Principal Investigator:
Dr SHI Danli, Research Assistant Professor, School of Optometry
This innovation addresses diabetic retinopathy (DR), a leading cause of blindness, by replacing invasive and expensive fundus fluorescein angiography (FFA) with a non-invasive, cost-effective screening solution empowered by generative artificial intelligence (GenAI). It converts colour fundus photography into high-resolution, realistic FFA images, preserving critical lesion details without the need for dye injections. It also supports ultra-widefield imaging and dynamic lesion-preserving video generation.
Validated by retinal specialists, this method enhances DR screening accuracy, reduces costs and improves patient comfort. Ongoing multi-centre clinical trials will assess its diagnostic performance, treatment outcomes and efficiency compared with traditional FFA. Offering a safe, scalable and impactful solution, this GenAI-driven innovation revolutionises DR evaluation while making the process more accessible and efficient in clinical practice.
