Search papers, labs, and topics across Lattice.
This paper addresses semi-supervised domain adaptation for medical image severity classification by incorporating ordinal relationships between classes. They introduce Cross-Domain Ranking, which learns rank scores by ranking sample pairs across source and target domains, and Continuous Distribution Alignment, which aligns the distributions of these rank scores. Experiments on ulcerative colitis and diabetic retinopathy datasets demonstrate improved severity classification performance by effectively aligning class-specific rank score distributions.
Forget aligning features – aligning *rankings* of samples across domains unlocks better severity classification in medical imaging when labels are scarce.
Semi-supervised domain adaptation leverages a few labeled and many unlabeled target samples, making it promising for addressing domain shifts in medical image analysis. However, existing methods struggle with severity classification due to unclear class boundaries. Severity classification involves naturally ordered class labels, complicating adaptation. We propose a novel method that aligns source and target domains using rank scores learned via ranking with class order. Specifically, Cross-Domain Ranking ranks sample pairs across domains, while Continuous Distribution Alignment aligns rank score distributions. Experiments on ulcerative colitis and diabetic retinopathy classification validate the effectiveness of our approach, demonstrating successful alignment of class-specific rank score distributions.
