Using AI to help close the eye health care gapJuly 11, 2019
CERA is working with the Nganampa Health Council and the Fred Hollows Foundation to help close the eye health care gap in remote Australia.
Health workers in one of Australia’s most remote Aboriginal communities are using an artificial intelligence (AI) tool developed at CERA to help detect and prevent leading causes of blindness.
Workers from the Nganampa Health Council, in the Anangu Pitjantjatjara Yankunytjatjara (APY) Lands in South Australia are trialling the tool in a research partnership with the Fred Hollows Foundation and CERA to assist health workers to detect diabetic retinopathy in the community and refer people for further treatment.
It is hoped the trial, to be conducted at seven clinics in the remote region, will lead to earlier detection and treatment of diabetic retinopathy. The debilitating eye disease is a major cause of bilateral vision impairment and blindness in Indigenous Australians.
Professor Mingguang He and Dr Jane Scheetz are leading the study, Using artificial intelligence to detect and refer diabetic retinopathy in remote Aboriginal communities, which will determine clinician acceptability and referral accuracy of the new technology.
Earlier trials of the technology have shown the AI system is able to achieve high levels of accuracy in detecting common blinding eye diseases including diabetic retinopathy, glaucoma and age-related macular degeneration – and this new trial will test its effectiveness in a real-world setting.
Jane recently visited Alice Springs to help train health care workers in the use of the tool and data collection for the two-year study.
The trial will see health workers take a photograph of the patients’ retina and upload it to the AI tool, which is able to provide a report on the health of the patients’ eyes within seconds. This report will be provided to patients in both English and Pitjantjatjara – determining whether the patient needs to be referred to a specialist for further treatment and assessment.
Currently, the region – home to about 2500 Aboriginal people – relies on the services of ophthalmologists in Adelaide to manually assess retinal photographs taken by health workers.
However, it’s hoped that using the AI will speed up the referral process by removing the need to send photographs to Adelaide for assessment.
“Indigenous Australians are at particular risk, experiencing up to three times higher rates of vision impairment and blindness and much higher rates of late-stage diabetic retinopathy compared to non-Indigenous Australians,” says Dr Scheetz.
“By working with Aboriginal health services and the Fred Hollows Foundation we hope our AI tool can increase access to eye health care in remote communities, lead to quicker diagnosis and treatment and prevent blindness in Aboriginal communities.”