Anterior cruciate ligament injuries diagnosis using slice‐aligning and multi‐level feature aggregation
MLFANet-SA combines slice selection and feature aggregation to improve lesion localization and classification of ACL injuries. Our model outperforms previous approaches in both public and private datasets. Its ability to reduce manual labeling while maintaining high accuracy suggests strong potential in diagnosing ACL injuries for radiologists.
Authors
- Ji He (ORCID: https://orcid.org/0000-0001-9811-6500)
- Wenchao Jiang (ORCID: https://orcid.org/0000-0001-6765-2466)
- Tongxin Zhang (ORCID: https://orcid.org/0000-0002-1719-6412)
- Sae‐Won Han (ORCID: https://orcid.org/0000-0003-3275-431X)
- Lei Zhu (ORCID: https://orcid.org/0000-0003-1685-7968)
- Long Li (ORCID: https://orcid.org/0000-0003-0472-7314)
- Yi Zheng (ORCID: https://orcid.org/0000-0001-5547-0956)
Institutions
- Guangdong University of Technology (CN)
- Nanfang Hospital (CN)
- Twelfth Guangzhou City People's Hospital (CN)
- Southern Medical University (CN)
- Guangzhou Medical University (CN)
Publication Details
- Journal
- Medical Physics
- Published
- 2025-11-01
- DOI
- https://doi.org/10.1002/mp.70130
- Primary Topic
- Knee injuries and reconstruction techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00