Abstract
Presentation Description :
In this talk, deep learning-based diabetic retinopathy (DR) classification using OCTA and a new interpretable method will be introduced. This method may provide a useful tool for DR clinics and transform their diagnosis by eliminating the need for traditional biomarker detection. Additionally, we will explore how deep learning, in conjunction with OCTA, unveils the intricate microvascular changes associated with DR. These new findings and insights will be shared in this talk.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.