MAUNet: a mixed attention U-net with spatial multi-dimensional convolution and contextual feature calibration for 3D brain tumor segmentation in multimodal MRI
Extensive experiments on the BraTS2019 and BraTS2020 datasets demonstrate the superiority of MAUNet, achieving average Dice scores of 84.5 and 83.8%, respectively. Ablation studies further validate the effectiveness of each proposed module, highlighting their contributions to improved segmentation accuracy. Our work provides a robust and efficient solution for automated brain tumor segmentation, offering significant potential for clinical applications.
Authors
- Wenna Chen (ORCID: https://orcid.org/0000-0003-2543-8271)
- Jincan Zhang
- Xinghua Tan (ORCID: https://orcid.org/0009-0005-3812-533X)
- Ganqin Du (ORCID: https://orcid.org/0000-0001-9097-2505)
- Rongfu Lv (ORCID: https://orcid.org/0009-0008-9709-2613)
- C.W. Cai (ORCID: https://orcid.org/0009-0004-9493-0539)
Institutions
- Henan University of Science and Technology (CN)
- First Affiliated Hospital of Henan University of Science and Technology (CN)
Publication Details
- Journal
- Frontiers in Neuroscience
- Published
- 2025-10-08
- DOI
- https://doi.org/10.3389/fnins.2025.1682603
- Primary Topic
- Advanced Neural Network Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- University of Pennsylvania
- Foundation of Henan Educational Committee