Moorfields Eye Hospital launches AI service for diabetic retinopathy in rural Australia

Moorfields Eye Hospital NHS Foundation Trust, in collaboration with the UCL Institute of Ophthalmology, Google, Topcon, and the INSIGHT Health Data Research Hub, is launching a mobile artificial intelligence (AI) service aimed at diagnosing diabetic retinopathy in remote regions of Australia. This initiative targets health inequities affecting rural and Indigenous Australians, particularly in the Pilbara region of Western Australia, which is home to a mere 45,000 residents spread over an area nearly twice the size of the UK. Access to quality health services in this region is highly restricted.

The service’s rollout follows a successful one-year pilot programme and a substantial grant of $5 million AUD from the West Australian government. The AI model, referred to as the RETFound AI foundation, is designed to provide “instant point-of-care diagnosis.” The next phase of implementation will involve refining the diabetic retinopathy algorithm with local data to enhance its accuracy specifically for Indigenous populations. There are future plans to expand the capabilities of the system, potentially allowing it to detect cardiovascular diseases by analysing eye scans.

Pearse Keane, a consultant ophthalmologist at Moorfields Eye Hospital, commented on the initiative’s significance, stating that there is “immense potential for medical AI to reduce health inequalities in low-resource settings.” He emphasised the AI’s role in overcoming challenges posed by “geographic isolation and limited access to health services” which are prevalent in the outback.

Mark Chia, an Australian ophthalmologist and member of the Lions Outback Vision team, also praised the collaboration, indicating that it has facilitated the combination of “world-leading data science and medical AI expertise with the experience and ingenuity of eyecare professionals in Australia.” This concerted effort aims to create an innovative approach to preserving vision and saving lives in underserved regions.

In addition to the deployment of the diabetic retinopathy AI model, Moorfields Eye Hospital has been active in advancing AI technologies for other eye health conditions. The institution has previously reported on the RETFound model’s development, detailing how it was trained on a dataset encompassing 1.6 million images. This foundational work also extends to identifying retinopathy of prematurity, a leading cause of childhood blindness, leveraging a sample comprising 7,414 images of at-risk newborns.

Furthermore, Moorfields has secured a £16.7 million EPR (Electronic Patient Record) contract with Meditech, scheduled to commence on 30 October 2024. This contract will last for an initial decade, with an option for a five-year extension, potentially extending the total duration to 15 years.

As these initiatives unfold, the integration of AI into ophthalmology continues to showcase its potential to bridge the gap in healthcare accessibility, particularly in geographic regions that often lack sufficient medical resources.

Source: Noah Wire Services