Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
You can not select more than 25 topics
Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
Tong Zhang
90d701c96f
|
11 months ago | |
|---|---|---|
| imgs | 1 year ago | |
| scripts | 1 year ago | |
| util | 1 year ago | |
| .gitignore | 1 year ago | |
| main_segmentation.py | 1 year ago | |
| models_3dvit.py | 1 year ago | |
| package.json | 1 year ago | |
| readme.md | 11 months ago | |
| requirements.txt | 1 year ago | |
| tags | 1 year ago | |
| unetr.py | 1 year ago | |
| utils.py | 1 year ago | |
readme.md
基于3D Transformer架构的掩码预训练及全心脏分割网络
Requirements
pip install -r requirements.txt
基于UNETR的三维全心脏分割网络 Quick Start
Dataset
We use MM-WHS in for 3D segmentation tasks. Before runnning the code, you need to move data to share memory.
cp /path/to/MM-WHS /dev/shm
Re-implement
Open the root directory, Exp-4000 can be runned by following command.
bash ./scripts/exp4000_seg_whs_train_vitBase_scratch.sh
Run on Openi
You can choose dockerhub.pcl.ac.cn:5000/user-images/openi:jiaxin-v0.0 as docker image.
基于3D-Transformer架构的预训练及全心脏分割网络
Python Shell Text
