For quite some time, tech firms have attempted to leverage artificial intelligence (A.I.) and machine learning to solve health issues. Now they’ve hit a new milestone: A Google platform utilizes “deep learning” to predict cardiovascular health—by analyzing the patient’s eye.
“Using deep learning algorithms trained on data from 284,335 patients, we were able to predict CV risk factors from retinal images with surprisingly high accuracy for patients from two independent datasets of 12,026 and 999 patients,” Lily Peng, product manager of the Google Brain Team, wrote in a Feb. 19 blog posting about the new technology. “In addition to predicting the various risk factors (age, gender, smoking, blood pressure, etc) from retinal images, our algorithm was fairly accurate at predicting the risk of a [cardiovascular] event directly.”
Retinal scanning in conjunction with deep learning could not only predict whether the patient was a smoker (with 71 percent accuracy), but also the patient’s systolic blood pressure.
“Our approach uses deep learning to draw connections between changes in the human anatomy and disease, akin to how doctors learn to associate signs and symptoms with the diagnosis of a new disease,” Peng continued. “This could help scientists generate more targeted hypotheses and drive a wide range of future research.”
However, it’s still early days for the technology, which will undergo further refinement and modification in the months and years ahead. Google’s team wants to test the underlying algorithms on even larger datasets, in hope of improving the accuracy of results.
A year ago, Google tested a machine-learning algorithm capable of predicting 92.4 percent of tumors in a sampling of medical slides. Nor is Google the only tech firm using its technology to improve overall health: Apple has spent the past three years gradually expanding the capabilities of its ResearchKit, which allows doctors and researchers to crowdsource data from users on health issues such as asthma. IBM’s Watson A.I. platform, meanwhile, is available to oncologists, although some have questioned its abilities.
A.I. is a long way off from replacing human doctors. But within the next few years, the technology may come to the forefront as a valuable tool in hospitals everywhere.