Benign and malignant bone lesion diagnosis based on self-supervised and radiomics fusion using SPECT/CT images
SPARC-Net, integrating SPECT-guided self-supervised learning with CT-based radiomics, improves classification of benign and malignant lesions, enhancing the accuracy, robustness, and interpretability of SPECT/CT imaging for bone tumor diagnosis.
Authors
- Hong Shangguan
- X. H. Bai (ORCID: https://orcid.org/0000-0003-0489-3857)
- Yingxin Dai (ORCID: https://orcid.org/0009-0009-6827-9312)
- Guoxu Zhang (ORCID: https://orcid.org/0009-0005-1156-165X)
- Zhaomin Yao (ORCID: https://orcid.org/0000-0001-6373-8721)
- Yusong Pei
- Miao Liu (ORCID: https://orcid.org/0000-0002-7975-6702)
- Zhiguo Wang (ORCID: https://orcid.org/0000-0002-0811-2045)
- Weiming Xie
- Ying Zhan (ORCID: https://orcid.org/0000-0002-5768-6184)
- Xiaodan Wu (ORCID: https://orcid.org/0009-0005-4196-5751)
Institutions
- General Hospital of Shenyang Military Region (CN)
- Xijing Hospital (CN)
- Northeastern University (CN)
- Air Force Medical University (CN)
Publication Details
- Journal
- Journal of bone oncology
- Published
- 2025-10-31
- DOI
- https://doi.org/10.1016/j.jbo.2025.100723
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00