VoxHRNet
This is the official implementation of the following paper:
Whole Brain Segmentation with Full Volume Neural Network
Yeshu Li, Jonathan Cui, Yilun Sheng, Xiao Liang, Jingdong Wang, Eric I-Chao Chang, Yan Xu
Network
Installation
The following environments/libraries are required:
- Python 3
- yacs
- SimpleITK
- apex
- pytorch
- nibabel
- numpy
- scikit-image
- scipy
Quick Start
Data Preparation
Download the LPBA40 and Hammers n30r95 datasets.
After renaming, your directory tree should look like:
$ROOT
├── data
│ └── LPBA40_N4_RN
│ ├── aseg_TEST001.nii.gz
│ ├── ...
│ ├── aseg_TEST010.nii.gz
│ ├── aseg_TRAIN001.nii.gz
│ ├── ...
│ ├── aseg_TRAIN027.nii.gz
│ ├── aseg_VALIDATE001.nii.gz
│ ├── ...
│ ├── aseg_VALIDATE003.nii.gz
│ ├── orig_TEST001.nii.gz
│ ├── ...
│ ├── orig_TEST010.nii.gz
│ ├── orig_TRAIN001.nii.gz
│ ├── ...
│ ├── orig_TRAIN027.nii.gz
│ ├── orig_VALIDATE001.nii.gz
│ ├── ...
│ └── orig_VALIDATE003.nii.gz
└── VoxHRNet
├── voxhrnet.py
├── ...
└── train.py
Create a YACS configuration file and make changes for specific training/test settings accordingly. We use config_lpba.yaml as an example as follows.
Train
Run
python3 train.py --cfg config_lpba.yaml
Test
Run
python3 test.py --cfg config_lpba.yaml
Pretrained Models
For the LPBA40 dataset, we number the subjects from 1-40 alphabetically and split them into 4 folds sequentially. The k-th fold is selected as the test set in the k-th split.
For the Hammers n30r95 dataset, the first 20 subjects and last 10 subjects are chosen as the training and test set respectively.
Their pretrained models can be found in the release page of this repository.
Citation
Please cite our work if you find it useful in your research:
@article{LI2021101991,
title = {Whole brain segmentation with full volume neural network},
journal = {Computerized Medical Imaging and Graphics},
volume = {93},
pages = {101991},
year = {2021},
issn = {0895-6111},
doi = {https://doi.org/10.1016/j.compmedimag.2021.101991},
url = {https://www.sciencedirect.com/science/article/pii/S0895611121001403},
author = {Yeshu Li and Jonathan Cui and Yilun Sheng and Xiao Liang and Jingdong Wang and Eric I.-Chao Chang and Yan Xu},
keywords = {Brain, Segmentation, Neural networks, Deep learning},
abstract = {Whole brain segmentation is an important neuroimaging task that segments the whole brain volume into anatomically labeled regions-of-interest. Convolutional neural networks have demonstrated good performance in this task. Existing solutions, usually segment the brain image by classifying the voxels, or labeling the slices or the sub-volumes separately. Their representation learning is based on parts of the whole volume whereas their labeling result is produced by aggregation of partial segmentation. Learning and inference with incomplete information could lead to sub-optimal final segmentation result. To address these issues, we propose to adopt a full volume framework, which feeds the full volume brain image into the segmentation network and directly outputs the segmentation result for the whole brain volume. The framework makes use of complete information in each volume and can be implemented easily. An effective instance in this framework is given subsequently. We adopt the 3D high-resolution network (HRNet) for learning spatially fine-grained representations and the mixed precision training scheme for memory-efficient training. Extensive experiment results on a publicly available 3D MRI brain dataset show that our proposed model advances the state-of-the-art methods in terms of segmentation performance.}
}
Acknowledgement
A large part of the code is borrowed from HRNet.
Contributing
This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.opensource.microsoft.com.
When you submit a pull request, a CLA bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., status check, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.
This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact [email protected] with any additional questions or comments.
7 Nov 27, 2022
5 Nov 30, 2022
1 May 31, 2022
32 Feb 21, 2022
53 Dec 4, 2022
148 Dec 1, 2022
117 Dec 11, 2022
106 Dec 28, 2022
16 Sep 11, 2021
9 May 25, 2022
4 Dec 14, 2021
58 Dec 12, 2022
9 Dec 04, 2022
51 Dec 11, 2022
11.3k Dec 31, 2022
338 Jan 03, 2023
12 Oct 23, 2022
3 Dec 05, 2022
148 Dec 21, 2022
1 Jan 18, 2022
4 Oct 04, 2022
89 Dec 09, 2022
3 Feb 06, 2022
12 Oct 06, 2022
36 Nov 15, 2022
5 Aug 19, 2022
15 Dec 22, 2022
11 Oct 15, 2022
4 Aug 24, 2022