Google's ARDA eye AI crosses one million patients screened
TL;DR
- Google's ARDA diabetic retinopathy tool has now screened more than a million patients across hospital systems in India, Thailand and Australia.
- Google has previously set a target of six million AI-supported screenings across India and Thailand within ten years, up from 600,000 globally in October 2024.
- ARDA is licensed to three commercial partners (Forus Health, AuroLab, Perceptra) that operate the screening service at no cost to patients.
Over a million patients across clinics in India, Thailand and Australia have now been screened by a single Google deep-learning tool for diabetic retinopathy, according to a Comment published in Nature Medicine on 23 September 2026.
The piece is a scaling retrospective, not a new model paper. Its authors are Google's Richa Tiwari, Rajroshan Sawhney and Kasumi Widner, alongside Paisan Ruamviboonsuk of Rajavithi Hospital in Bangkok and Angus Turner of the Lions Eye Institute in Western Australia. Scaling the tool across three highly distinct settings, they write, "offers cross-cutting insights that may inform the expansion of healthcare artificial intelligence globally."
The three anchor sites are the Aravind Eye Care System in Madurai, Rajavithi Hospital in Bangkok, and Lions Outback Vision at the Lions Eye Institute in Western Australia. The tool, called Automated Retinal Disease Assessment or ARDA, is licensed to three commercial partners, Forus Health, AuroLab and Perceptra, which run the service at no cost to patients.
Google has previously set a target of six million AI-supported screenings across India and Thailand within ten years, up from more than 600,000 globally as of October 2024.
Two researchers in our Who's Who tracker shared the paper the day it published. The public abstract carries no per-country breakdown of the million-patient figure and no deployed-site accuracy numbers.
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