ATLASS: An AnaTomicaLly-Aware Self-Supervised Learning Framework for Generalizable Retinal Disease Detection

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Our framework enhances retinal disease classification, highlighting the role of self-supervised learning and anatomical relevance.

Integration of anatomical knowledge significantly improves performance in retinal disease detection and classification.
The use of specialized masking techniques with retinal structures enhances feature learning without extensive data annotation.
We employed a Vision Transformer with dual-masking strategies to ensure structural integrity during model training.
Comprehensive evaluation indicates strong performance across diverse retinal disease tasks, reinforcing clinical applicability.

Khan et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark