Subscribe for daily wellness inspiration

Like onlymyhealth on Facebook!

New machine learning developed by google researchers to detect diabetes retinopathy

By  ,  Onlymyhealth editorial team
Nov 30, 2016
4.8 / 5(4 Ratings)

And here comes the good news for patients suffering from diabetes facing related complications. It is a deep learning algorithm capable of interpreting signs of DR in retinal photographs. This new  potential technique is great for doctors as it will help them to  screen more patients, especially in underserved communities with restricted funds.

 

diabetes neuropathy

 

The techniques used previously are laser and anti-VEGF therapy factor in efficacy, safety, cost, and pragmatic issues. Google research team began studying whether machine learning could be used to screen for diabetic retinopathy (DR).

 

This study was published in the Journal of the American Medical Association. This paper demonstrates the technology benefits. It is estimated that the deep neural networks can be trained, using large data sets and without having to specify lesion-based features, to identify diabetic retinopathy or diabetic macular edema in retinal fundus images with high sensitivity and high specificity.

 

These results demonstrate that deep neural networks can be trained, using large data sets and without having to specify lesion-based features, to identify diabetic retinopathy or diabetic macular edema in retinal fundus images with high sensitivity and high specificity.


However, this  algorithm has been trained to identify only diabetic retinopathy and diabetic macular edema.

 

News Source: BetaNews

Image Source: Getty

Read More: Health News

Related Articles
4.8 / 5(4 Ratings)
Write a Review
Post a Comment
Disclaimer +
Though all possible measures have been taken to ensure accuracy, reliability, timeliness and authenticity of the information; Onlymyhealth assumes no liability for the same. Using any information of this website is at the viewers’ risk. Please be informed that we are not responsible for advice/tips given by any third party in form of comments on article pages . If you have or suspect having any medical condition, kindly contact your professional health care provider.