Google’s AI retina-scanning tool can predict cardiovascular risk

The technology could reveal heart health conditions after comparing eye scans with a matrix for cardiovascular risks. The algorithm has proved correct in 70% of the cases it has been tested so far.

eye scanAn algorithm developed by Google is indeed set to prove that your eyes are the window to your soul. (Source: Freepik)

Listen to this article
Your browser does not support the audio element.

Can a look inside your eyes reveal the condition of your heart? An algorithm developed by Google is indeed set to prove that your eyes are the window to your soul. A deep learning model based on images of the retinas of nearly 285,000 people can predict risk factors such as age, gender, blood pressure and smoking habits, some of the leading predictors of heart disease.

The technology could reveal heart health conditions after comparing eye scans with a matrix for cardiovascular risks. The algorithm has proved correct in 70% of the cases it has been tested so far. Google’s new artificial intelligence technique could eliminate the need for a long series of conventional tests and scans usually required to detect cardiovascular risks.

So how has the quest to develop a device to detect diabetic retinopathy helped determine the risk of cardiovascular disease? Google already had access to retina images from a US and UK dataset that it used to train its deep learning model. He studied anatomical features of the eye such as the optic disc or blood vessel to generate predictions about age, gender, smoking habits and blood pressure. This model was validated using an additional 13,000 records from these datasets.

Last year, the University of Leeds conducted a similar study into the AI ​​deep learning process, in which it analyzed the retinal and heart scans of more than 5,000 people. The AI ​​system identified associations between pathology in the retina and changes in the patient’s heart. Once it learned the image patterns, the AI ​​system could estimate the size and pumping efficiency of the heart’s left ventricle from scans of the retina alone. An enlarged ventricle is linked to an increased risk of heart disease. Combined with basic patient demographics, their age and gender, the AI ​​system could predict their risk of a heart attack over the next 12 months.

How did the idea come about? Google was working with its partner hospitals in India to develop an AI tool that could screen for diabetic retinopathy without trained personnel. Diabetic retinopathy is a complication in which uncontrolled blood sugar for several years damages blood vessels in the retina, sometimes causing blindness. The problem is, people don’t perceive any change in their vision until it’s very late. Laser treatment can halt the progress of the disease but it cannot restore vision, said Dr R Kim of Aravind Eye Hospital, one of Google’s partners. The simple solution was to screen those who came to diabetes clinics. We diagnosed diabetic retinopathy remotely, with diabetes clinics sending us the photographs online. What Google’s AI did was ensure that an eye doctor didn’t have to go through every single photograph to check the condition, said Dr. Kim. This AI-enabled device has already received CE certification for use in Europe and has also been used by Aravind Hospital to collect cases in remote areas. The hospital alone has so far used it to screen 200,000 people.

Dr. Rajiv Raman of Sankara Netralaya, who also collaborated on the initial study in 2016, is now investigating whether these images can predict the onset of diabetic retinopathy. The challenge is to see if it can work with real-world data. The company has shown that the retinal fundus images we take to check for diabetes retinopathy can be used to check for cardiovascular disease risk. Now, we have to prove with certainty that it happens, he said.

We have to collect data from thousands of patients. Along with images of the retina, we need to collect data on parameters such as blood pressure, blood glucose levels and cholesterol levels, said Dr. Kim. Patients would then need to be followed up for two years to see if the algorithm can accurately predict cardiovascular disease risk.

First published on: 2023-07-02 at 07:34 CEST



#Googles #retinascanning #tool #predict #cardiovascular #risk
Image Source : indianexpress.com

Leave a Comment