Prediction of neoadjuvant therapy response in breast cancer based on interpretable artificial intelligence
We developed an interpretable AI model that predicts response to neoadjuvant therapy in breast cancer using H&E slides. The model's predictions are biologically interpretable, correlating with TME dynamics and spatial TIL patterns, offering a novel strategy for personalizing NAT treatment strategies.
Authors
- Hongqun Tang (ORCID: https://orcid.org/0000-0001-5837-7853)
- Hui Xu (ORCID: https://orcid.org/0000-0002-7238-2779)
- Wei Qu (ORCID: https://orcid.org/0000-0003-2075-9033)
- Hongliang Ji
- Jianhong Tu
- Jing Huang (ORCID: https://orcid.org/0000-0002-0814-838X)
- Libin Deng (ORCID: https://orcid.org/0000-0001-8648-6985)
- Yiwei Wang (ORCID: https://orcid.org/0000-0003-2999-8727)
- Shiwei Zhang (ORCID: https://orcid.org/0009-0001-9332-4930)
- Xin Shu (ORCID: https://orcid.org/0009-0000-8337-331X)
- Yao Zhou (ORCID: https://orcid.org/0000-0002-9680-5246)
- Hao Fang (ORCID: https://orcid.org/0000-0001-8530-3999)
- Fan Wang (ORCID: https://orcid.org/0009-0009-3743-1107)
- Fan Niu
Institutions
- Nanchang University (CN)
- First Affiliated Hospital of Jiangxi Medical College (CN)
- Integrated Chinese Medicine (China) (HK)
- Third Affiliated Hospital of Nanchang University (CN)
- Third Hospital of Nanchang (CN)
- Second Hospital of Nanchang (CN)
- Buchang Pharma (China) (CN)
- Clinical Research Management (US)
Publication Details
- Journal
- International Journal of Surgery
- Published
- 2025-09-24
- DOI
- https://doi.org/10.1097/js9.0000000000003326
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00