Attention-based deep learning network for predicting World Health Organization meningioma grade and Ki-67 expression based on magnetic resonance imaging
Question Can artificial intelligence accurately assess meningioma WHO grade and Ki-67 expression from preoperative MRI to guide personalised treatment and follow-up strategies? Findings The attention-enhanced nn-Unet segmentation achieved high accuracy, while 2.5D deep learning models with attention mechanisms achieved accurate prediction of grades and Ki-67. Clinical relevance Our fully automated 2.5D deep learning model, enhanced with attention mechanisms, accurately predicts WHO grades and Ki-67 expression levels in meningiomas, offering a robust, objective, and non-invasive solution to support clinical diagnosis and optimise treatment planning.
Authors
- Yongping You (ORCID: https://orcid.org/0000-0003-4753-5515)
- Qing Yang (ORCID: https://orcid.org/0000-0003-4844-4690)
- Zhichao Wang (ORCID: https://orcid.org/0000-0002-0136-1911)
- Junxia Zhang (ORCID: https://orcid.org/0000-0002-8753-1650)
- Xiefeng Wang (ORCID: https://orcid.org/0000-0001-7031-5532)
- Zhiyan Shen
- Chenfei Lu (ORCID: https://orcid.org/0000-0002-6217-3869)
- Xing Cheng (ORCID: https://orcid.org/0000-0003-2704-1323)
- Xiao Fan (ORCID: https://orcid.org/0000-0002-0195-7790)
- Z. Q. Liu (ORCID: https://orcid.org/0009-0009-0526-3612)
- Chen Li (ORCID: https://orcid.org/0000-0002-4546-500X)
- Chunfa Qian
- Kefan Song
- Jintan Li
- Jianxing Yin
- Huaning Li
Institutions
- Nanjing Brain Hospital (CN)
- Jiangsu Cancer Hospital (CN)
- Jiangsu Province Hospital (CN)
- Nanjing Medical University (CN)
Publication Details
- Journal
- European Radiology
- Published
- 2025-08-20
- DOI
- https://doi.org/10.1007/s00330-025-11958-7
- Citations
- 2
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 2.72
Funders
- National Natural Science Foundation of China
- Government of Jiangsu Province