COROLLA: An Efficient Multi-Modality Fusion Framework with Supervised Contrastive Learning for Glaucoma Grading
This repo covers an reference implementation for the following papers in PyTorch：
(1) COROLLA: An Efficient Multi-Modality Fusion Framework with Supervised Contrastive Learning for Glaucoma Grading paper
For the supervised contrastive part, this repo is based on the following paper:
(1) Supervised Contrastive Learning. Paper / code
The contribution of COROLLA is finding a new modality generated from OCT volumes to achieve better and faster dignosis performance on glaucoma grading task. We use OCT-Explorer to get the thickness map. The dataset is available here.