Automatic Grading of Age-Related Cataract via Progressive Ordinal Attention and Topology-Constrained Prototype Learning.

In this study, we propose a progressive ordinal attention network (progressive ordinal attention network) for automated four-level cataract classification.

A spatial gating module is further incorporated to suppress background interference and enhance feature representation of the lens region.

The model shows improved capability in distinguishing adjacent severity levels and exhibits strong agreement with clinical annotations.