Google ARDA retinal AI tops one million patients screened
TL;DR
- Google's diabetic retinopathy screening tool ARDA has crossed one million patients across sites in India, Thailand and Australia.
- The Nature Medicine Comment, published 23 September 2026, is framed as a deployment-lessons report rather than a new accuracy paper.
- Google licenses ARDA to Forus Health, AuroLab and Perceptra, targeting six million AI-assisted screenings in India and Thailand over ten years.
Google's diabetic retinopathy screening tool has crossed one million patients, and the people who ran the deployments are now writing up what it took to get there. The venue is a Nature Medicine Comment published 23 September 2026, anchored in three unlike-each-other sites: the Aravind Eye Care System in Madurai, Rajavithi Hospital in Bangkok, and Lions Outback Vision at the Lions Eye Institute in Western Australia.
The authors frame the piece as "scaling a successful deep learning tool across three highly distinct settings – in India, Thailand and Australia," aimed at "cross-cutting insights that may inform the expansion of healthcare artificial intelligence globally."
The tool is ARDA, Google's Automated Retinal Disease Assessment. Google does not run the clinics. It licenses ARDA to Forus Health, AuroLab and Perceptra, who deliver the service at no cost to patients, with a stated target of six million AI-assisted screenings across India and Thailand within ten years.
The byline carries the point. Corresponding author Richa Tiwari of Google writes alongside Sunny Virmani and Yossi Matias of Google and three clinician co-authors from the deployment sites: Paisan Ruamviboonsuk at Rajavithi Hospital, Kim Ramasamy at Aravind Eye Care, and Angus Turner at the Lions Eye Institute. This is a deployment-lessons report rather than a new model paper. Two researchers we track in our Who's Who posted the paper's source link on Bluesky.
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