MMGan: a multimodal MR brain tumor image segmentation method
Leiyi Gao, Jiao Li, Ruixin Zhang, Hailu Hanna Bekele, Junzhu Wang, Yining Cheng, Hongxia Deng · 2023
Read MMGan: a multimodal MR brain tumor image segmentation method by Leiyi Gao, Jiao Li, Ruixin Zhang, Hailu Hanna Bekele, Junzhu Wang, Yining Cheng, Hongxia Deng (2023) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-10728273.txt.
Description
Leiyi Gao and six co-authors present a neural network that outlines brain tumors in MRI scans. Their 2023 paper joins two ideas: deep residual learning, which lets very deep networks train well, and adversarial training, in which a second network pushes the first to make cleaner outlines. The outlining part follows the U-Net shape common in medical imaging.
The team trained and tested the system, named MMGan, on the BRATS challenge data set of brain tumor scans. It beat leading rival methods on accuracy and sensitivity for the whole tumor, the tumor core, and the active rim. Its overlap score for the core reached 0.86. Cleaner outlines help doctors plan surgery and track growth over time. The authors suggest gains could grow with other loss functions.
Catalog record
| CREATOR | Leiyi Gao, Jiao Li, Ruixin Zhang, Hailu Hanna Bekele, Junzhu Wang, Yining Cheng, Hongxia Deng |
|---|---|
| YEAR | 2023 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 58 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |