Google researchers use AI to analyse eye images for heart attack

20 February 2018 (Last Updated November 22nd, 2018 11:31)

Health researchers at Google have used artificial intelligence (AI) to predict cardiovascular (CV) risk factors through analysis of retinal images.

Google researchers use AI to analyse eye images for heart attack
Google’s AI accurately predicts CV risk by assessing retinal images. Credit: TO17.

Health researchers at Google have used artificial intelligence (AI) to predict cardiovascular (CV) risk factors through analysis of retinal images.

The new research is based on the team’s previous findings of using deep learning techniques for accurate diagnosis of diabetic eye disease using medical imaging.

During the latest study, it was found that deep learning algorithms, which were trained on data from 284,335 patients, could predict CV risk factors from retinal images with high accuracy for individuals from two separate datasets of 12,026 and 999 patients.

They further observed that the algorithm could detect patients who would develop a major CV event in the future in 70% of the cases. This accuracy is said to be comparable to other CV risk calculators that require a blood test for cholesterol.

In a statement, the team said: “We show that in addition to detecting eye disease, images of the eye can very accurately predict other indicators of CV health.

“This discovery is particularly exciting because it suggests we might discover even more ways to diagnose health issues from retinal images.”

“This discovery is particularly exciting because it suggests we might discover even more ways to diagnose health issues from retinal images.”

The researchers additionally explored the way the algorithm made CV risk predictions to ensure that the approach is reliable and can aid in further research.

Researchers added: “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.

“This could help scientists generate more targeted hypotheses and drive a wide range of future research.”

They believe that further studies are required using larger and more comprehensive datasets for validating the CV risk prediction capability of the algorithm.