Google has developed an AI that detects the risks of heart disease by scanning the eyes

Google has developed an AI that detects the risks of heart disease by scanning the eyes

Google has developed an artificial intelligence capable of detecting risks of cardiovascular disease by scanning the eye of a patient, noted the American website The Verge.

The team of researchers from Verily, the life sciences research subsidiary of the American giant, published on February 19, 2018 the results of its study entitled “predictions of cardiovascular risk factors from photos of retinas through deep learning”.

To train their detection algorithm, the researchers “fed” it with photos of the retina of more than 284,000 people. Thanks to these images, they detected recurring patterns that allowed them to assimilate certain characteristics to personal patient information, such as:

  • age,
  • kind,
  • whether they smoke or not,
  • voltage,
  • if they have ever had heart attacks.

This information is usually used to determine the risks of cardiovascular accidents in a patient.

Doctors already observe the retina to deduce this kind of information, but they usually have to supplement it with blood tests. According to Professor Alun Hughes, interviewed by The Verge, AI will be able to significantly speed up these analyzes, but that it will still have to be trained to reduce the margin of error.

The researchers who worked on this experiment are Ryan Poplin, Avinash V. Varadarajan, Katy Blumer1, Yun Liu1, Michael V. McConnell, Greg S. Corrado1, Lily Peng and Dale R. Webster.

Receive our latest news

Every day, the main Business Insider news

Go to our cases Get a free quote