Weakly supervised learning in thymoma histopathology classification: an interpretable approach
This model offers a significant advancement in thymoma classification, combining high accuracy with interpretability. By integrating weakly supervised learning, MIL, and attention mechanisms, it provides an interpretable AI framework that is applicable in clinical settings. The model reduces the diagnostic burden on pathologists and has the potential to improve patient outcomes by making AI tools more transparent and clinically relevant.
Authors
- Yangyang Fan (ORCID: https://orcid.org/0000-0003-2505-1915)
- Xiaoyu Yan (ORCID: https://orcid.org/0009-0008-3356-6939)
- Lei Cui (ORCID: https://orcid.org/0000-0003-4365-5173)
- Chunbao Wang (ORCID: https://orcid.org/0000-0002-8795-0142)
- Xian Du (ORCID: https://orcid.org/0000-0003-3515-2840)
- Hansheng Li (ORCID: https://orcid.org/0000-0002-0923-196X)
- Jie Lian (ORCID: https://orcid.org/0000-0002-2005-2022)
- Li Chen (ORCID: https://orcid.org/0009-0002-8277-9494)
- Hong Chang (ORCID: https://orcid.org/0000-0002-3390-8483)
- Yue Jiang (ORCID: https://orcid.org/0000-0002-9427-6627)
- Xiao Teng
- Zhe Yang (ORCID: https://orcid.org/0009-0008-5827-8095)
- Caihong Ran
- Xiaolin Wang
Institutions
- Northwest University (CN)
- First Affiliated Hospital of Xi'an Jiaotong University (CN)
- Xi'an Jiaotong University (CN)
Publication Details
- Journal
- Frontiers in Medicine
- Published
- 2024-12-11
- DOI
- https://doi.org/10.3389/fmed.2024.1501875
- Citations
- 2
- Primary Topic
- Myasthenia Gravis and Thymoma
- Type
- article
- Field-Weighted Citation Impact
- 0.74
Funders
- Strong