TDSI_super-resolution_DEMO

Medical Image Super-Resolution Models

This repository contains two deep learning models for medical image segmentation super-resolution.

Models

1. ConvKAN Model

A KAN (Kolmogorov-Arnold Network) based model for medical image super-resolution.

2. MLP-GAN Model

A GAN-based model with UNet generator for super-resolution segmentation.

Project Structure

NewDataSet/
├── model_convKAN/
│   ├── notebooks/          # Jupyter notebooks for inference
│   ├── src/               # Source code
│   └── README.md          # Detailed documentation
│
└── model_mlpGAN/
    ├── inference_demo.ipynb              # Inference demonstration
    ├── Copie de 210_justnewdataset.pth  # Pre-trained model weights
    ├── P001_img.nrrd                    # Sample image data
    └── P001_seg.nrrd                    # Sample segmentation data

Usage

Each model folder contains:

• Inference notebooks for demonstration
• Pre-trained model weights
• Sample data for testing

Refer to the individual model folders for specific usage instructions.