In a recent interview on RTHK’s Hong Kong United television programme, Professor He Mingguang offered sight-saving hope to diabetes sufferers. The Director of the Research Centre for SHARP Vision and Chair Professor of Experimental Ophthalmology at the PolyU School of Optometry discussed his innovative AI-enabled self-testing retinal fundus camera, designed to enhance the efficiency and accuracy of diabetic retinopathy detection.

The development of an AI-driven diabetic eye screening system fulfils the vital need in healthcare for precisely and reliably identifying diabetic retinopathy. The condition is common among the 10% of the population in Hong Kong afflicted with diabetes, who need regular eye examinations. Conventional diagnostic techniques require extensive equipment and skilled personnel, which makes screenings laborious and expensive. 

New eye exams only take two minutes

To address these difficulties, the research team devised a portable, self-service fundus camera integrated with an AI analysis system. The new method improves diagnostic precision and enables the rapid, economical screening of large numbers of patients. Following five years of research and rigorous clinical testing, the system has attained remarkable accuracy rates of over 90%.

This innovation is expected to have a substantial impact on patients and the healthcare sector. Offering inexpensive and accessible tests available around the clock, the AI system may draw more people to eyecare. The rapid identification of people requiring therapy also permits prompt referrals to professionals. The new screening procedure is fast. While conventional examinations often require around half an hour, the AI approach takes just two minutes. This efficiency is crucial, especially as demand for basic eyecare is increasing. The need is particularly acute among individuals aged 65 and older, who make up 10-15% of the population and are more vulnerable to ocular diseases. 

The growing shortage of ophthalmologists and optometrists to meet this growing demand makes the integration of AI a viable solution for enhancing access to eyecare services. By streamlining the screening process, this technology promises to improve patient outcomes, as well as alleviate pressure on the healthcare system.
 

During a recent visit to PolyU by the Ministry of Education, a delegation explored various projects, including this AI-enabled self-testing retinal fundus camera that improves the accuracy of detecting diabetic retinopathy, highlighting the industry and academia integration in advancing medical innovation at PolyU.

Watch the interview (04:09-06:33).
Click here to learn more about the invention.