Addressing Domain Shift via Imbalance-Aware Domain Adaptation in Embryo Development Assessment

TL;DR

This paper introduces IADA, a novel framework that addresses both domain shift and class imbalance in medical imaging, significantly improving embryo development assessment accuracy across diverse imaging conditions.

Contribution

The paper proposes IADA, a new method combining adaptive feature learning, balanced domain alignment, and adaptive thresholding to handle dual challenges in medical imaging.

Findings

Up to 25.19% higher accuracy in embryo assessment.

Robust generalization with up to 12.56% AUC improvement.

Effective across four different imaging modalities.

Abstract

Deep learning models in medical imaging face dual challenges: domain shift, where models perform poorly when deployed in settings different from their training environment, and class imbalance, where certain disease conditions are naturally underrepresented. We present Imbalance-Aware Domain Adaptation (IADA), a novel framework that simultaneously tackles both challenges through three key components: (1) adaptive feature learning with class-specific attention mechanisms, (2) balanced domain alignment with dynamic weighting, and (3) adaptive threshold optimization. Our theoretical analysis establishes convergence guarantees and complexity bounds. Through extensive experiments on embryo development assessment across four imaging modalities, IADA demonstrates significant improvements over existing methods, achieving up to 25.19\% higher accuracy while maintaining balanced performance across classes. In challenging scenarios with low-quality imaging systems, IADA shows robust generalization with AUC improvements of up to 12.56\%. These results demonstrate IADA's potential for developing reliable and equitable medical imaging systems for diverse clinical settings. The code is made public available at \url{https://github.com/yinghemedical/imbalance-aware_domain_adaptation}