Google AI can predict heart disease risk from eye images

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

PTI
Washington, Publish Date: Feb 21 2018 12:44AM | Updated Date: Feb 21 2018 12:44AM
Google AI can predict heart disease risk from eye imagesRepresentational Pic

Researchers at Google have developed a new artificial intelligence system that can accurately predict the risk of heart diseases by scanning images of people's retina.

The discovery may point to more ways to diagnose health issues from retinal images, researchers said.

"Using deep learning algorithms trained on data from 284,335 patients, we 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," Lily Peng from the Google Brain Team wrote in a blog post.

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

While doctors can typically distinguish between the retinal images of patients with severe high blood pressure and normal patients, the algorithm could go further to predict the systolic blood pressure within 11 milimetre Hg on average for patients overall, including those with and without high blood pressure.

The algorithm was fairly accurate at predicting the risk of a cardiovascular event directly, said Peng.

"Our algorithm used the entire image to quantify the association between the image and the risk of heart attack or stroke," she said.

Given the retinal image of one patient who later experienced a major cardiovascular event (such as a heart attack) and the image of another patient who did not, the algorithm could pick out the heart patient 70 per cent of the time.

This performance approaches the accuracy of other cardiovascular risk calculators that require a blood draw to measure cholesterol.

"We opened the 'black box' by using attention techniques to look at how the algorithm was making its prediction. These techniques allow us to generate a heatmap that shows which pixels were the most important for a predicting a specific cardiovascular risk factor," said Peng.

For example, the algorithm paid more attention to blood vessels for making predictions about blood pressure, as shown in the image above. Explaining how the algorithm is making its prediction gives doctor more confidence in the algorithm itself.

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, colours, values and shapes that are present in real images.

"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," Peng said.

This could help scientists generate more targeted hypotheses and drive a wide range of future research. PTI 

Latest News

  1. Sopore gunfight: Police identifies one slain militant as Tujar Sharief resident
  2. Sopore gunfight: Police identifies one slain militant as Tujar Sharief resident
  1. CRPF man goes missing from Srinagar camp
  2. CRPF man goes missing from Srinagar camp
  1. Samsung unveils its first triple camera Smart phone Galaxy A7
  2. Samsung unveils its first triple camera Smart phone Galaxy A7
  1. Congress throwing muck, conspiring to topple me: Modi
  2. Congress throwing muck, conspiring to topple me: Modi
  1. Landslide blocks circular road in Jammu, clearance operation begins
  2. Landslide blocks circular road in Jammu, clearance operation begins
  1. Resignation of SPOs 'negligible', restricted to certain areas of south Kashmir: Govt
  2. Resignation of SPOs 'negligible', restricted to certain areas of south Kashmir: Govt
  1. Governor's rule not about deciding Article 35-A: Jammu and Kashmir govt
  2. Governor's rule not about deciding Article 35-A: Jammu and Kashmir govt
  1. Two militants killed in Sopore gunfight: Army
  2. Two militants killed in Sopore gunfight: Army
  1. Junaid Mattu resigns from NC, to contest Kashmir municipal elections
  2. Junaid Mattu resigns from NC, to contest Kashmir municipal elections
  1. Jammu and Kashmir govt asks forest dept to setup VLCs for protection of forests
  2. Jammu and Kashmir govt asks forest dept to setup VLCs for protection of forests
  1. Indo-Pak trade potential stands at USD 37 billion: World Bank
  2. Indo-Pak trade potential stands at USD 37 billion: World Bank
  1. Body of 30-yr-old man found in Srinagar
  2. Body of 30-yr-old man found in Srinagar
  1. VDC member, wife injured when rifle went off in Reasi
  2. VDC member, wife injured when rifle went off in Reasi
  1. World Bank will offer new initiative on Indus Treaty dispute: Pakistan
  2. World Bank will offer new initiative on Indus Treaty dispute: Pakistan
  1. Panchayat ghar set on fire in Bandipora, 8th such incident in valley
  2. Panchayat ghar set on fire in Bandipora, 8th such incident in valley
  1. SC extends deadline till Nov 15 for UP govt to submit vision document on protection of Taj Mahal
  2. SC extends deadline till Nov 15 for UP govt to submit vision document on protection of Taj Mahal
  1. Jammu and Kashmir: 11 drivers booked for overloading in Rajouri
  2. Jammu and Kashmir: 11 drivers booked for overloading in Rajouri
  1. Fresh exchange of fire at Sopore gunfight site
  2. Fresh exchange of fire at Sopore gunfight site
  1. Body of woman severed at head found in UP's Ballia
  2. Body of woman severed at head found in UP's Ballia
  1. Video | Govt announces one-month additional salary for employees on poll duty
  2. Video | Govt announces one-month additional salary for employees on poll duty
  1. Three Kashmir families claim kin among five militants killed in Bandipora gunfight
  2. Three Kashmir families claim kin among five militants killed in Bandipora gunfight
  1. Ganderbal family claims kin killed in Bandipora gunfight
  2. Ganderbal family claims kin killed in Bandipora gunfight
  1. Two south Kashmir families reach Bandipora, claim sons killed in Sumlar gunfight
  2. Two south Kashmir families reach Bandipora, claim sons killed in Sumlar gunfight
  1. Patna doctors' strike enters 2nd day, 12 patients dead
  2. Patna doctors' strike enters 2nd day, 12 patients dead
  1. Army pays tribute to soldier killed in Tangdhar gunfight
  2. Army pays tribute to soldier killed in Tangdhar gunfight
  1. Adobe, Microsoft and SAP announce Open Data Initiative
  2. Adobe, Microsoft and SAP announce Open Data Initiative
  1. Publicise criminal records of candidates: SC to political parties
  2. Publicise criminal records of candidates: SC to political parties
  1. JKBOSE to conduct class 11th annual examination from current session
  2. JKBOSE to conduct class 11th annual examination from current session
  1. Foreign tourist couple trapped in snow rescued in Kargil
  2. Foreign tourist couple trapped in snow rescued in Kargil
  1. Values of Nelson Mandela needed more than ever: Swaraj
  2. Values of Nelson Mandela needed more than ever: Swaraj
  1. Man dead, another injured in Kangan road mishap
  2. Man dead, another injured in Kangan road mishap
  1. Sopore shooting: searches continue in Tujjar; educational institutions closed, Internet suspended
  2. Sopore shooting: searches continue in Tujjar; educational institutions closed, Internet suspended
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