Figure 1.

The proposed CLIP/CLOOB framework: contrastive pre-training of the encoders ($h^{\mathbf{x}},h^{\mathbf{y}}$) of the two retinal imaging modalities (fundus images—$\mathbf{x}$ and OCT volumes—$\mathbf{y}$), followed by using the pre-trained encoders for downstream predictive tasks.