Deep learning for Three‐Class Classification of ground-glass nodules on non-enhanced chest CT: A multicenter comparative study of CNN architectures
The Res2Net 3D model accurately differentiates GGN subtypes using non-contrast CT, showing high performance, especially for invasive adenocarcinoma.
Authors
- Hong Yu (ORCID: https://orcid.org/0000-0003-0667-8413)
- Meihua Shao (ORCID: https://orcid.org/0000-0003-1203-9730)
- Jianfei Fu (ORCID: https://orcid.org/0000-0002-3036-1056)
- Linyang He (ORCID: https://orcid.org/0000-0001-5181-8337)
- Lin Zhu (ORCID: https://orcid.org/0000-0002-6772-5730)
- Jian Wang (ORCID: https://orcid.org/0000-0003-1194-9046)
- Cui Zhang (ORCID: https://orcid.org/0000-0002-1727-0563)
- Guohua Cheng (ORCID: https://orcid.org/0000-0002-8640-1037)
- Hengfeng Shi
Institutions
- Shanghai Jiao Tong University (CN)
- Shanghai Chest Hospital (CN)
- Tongde Hospital of Zhejiang Province (CN)
- Lishui Central Hospital (CN)
- Anqing City Hospital (CN)
Publication Details
- Journal
- European Journal of Radiology Open
- Published
- 2025-10-09
- DOI
- https://doi.org/10.1016/j.ejro.2025.100690
- Primary Topic
- Lung Cancer Diagnosis and Treatment
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Anhui Provincial Key Research and Development Plan
- Science and Technology Program of Zhejiang Province