Google AI scans retina to predict heart disease risk

A new robot created by Google reads your retinas to assess your risk of heart disease.

The machine analyses scans of the back of your eye to accurately predict risk factors including age, blood pressure, and whether or not you smoke.

It then uses this data to analyse your risk of suffering a life-threatening cardiac event, such as a heart attack.

The device, powered by artificial intelligence, is far less invasive than a traditional diagnostic blood test but is just as accurate, according to Google.

A new robot created by Google analyses scans of the back of your eye to accurately predict risk factors including age, blood pressure, and whether or not you smoke. The AI uses standard retinal scans (left image) to analyse retinal blood vessels (green in left image)

Google worked with California-based health-tech subsidiary Verily to create the AI algorithm, which tracked blood vessels at the back of the eye.

Previous research has shown that the shape and size of retinal vessels reflect a person’s overall health, including their risk of heart disease and stroke.

Patients with a high blood pressure or who smoke are more likely to have weaker, thinner and damaged vessels than young and healthy individuals.

Using retinal images, Google says it was able to quantify this link and predict a patient’s risk of a heart attack or other major cardiovascular event.

The algorithm was able to tell whether or not a patient would suffer a cardiovascular event in the next five years with a 70 per cent accuracy rate, Google said.

The results were similar to those achieved via testing methods that require blood be drawn to measure a patient’s cholesterol, which are typically 72 per cent accurate.

Data collected by the AI can then be used to analyse your risk of suffering a life-threatening cardiac event, such as a heart attack. The device, powered by artificial intelligence, is far less invasive than a traditional diagnostic blood test but is just as accurate, according to Google

Data collected by the AI can then be used to analyse your risk of suffering a life-threatening cardiac event, such as a heart attack. The device, powered by artificial intelligence, is far less invasive than a traditional diagnostic blood test but is just as accurate, according to Google

Study coauthor Dr Michael McConnell, a medical researcher at Verily, said: ‘Cardiovascular disease is the leading cause of death globally.

‘There’s a strong body of research that helps us understand what puts people at risk: Daily behaviours including exercise and diet in combination with genetic factors, age, ethnicity, and biological sex all contribute.

‘However, we don’t precisely know in a particular individual how these factors add up, so in some patients we may perform sophisticated tests … to help better stratify an individual’s risk for having a cardiovascular event such as a heart attack or stroke. 

‘This paper demonstrates that deep learning applied to a retinal fundus image, a photograph that includes the blood vessels of the eye, can frequently predict these risk factors – from smoking status to blood pressure – as well as predict the occurrence of a future major cardiovascular event on par with current measures.’

HOW DOES ARTIFICIAL INTELLIGENCE LEARN?

AI systems rely on artificial neural networks (ANNs), which try to simulate the way the brain works in order to learn.

ANNs can be trained to recognise patterns in information – including speech, text data, or visual images – and are the basis for a large number of the developments in AI over recent years.

Conventional AI uses input to ‘teach’ an algorithm about a particular subject by feeding it massive amounts of information.   

AI systems rely on artificial neural networks (ANNs), which try to simulate the way the brain works in order to learn. ANNs can be trained to recognise patterns in information - including speech, text data, or visual images

AI systems rely on artificial neural networks (ANNs), which try to simulate the way the brain works in order to learn. ANNs can be trained to recognise patterns in information – including speech, text data, or visual images

Practical applications include Google’s language translation services, Facebook’s facial recognition software and Snapchat’s image altering live filters.

The process of inputting this data can be extremely time consuming, and is limited to one type of knowledge. 

A new breed of ANNs called Adversarial Neural Networks pits the wits of two AI bots against each other, which allows them to learn from each other. 

This approach is designed to speed up the process of learning, as well as refining the output created by AI systems. 

The researchers trained the algorithm using a dataset of nearly 300,000 patients, including eye scans and general medical data.

The robot used a deep learning analysis to spot patterns in this information, learning to link telltale signs in the eye scans with the metrics needed to predict disease risk, such as age or blood pressure.

Google said its AI is a proof of concept for now, with any practical implications in medicine ‘years away’, according to study coauthor Dr Lily Peng.

She told USA Today: ‘It’s not just when it’s going to be used, but how it’s going to be used.

‘I am very excited about what this means for discovery. ‘We hope researchers in other places will take what we have and build on it.’ 



Read more at DailyMail.co.uk