Thyroid pathology image classification via multi-scale feature fusion and multi-instance learning
We propose a weakly supervised MIL framework integrating multi-scale analysis and cross-model feature fusion for thyroid cancer diagnosis. Our method showed promising and consistent results across internal and external datasets. While further clinical validation and workflow integration are needed, the results suggest the potential of this approach to assist pathologists in diagnostic workflows, particularly in resource-constrained settings.
Authors
- Guanxin Liu (ORCID: https://orcid.org/0009-0000-3421-2616)
- Kaisheng Xie
- Yuwei Jiang (ORCID: https://orcid.org/0000-0002-9533-0760)
- Xiaobo Yang (ORCID: https://orcid.org/0000-0002-5257-5100)
- Mengmeng Sun (ORCID: https://orcid.org/0000-0003-0177-1723)
- Ning Liao (ORCID: https://orcid.org/0009-0002-7465-953X)
- Lu Wang (ORCID: https://orcid.org/0000-0002-9437-6898)
- Kai Zhang (ORCID: https://orcid.org/0000-0002-1347-0373)
- Xiangzhi Li (ORCID: https://orcid.org/0000-0001-9407-0277)
- Bingfang He
- Shimei Zhao
- Yajun Ying
Institutions
- Guangxi University of Science and Technology (CN)
- Wenzhou Medical University (CN)
- Zhejiang Taizhou Hospital (CN)
- Hechi University (CN)
Publication Details
- Journal
- Diagnostic Pathology
- Published
- 2025-10-14
- DOI
- https://doi.org/10.1186/s13000-025-01703-5
- Primary Topic
- AI in cancer detection
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Natural Science Foundation of China
- Middle-aged and Young Teachers' Basic Ability Promotion Project of Guangxi