Creation dataset by Brainweb Simulator.
The Following figure shows the configs in Simulator
dataset generated in three mode: ""MS mild"", ""MS moderate"", """MS severe""
- convert 3D T1 (.mnc) to 2D
- normalization
- remove extra tissues from T1
- remove extra tissues from mask
- Install dependencies:
pip install -r requirements.txt
- Training
python train.py --dataset /path/to/hdf5 --BS_train batch_size --BS_test batch_size --img_h image_height --img_w image_width --epochs epochs_number --model_name /model/path
Example:**
python train.py --dataset sample/dataset_without_noise_csf_1.hdf5 --BS_train 32 --BS_test 32 --img_h 256 --img_w 176 --epochs 40 --model_name csf_tissue
python test.py --images /path/to/images/T1 --labels /path/to/groundtruth --model_path /path/to/model --save_output True/False
| CSF | GM | WM | |
|---|---|---|---|
| mIOU | 0.69 | 0.83 | 0.72 |
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent