Multi-Class Segmentation - PraNet-V2

📌 Overview

Welcome to the multi-class segmentation zone of PraNet-V2! 🌈🏥

This directory contains all the goodies for training, testing, and inference on multi-class semantic segmentation tasks, including ACDC and Synapse datasets. 🏆

📂 Directory Structure

multi-class/
├── EMCAD/            # EMCAD model implementation
├── MERIT/            # MERIT model implementation
├── MIST/             # MIST model implementation
├── model_pth/        # Pre-trained model directory
├── dataset/          # Dataset directory
├── README.md         # This file

🚀 How to Run

Training

cd ./multi-class/MIST
python -W ignore Synapse_train.py --dual

Inference

cd ./multi-class/EMCAD
python -W ignore test_synapse.py --dual

Evaluation

# Evaluation results are logged in the `test_log/` directory

📥 Dataset Download

Please refer to the main README for dataset download instructions and place the datasets in the multi-class/dataset/ directory.

🏆 Citation

If you find our work useful, please consider citing us! 🏆👇

@article{hu2025pranet2,
  title={PraNet-V2: Dual-Supervised Reverse Attention for Medical Image Segmentation},
  author={Hu, Bo-Cheng and Ji, Ge-Peng and Shao, Dian and Fan, Deng-Ping},
  journal   = {arXiv preprint arXiv:2504.10986},
  year={2025},
  url       = {https://arxiv.org/abs/2504.10986}
}