Google working with Aravind Eye Hospital to train its AI in diabetic retinopathy screening

Sriram Sharma June 12, 2017

Google partners with India’s Aravind Eye Hospital to use AI to detect diabetic retinopathy

India’s largest eye care provider, Aravind Eye Hospital, has been quietly working for over four years with Google on a project to use artificial intelligence (AI) in ophthalmology. Aravind Eye Hospital, which has branches across India, is headquartered in Madurai, Tamil Nadu.

Lily Peng, product manager at Google, who released a paper on the research a year ago, said at the 2017 WIRED Business Conference that Google had just finished a clinical study in India, and that work was underway to get the technology into routine use with patients. India, considered the diabetes capital of the world, has over 70 million diabetes patients who are at risk of blindness due to the disease.

India’s largest eye care provider, Aravind Eye Hospital, has been quietly working for over three to four years with Google on a project to use artificial intelligence in ophthalmology  

At the TensorFlow Dev Summit earlier this year, Peng had said that that Google’s machine learning algorithm was very close to an ophthalmologist in terms of performance.

“This particular project came in around four years ago; then we started looking at how we could develop it,” said Dr R Kim, chief medical officer at Aravind Eye Hospital. He said that the hospital has been working on an automated diabetic retinopathy screening since 2003. “It’s not a fully automated system, it doesn’t use AI. It’s a semi-automated technology for DR (diabetic retinopathy) screening,” he said.

According to Dr Kim, Google’s AI is far superior to anything he has ever seen in DR screening.

“With AI, we will be able to grade diabetic retinopathy to a certain level of identification, mainly for screening” — Dr R Kim, chief medical officer at Aravind Eye Hospital  

“With AI, we will be able to grade diabetic retinopathy to a certain level of identification, mainly for screening. This is basically a screening tool and we have validated it. We have done one validation, but we’re yet to publish anything on it. A research paper will be published soon on the partnership,” he added.

The learnings from this will be applied to Aravind’s eye care practice, he said. “We’re planning for the future, and looking at how to implement it in our routine work. Many people have worked with us earlier to develop an automated imaging solution, but we never made as much progress as with the Google partnership,” he said.

He, however, declined to quantify its accuracy in terms of a percentage value. “At this point, it’s going through the screening validation. I have a feeling it could turn out to be a good diagnostic tool soon. Depends on how the sensitivity increases — it could be much more sensitive than human graders,” Kim said.

“At this point, it’s going through the screening validation. I have a feeling it could turn out to be a good diagnostic tool soon” — Dr Kim  

Other preventable eye diseases can similarly be detected using early screening, he said. “Glaucoma and ARMD (age-related macular degeneration), is another condition where people are trying to look at AI for early identification. These conditions are treatable in early stages and not in late stages,” he said.

“I think the impact of AI is not going to be on the field of ophthalmology per se. It will be seen at a patient care level, a diabetes management level,” said Anand Sivaraman, director of Bengaluru-based Remidio, maker of a smartphone-based ophthalmic imaging system.

“I think the impact of AI is not going to be on the field of ophthalmology per se. It will be seen at a patient care level, a diabetes management level” — Anand Sivaraman, director, Remidio  

However, Sivaraman said the challenge lies in making sure there are no false negatives, which could deprive patients of treatment or consultation.

“We are looking at a future where such AI algorithms come on the smartphone. First, this will give people access to a low-cost ophthalmic system, and second, we can start thinking of a B2C model, where a consumer can take a selfie (of the eye) and start using such algorithms,” he said.

Google is not the only tech giant working on using AI to prevent blindness. In late 2016, Microsoft India had announced a collaboration with the Hyderabad-based L V Prasad Eye Institute and launched the Microsoft Intelligent Network for Eyecare (MINE) with the aim of providing eye care at scale using AI. Microsoft scientists also used search engine queries to detect users who are suffering from pancreatic cancer.

FactorDaily has written to Google India representatives for comment and will update the story when they respond.

Updated at 4.50pm on June 12, 2017, to add Anand Sivaraman’s quotes.



Newsletter