Abstract

The emergence of artificial intelligence offers great potential to revolutionize practice pattern in highly image driven specialties of medicine, including Ophthalmology. In the 2 years, our research team has been collecting pairwise images, such as full colour fundus photograph (FCFP) and fundus fluorescein angiogram (FFA), on the same patients. Based on these images and Generative Adversarial Network (GAN), we developed algorithm that can “generate” FFA‐style of images based on traditional FCFP. With this technology, we can “visualize” the microaneurysm, retinal ischemia, retinal neovascularization from FCFP much better than ever before. In this lecture, we will summarize the progression on this “generative AI” work including the accuracy, challenges and opportunities.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call