Google's retinopathy AI screens 1M patients in three countries
TL;DR
- Over one million patients have been screened using Google's diabetic retinopathy AI across sites in India, Thailand and Australia.
- The three anchor deployments are Aravind Eye Care in Madurai, Rajavithi Hospital in Bangkok, and Lions Outback Vision in Western Australia.
- Google licenses the model to Forus Health, AuroLab and Perceptra, with partners targeting six million screenings over the next decade.
Over one million patients have been screened using Google's diabetic retinopathy detection model across sites in India, Thailand and Australia, according to a Comment published in Nature Medicine on 23 September 2026 by authors from Google and its three anchor deployment partners.
The three deployment sites are Aravind Eye Care System in Madurai, Rajavithi Hospital in Bangkok, and Lions Outback Vision at the Lions Eye Institute in Western Australia. Google licenses the underlying model, known as ARDA, to three commercial partners, Forus Health, AuroLab and Perceptra, which run the screening programs in-country. The authors, led by Richa Tiwari, Rajroshan Sawhney and Kasumi Widner of Google alongside Paisan Ruamviboonsuk of Rajavithi and Angus Turner of Lions Eye, describe the deployment as one that "offers cross-cutting insights that may inform the expansion of healthcare artificial intelligence globally."
The commercial framing sits on Google's own blog post: partner organizations are targeting six million AI-supported screenings across India and Thailand over the next decade, working with providers including Forus Health, AuroLab and Perceptra plus the Thai Ministry of Public Health's Department of Medical Services. The same post notes that "Nearly half of people with diabetes will develop diabetic retinopathy," and cites 537 million adults living with diabetes worldwide, with 227 million of them in the Asia Pacific region.
The paper is a lessons-learned Comment rather than a fresh clinical trial. The published preview does not carry per-site sensitivity or specificity numbers for Aravind, Rajavithi or Lions Outback, and the commercial terms of the licensing deals with Forus, AuroLab and Perceptra are not disclosed.
Shared on Bluesky by 2 AI experts
-
Over a million patients were screened for diabetic eye disease as Google scaled its clinical tool across India, Thailand, and Australia. One of the biggest challenges wasn't the model itself; it was sorting out camera o…
View on Bluesky →
Originally reported by nature.com
Read the original article →Original headline: Practical lessons in the global scaling of clinical AI: from one hospital to over a million patients screened - Nature Medicine