Implementation of the network for meningioma segmentation.
The network is a modified 3D UNet. We use the ResNeSt blocks, proposed by Zhang et al. [arXiv], in the encoder.
The code has been only tested in the environment as following
- Ubuntu 18.04.6 LTS
- Python 3.6.13
- Pytorch 1.1.0
Haolin Chen, Shuqi Li, Youming Zhang, Lizhi Liu, Xiaofei Lv, Yongju Yi, Guangying Ruan, Chao Ke, Yanqiu Feng. "Deep learning-based automatic segmentation of meningioma from multiparametric MRI for preoperative meningioma differentiation using radiomic features: A multicentre study" (under review)
See LICENSE