ADAptation: Reconstruction-based Unsupervised Active Learning for Breast Ultrasound Diagnosis
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908428920684544 |
|---|---|
| author | Duan, Yaofei Huang, Yuhao Yang, Xin Han, Luyi Xie, Xinyu Zhu, Zhiyuan He, Ping Chan, Ka-Hou Cui, Ligang Im, Sio-Kei Ni, Dong Tan, Tao |
| author_facet | Duan, Yaofei Huang, Yuhao Yang, Xin Han, Luyi Xie, Xinyu Zhu, Zhiyuan He, Ping Chan, Ka-Hou Cui, Ligang Im, Sio-Kei Ni, Dong Tan, Tao |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_00474 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ADAptation: Reconstruction-based Unsupervised Active Learning for Breast Ultrasound Diagnosis Duan, Yaofei Huang, Yuhao Yang, Xin Han, Luyi Xie, Xinyu Zhu, Zhiyuan He, Ping Chan, Ka-Hou Cui, Ligang Im, Sio-Kei Ni, Dong Tan, Tao Computer Vision and Pattern Recognition 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. |
| title | ADAptation: Reconstruction-based Unsupervised Active Learning for Breast Ultrasound Diagnosis |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2507.00474 |