Get all your news in one place.
100’s of premium titles.
One app.
Start reading
AAP
AAP
Health
Michael Ramsey

Artificial intelligence targets Indigenous vision loss

AI has shown promise in detecting early signs of diabetes-induced blindness in Indigenous patients. (Dave Hunt/AAP PHOTOS) (AAP)

Artificial intelligence could be used to help close the vision gap for Indigenous people who are at much higher risk of losing their sight.

A research partnership between Google and the Perth-based Lions Eye Institute has examined the technology's effectiveness in screening for diabetic retinopathy, a condition which causes lesions at the back of the retina that can lead to blindness.

Aboriginal and Torres Strait Islander people are three times more likely to lose their vision than non-Indigenous Australians.

They are 14 times more likely than non-Indigenous Australians to suffer diabetes-induced blindness, according to the institute.

A retrospective study, published in the British Journal of Ophthalmology, assessed the performance of Google's AI model in screening for diabetic retinopathy using anonymised data from Indigenous patients at Perth's Derbarl Yerrigan health service.

It found the deep-learning system showed "improved sensitivity and similar specificity" compared to detection by retinal specialists.

The finding is significant given concerns about racial bias which can affect the performance of artificial intelligence in relation to minority groups.

Lion's Eye Institute chief executive Angus Turner said it was important to understand how the technology could be applied in a real-world setting to earlier detect and treat preventable vision impairments.

"Diabetic retinopathy screening represents one of the earliest and most well-developed use cases for AI technology and research like this is critical to understanding its role and potential in a clinical setting," Dr Turner said.

"The findings have promising implications for validating algorithms in diverse populations, which is of critical importance for those that suffer disproportionately from diabetes-related complications."

Google's artificial intelligence is already used by clinicians and community health workers to help screen for diabetic retinopathy in India and Thailand.

Derbarl Yerrigan chief executive Tracey Brand said improving screening would allow patients to receive timely treatment from visiting specialists at community-controlled clinics.

Sign up to read this article
Read news from 100’s of titles, curated specifically for you.
Already a member? Sign in here
Related Stories
Top stories on inkl right now
Our Picks
Fourteen days free
Download the app
One app. One membership.
100+ trusted global sources.