CDRMamba: A Framework for Automated CranioMaxilloFacial Defect Reconstruction Using Mamba-Based Modeling

This repository contains the code for the paper "CDRMamba: A Framework for Automated CranioMaxilloFacial Defect Reconstruction Using Mamba-Based Modeling".

Directory Structure

.
├── config.py
├── dataset
│   ├── dataset_lits_train.py
│   ├── dataset_lits_val.py
│   └── transforms.py
├── models
│   ├── __init__.py
│   ├── CDRMamba.py
│   └── TSSMamba.py
├── requirements.txt
├── train.py
├── test.py
└── utils
    ├── common.py
    ├── logger.py
    ├── loss.py
    ├── metrics.py
    └── weights_init.py

Dependencies

Please ensure you have the following dependencies installed:

• Python 3.8
• PyTorch
• tqdm
• SimpleITK
• numpy You can set up the conda environment using the requirements.txt file:

conda create --name myenv --file requirements.txt
conda activate myenv

Train and test

You can train the model using the following command:

python train.py

Test command:

python test.py

Configuration

You can set the training parameters such as learning rate, batch size, etc., in the config.py file.

Weights and Dataset

The weights and dataset are open and can be obtained by contacting 22210860059@m.fudan.edu.cn.