Google AI shows how eyes could reveal heart attack risk

The study, published in the journal Nature Biomedical Engineering, showed that deep learning applied to a retinal fundus image, a photograph that includes the blood vessels of the eye, can predict risk factors for heart diseases - from blood pressure to smoking status.

Indo-Asian News Service
San Francisco, Publish Date: Feb 20 2018 1:32PM | Updated Date: Feb 20 2018 1:32PM
Google AI shows how eyes could reveal heart attack riskGoogle AI shows how eyes could reveal heart attack risk

Knowing the risk of a heart attack or a stroke could soon become as simple as an eye test as researchers at Google and its health-tech arm Verily have found that Artificial Intelligence (AI) and Machine Learning (ML) could help identify signals of heart diseases through retinal images.

"AI offers us the potential for new, less invasive tests for heart health -- predicting cardiovascular results from retinal images with computer vision -- encouraging early results!," Google CEO Sundar Pichai said on Tuesday while referring the study in a tweet.

The study, published in the journal Nature Biomedical Engineering, showed that deep learning applied to a retinal fundus image, a photograph that includes the blood vessels of the eye, can predict risk factors for heart diseases -- from blood pressure to smoking status.

The algorithm that the researchers produced can even help predict the occurrence of a future major cardiovascular event on par with current measures, said Michael McConnell, Head of Cardiovascular Health Innovations at Verily in blog post. 

Cardiovascular disease is the leading cause of death globally and researchers know that lifestyle factors including exercise and diet in combination with genetic factors, age, ethnicity, and sex all contribute to it. 

However, they do not exactly know how these factors add up in a particular individual, and so in some patients it becomes essential to perform sophisticated tests, like coronary calcium CT scans, to help better stratify an individual's risk for having a heart attack or a stroke, and such other cardiovascular events. 

In this study, using deep learning algorithms trained on data from 284,335 patients, the researchers were able to predict cardiovascular risk factors from retinal images with surprisingly high accuracy for patients from two independent datasets of 12,026 and 999 patients. 

The algorithm could distinguish the retinal images of a smoker from that of a non-smoker 71 per cent of the time, the study found. 

"In addition, while doctors can typically distinguish between the retinal images of patients with severe high blood pressure and normal patients, our algorithm could go further to predict the systolic blood pressure within 11 mmHg on average for patients overall, including those with and without high blood pressure," study co-author Lily Peng, Product Manager, Google Brain Team, said.

"One of the exciting aspects of this study is the generation of 'attention maps' to show which aspects of the retina contributed most to the algorithm, thus providing a window into the 'black box' often associated with machine learning," McConnell, who is also a co-author of the study, said. 

This can give clinicians greater confidence in the algorithm, and potentially provide new insights into retinal features not previously associated with cardiovascular risk factors or future risk, McConnell said. 

The findings suggest that a simple retinal image could one day help understand the health of a patient's blood vessels, key to cardiovascular health.

"This is promising, but early research -- more work must be done to develop and validate these findings on larger patient cohorts before this can arrive in a clinical setting," McConnell added.

Latest News

  1. ED imposes Rs 14.4 lakh penalty on Hurriyat (G) chairman Geelani
  2. ED imposes Rs 14.4 lakh penalty on Hurriyat (G) chairman Geelani
  1. PSO of Anantnag municipality chairman injured in suspected militant attack
  2. PSO of Anantnag municipality chairman injured in suspected militant attack
  1. Shopian gunfight ends, two militant bodies recovered; searches on
  2. Shopian gunfight ends, two militant bodies recovered; searches on
  1. Elect your candidates on promise versus performance basis: Altaf Bukhari to workers
  2. Elect your candidates on promise versus performance basis: Altaf Bukhari to workers
  1. Er Rasheed to fight Lok Sabha polls from Baramulla constituency
  2. Er Rasheed to fight Lok Sabha polls from Baramulla constituency
  1. Policemen hit by stone during clashes in Shopian
  2. Policemen hit by stone during clashes in Shopian
  1. Governor’s grievance cell disposes of 55195 complaints
  2. Governor’s grievance cell disposes of 55195 complaints
  1. New Delhi won’t achieve anything by banning JeI, mass arrests: Sajad Lone
  2. New Delhi won’t achieve anything by banning JeI, mass arrests: Sajad Lone
  1. Friday prayers barred at Jamia Masjid in Srinagar, third time this month
  2. Friday prayers barred at Jamia Masjid in Srinagar, third time this month
  1. Govt should come out with 'more facts' on Balakot strikes, 'baffled' by PM's reaction: Pitroda
  2. Govt should come out with 'more facts' on Balakot strikes, 'baffled' by PM's reaction: Pitroda
  1. Farooq leads campaigning of alliance candidates in Jammu
  2. Farooq leads campaigning of alliance candidates in Jammu
  1. Hajin Gunfight: Police identifies two slain militants
  2. Hajin Gunfight: Police identifies two slain militants
  1. Pulwama resident allegedly affiliated with Jaish held in south Delhi
  2. Pulwama resident allegedly affiliated with Jaish held in south Delhi
  1. Ten injured in Shopian clashes, one suffers gunshot wound in neck
  2. Ten injured in Shopian clashes, one suffers gunshot wound in neck
  1. Pakistan might discover a very big reserve of oil and gas: PM Imran Khan
  2. Pakistan might discover a very big reserve of oil and gas: PM Imran Khan
  1. India boycotts Pak Day event as Kashmiri Hurriyat leaders invited
  2. India boycotts Pak Day event as Kashmiri Hurriyat leaders invited
  1. Jammu and Kashmir: Four militants, minor boy killed in gunfights in Shopian, Bandipora
  2. Jammu and Kashmir: Four militants, minor boy killed in gunfights in Shopian, Bandipora
  1. Clashes break out near gunfight site in Shopian
  2. Clashes break out near gunfight site in Shopian
  1. Militant killed in ongoing Shopian gunfight: Army
  2. Militant killed in ongoing Shopian gunfight: Army
  1. Two militants, minor boy killed in Hajin gunfight
  2. Two militants, minor boy killed in Hajin gunfight
  1. Mirwaiz placed under house arrest ahead of JRL protest call
  2. Mirwaiz placed under house arrest ahead of JRL protest call
  1. 44 dead after blast rocks industrial park in China
  2. 44 dead after blast rocks industrial park in China
  1. 93 people dead as ferry sinks in Iraq: Minister
  2. 93 people dead as ferry sinks in Iraq: Minister
  1. Sopore resident among two militants slain in Baramulla gunfight; thousands attend last rites
  2. Sopore resident among two militants slain in Baramulla gunfight; thousands attend last rites
  1. Gunfight underway in south Kashmir's Shopian
  2. Gunfight underway in south Kashmir's Shopian
  1. Sopore gunfight: Fresh exchange of fire after nightlong lull
  2. Sopore gunfight: Fresh exchange of fire after nightlong lull
x
This site uses cookies to deliver our services and to show you relevant news and ads. By using our site, you acknowledge that you have read and understand our Cookie Policy, Privacy Policy, and our Terms of Service.That's Fine