Taiwan-based non-profit develops AI software for diabetic eye disease diagnosis
The Industrial Technology Research Institute, a non-profit R&D organisation in Taiwan, has unveiled its AI-assisted system for detecting diabetic eye diseases.
WHAT IT’S ABOUT
Point-of-care AI-DR can instantly diagnose diabetic retinopathy and diabetic eye oedema in five to 10 seconds. It can locate from fundus images four main lesions – microaneurysms, haemorrhages, soft exudates, and hard exudates – and detect two anatomical landmarks, the optic disc and the macular area. It can further identify 14 other common ocular fundus abnormalities, including retinal-related diseases, changes in blood vessels, and optic nerve-related diseases.
AI-DR also provides five severity levels of diabetic retinopathy and produces binary classifications for both diabetic retinopathy and diabetic macular oedema.
The AI model has been trained using 150,000 fundus images. In a study, AI-DR was shown to have 98% sensitivity and 96% specificity in diagnosing diabetic retinopathy.
According to ITRI, AI-DR can support all available fundus cameras worldwide and be used to create other solutions such as edge AI systems, standalone web applications, and private and public cloud-based services. The organisation has recently provided a use license for AI-DR to Acer for its VeriSee DR diagnostic software.
WHY IT MATTERS
Last year, there are about 537 million adults between the age of 20 and 79 who have diabetes. The International Diabetes Foundation expects their count to reach 643 million by 2030 and 783 million by 2045.
Currently, more than half of people living with diabetes globally will develop DR and about one in 15 will develop DME, according to the US National Eye Institute.
ITRI said that with diagnostic solutions like AI-DR, the risks of blindness caused by diabetes complications can be lowered. It also emphasised that in diabetes management today, it is crucial to track changes in fundus symptoms as this can be better monitored than blood sugar levels.
Additionally, AI-DR’s intuitive design allows GPs and other non-ophthalmologists to perform quick diabetic retinopathy screening as trained ophthalmologists, which can lead to increased screening rates, reduced treatment costs and improved patient outcomes by enabling early detection.
Other existing diabetic eye screening software products around Asia-Pacific are also powered by AI, including the ORAiCLE platform by New Zealand-based Toku Eyes, Singapore Eye Lesion Analyzer by the Singapore Eye Research Institute, and VUNO Med-Fundus AI by South Korean medical AI company VUNO.
Meanwhile, a recent Google-backed study in Thailand has demonstrated the accurate detection of diabetic retinopathy using a new deep-learning system.
Also, Samsung is turning old Galaxy smartphone devices into AI-powered fundus cameras to screen for diabetic eye diseases as part of its upcycling programme to improve access to eye health screening in underserved communities around the world.