🧬 Multi-Modality Anatomical Image Fusion (MATLAB Implementation)

This repository contains the official MATLAB implementation of the method proposed in:

“A novel multimodality anatomical image fusion method based on contrast and structure extraction.” International Journal of Imaging Systems and Technology, 2022.

The code performs fusion of multi-modal medical images (e.g., CT–MRI, MR-T1–MR-T2, MR-Gad–MR-T1) using dense SIFT-based contrast analysis and structural layer extraction, followed by guided filtering and pyramid-based reconstruction.

---

📂 Repository Structure

├── main.m                # Main file to run the fusion pipeline
├── load_images.m         # Loads a sequence of multimodal images
├── fuseImage.m           # Core fusion method implementation
│
├── Pyramid_Decomposition/
├── Guided_Filter/
└── Dense_SIFT/

Make sure all folders remain in the same directory structure when running the code.

---

🛠️ Requirements

• MATLAB R2017b or later
• Image Processing Toolbox
• All subfolders added automatically via addpath inside fuseImage.m

No installation is needed beyond keeping the provided folders intact.

---

▶️ How to Run

1. Clone this repository

   git clone https://github.com/arbabsufyan/Proposed-Code
   cd <your-repo-folder>

2. Open MATLAB

3. Make sure MATLAB’s Current Folder is set to the repository directory.

4. Prepare your dataset

   • Place your input images inside one of the dataset folders (e.g. Dataset/CT-MRI/Pair 1)
   • Each folder should contain two or more images of the same size (.tif, .jpg, .gif, .bmp, or .png)

5. Run the main script

   main

6. The script will:

   • Load multimodal images
   • Display the input images
   • Perform fusion using the proposed method
   • Display the final fused image

7. The scale parameter of Dense SIFT can be adjusted inside main.m:

   F = fuseImage(I, 16);   % default = 16

---

📘 Example

To fuse MR-T1 and MR-T2 images:

I = load_images('./Dataset/MR-T1-MR-T2/Pair 1');
F = fuseImage(I, 16);
imshow(F);

---

📄 Citation

If you use this code in your research or publication, please cite our paper:

@article{sufyan2022novel,
  title={A novel multimodality anatomical image fusion method based on contrast and structure extraction},
  author={Sufyan, Arbab and Imran, Muhammad and Shah, Syed Attique and Shahwani, Hamayoun and Wadood, Arbab Abdul},
  journal={International Journal of Imaging Systems and Technology},
  volume={32},
  number={1},
  pages={324--342},
  year={2022},
  publisher={Wiley Online Library}
}

Your citation motivates future updates and helps support this research.