Google algorithm predicts cardiovascular risk from eye images

February 20, 2018 // By Rich Pell
Google algorithm predicts cardiovascular risk from eye images
Google (Mountain View, CA) has announced that it has shown that using deep learning techniques with medical images of the eye can - in addition to detecting eye disease - accurately predict indicators of cardiovascular health.

It is already known that deep learning techniques can help increase the accuracy of diagnoses for medical imaging, especially for diabetic eye disease. This latest discovery, says Google, suggests that it may be possible to discover even more ways to diagnose health issues from retinal images.

For this study, the company used deep learning algorithms trained on data from 284,335 patients, following which, say the Google researchers, they were able to predict cardiovascular risk factors from retinal images "with surprisingly high accuracy" for patients from two independent datasets. For example, their algorithm was able to distinguish the retinal images of a smoker from that of a non-smoker 71% of the time, as well as predict the systolic blood pressure within 11 mmHg on average for patients overall.

In addition to predicting the various risk factors - i.e., age, gender, smoking, blood pressure, etc. - from retinal images, the algorithm was also fairly accurate at predicting the risk of a cardiovascular event directly. This performance, say the researchers, approaches the accuracy of other cardiovascular risk calculators that require a blood draw to measure cholesterol.

"At the broadest level, we are excited about this work because it may represent a new method of scientific discovery," says Lily Peng MD PhD, Product Manager, Google Brain Team. "Traditionally, medical discoveries are often made through a sophisticated form of guess and test — making hypotheses from observations and then designing and running experiments to test the hypotheses."

"However, with medical images, observing and quantifying associations can be difficult because of the wide variety of features, patterns, colors, values and shapes that are present in real images," says Peng. "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


Vous êtes certain ?

Si vous désactivez les cookies, vous ne pouvez plus naviguer sur le site.

Vous allez être rediriger vers Google.