Google's diabetic retinopathy AI passes one million patients

More than one million patients have now been screened for diabetic retinopathy using a Google-developed deep learning tool deployed across three health systems in India, Thailand and Australia, according to a Comment published in Nature Medicine on 23 September 2026.

The paper, led by Richa Tiwari of Google with clinical co-authors from the Aravind Eye Care System in Madurai, Rajavithi Hospital in Bangkok and Lions Outback Vision at the Lions Eye Institute in Western Australia, argues that the three deployments offer "cross-cutting insights that may inform the expansion of healthcare artificial intelligence globally."

Rather than run the service itself, Google licenses the model to three commercial partners: Forus Health, AuroLab and Perceptra. Those partners handle in-country deployment and pursue local regulatory approvals, and are targeting a combined "6 million AI-supported screenings to resource-constrained communities at no cost to patients" over the next decade. In Thailand, Google's own account says the work is being run alongside the Ministry of Public Health's Department of Medical Services on implementation research and cost-effectiveness analysis.

The Nature Medicine piece sits behind a paywall; the abstract publishes no per-site accuracy figures and no country-level breakdown of the screening total.