The partnership was announced on Thursday at the Asia-Pacific AI for Social Good Summit held in Bangkok.
The initiative has already been rolled out across several clinics in India with help from Verily, formerly Google Life Sciences.
Now, after months of research in Thailand, Google is collaborating with Rajavithi Hospital to kick off clinical trials in select sites across the country.
Google hired 54 specialists to examine 130,000 images of the back of the eye, resulting in 880,000 diagnoses. Google then took those diagnoses and trained a machine to study a new image of the back of the eye and grade it on a five-point scale, hoping to help doctors better diagnose the disease.
Google is now working with Thai doctors to test if the algorithm works in Thailand.
Dr Paisan Ruamviboonsuk, assistant director of the Centre of Medical Excellence at Rajavithi Hospital, said there were more than five million patients with diabetes in Thailand, all of whom were at risk of diabetic eye disease, a condition that can lead to blindness but can be prevented by regular screenings.
“The results of the retrospective study show that the AI model was able to catch the eye disease better than trained human graders conducting similar screenings,” he said.
Lily Peng, Google Health’s product manager, added that the company’s goal was to make the benefits of AI available to everyone.
“The goal is to build this into a tool that can assist doctors in diagnosing the disease and help prevent more Thais from getting blind,” she said.
Meanwhile, Google and the United Nations Economic and Social Commission for Asia and the Pacific [ESCAP] used the occasion of on Thursday’s summit to sign an MoU for a partnership to share best practices and identify solutions to optimise the use of AI in Asia-Pacific.
For its part, Google is providing a grant to the Association of Pacific Rim Universities to create an Asia-Pacific AI for Social Good Research Network.
Published : December 13, 2018
By : The Nation