CFAI-NETt

Installation

We recommend the following platforms. In addition, you need to install the necessary packages using the following instructions. Please leave a message for any questions. And install a runtime environment that supports Mamba:

Python 3.8 / Pytorch 2.0.0 / NVIDIA GeForce RTX 4090 / CUDA 11.8.0 / Ubuntu
pip install -r requirements.txt
cd ./kernels/selective_scan
pip install -e .

Prepare data & Pretrained model

Dataset:

• Synapse Multi-Organ Dataset: Sign up in the official Synapse website and download the dataset , save in the dataset/synapse/ folder.
• ACDC Dataset: Download the preprocessed ACDC dataset from TransUNet and move into dataset/acdc/ folder.

ImageNet pretrained model:

You should download the pretrained VMamba-Tiny V2 model (vssm1_tiny_0230s_ckpt_epoch_264) from VMamba, and then put it in the model/pretrain/ folder for initialization.

Training

Using the following command to train & evaluate MSVM-UNet:

# Synapse Multi-Organ Dataset
python train_synapse.py
# ACDC Dataset
python train_acdc.py

test

Please put the trained checkpoints file into the inference.py file for testing:

python inference.py