall, including those with and without high blood pressure," lead author Lily Peng wrote in a Google blog.
Medical algorithms are traditionally created to redesign experiments to test hypotheses made from observations. But this algorithm found new ways to analyze existing medical data.
"With enough data, it's hoped that artificial intelligence can then create entirely new medical insight without human direction," the Verge reported.
This technology would make it more efficient for doctors to analyze cardiac risk without a blood test, which typically predicts events with 72 percent accuracy. But more tests are necessary before the AI can be used in a clinical setting.
“We look forward to developing and testing our algorithm on larger and more comprehensive datasets. To make this useful for patients, we will be seeking to understand the effects of interventions such as lifestyle changes or medications on our risk predictions and we will be generating new hypotheses and theories to test,” Peng wrote.
The research was published Monday in the journal "Biomedical Engineering."
© 2020 Cox Media Group