ADAptation: Reconstruction-based Unsupervised Active Learning for Breast Ultrasound Diagnosis

2025-07-01

AI Summary

To address performance degradation of deep models caused by cross-domain distribution shift in breast ultrasound diagnosis—and the high annotation cost in the target domain—this paper proposes a reconstruction-based unsupervised active learning framework. Methodologically: (1) diffusion models align image styles across domains; (2) a hyperspherical-constrained contrastive learning network enhances feature discriminability; and (3) a dual-scoring mechanism, driven by reconstruction error, jointly quantifies sample uncertainty and cross-domain representativeness. Experiments on four public breast ultrasound datasets demonstrate that the method significantly outperforms state-of-the-art active learning approaches under limited annotation budgets, exhibiting strong generalization and clinical applicability. The source code is publicly available.

Technology Category

• Machine Learning: Active Learning
• Computer Vision: Diffusion Models for Vision
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Deep learning-based diagnostic models often suffer performance drops due to distribution shifts between training (source) and test (target) domains. Collecting and labeling sufficient target domain data for model retraining represents an optimal solution, yet is limited by time and scarce resources. Active learning (AL) offers an efficient approach to reduce annotation costs while maintaining performance, but struggles to handle the challenge posed by distribution variations across different datasets. In this study, we propose a novel unsupervised Active learning framework for Domain Adaptation, named ADAptation, which efficiently selects informative samples from multi-domain data pools under limited annotation budget. As a fundamental step, our method first utilizes the distribution homogenization capabilities of diffusion models to bridge cross-dataset gaps by translating target images into source-domain style. We then introduce two key innovations: (a) a hypersphere-constrained contrastive learning network for compact feature clustering, and (b) a dual-scoring mechanism that quantifies and balances sample uncertainty and representativeness. Extensive experiments on four breast ultrasound datasets (three public and one in-house/multi-center) across five common deep classifiers demonstrate that our method surpasses existing strong AL-based competitors, validating its effectiveness and generalization for clinical domain adaptation. The code is available at the anonymized link: https://github.com/miccai25-966/ADAptation.

Problem

• Addresses performance drops in deep learning models due to distribution shifts
• Reduces annotation costs while maintaining model performance via active learning
• Bridges cross-dataset gaps using diffusion models and dual-scoring mechanism

Innovation

• Uses diffusion models for domain style translation
• Hypersphere-constrained contrastive learning network
• Dual-scoring balances uncertainty and representativeness