Spener

This is the official code repository of our work "Unsupervised Self-prior Embedding Implicit Neural Respresentation for Iterative Sparse-View CT Reconstruction" accepted by AAAI 2025. [arXiv][AAAI Proceedings]

Overview

Fig. 1. Overview of proposed Spener.

Results

Fig. 2. Qualitative results of CT image reconstructed by the compared methods under two dose settings, with both results reconstructed from AAPM dataset with 90 views.

File Tree

Spener
│   dataset.py                  # dataloader
│   netarch.py                  # network architecture
│   README.md                   # readme file
│   requirements.txt            # requirements file
│   train.py                    # train file
│   utils.py                    # tools
│
├── config
│       demo_config.json        # demo configuration file
│
├── demo_data
│       demo_AAPM_slice.nii.gz  # demo data 
│
├── Fig
│       pipeline.png     
│       results1.png     
│
└── output
    └── SV_90_I0_1e+06
            final_recon.nii.gz   # Spener reconstruction
            gt_img.nii.gz        # GT image
            sv_fbp.nii.gz        # reconstruction from sparse-view

Main Requirements

To run this project, you will need the following packages:

• PyTorch
• tinycudann
• torch-radon
• bm3d
• SimpleITK
• tqdm
• NumPy
• other dependencies

the repo of torch-radon is at https://github.com/carterbox/torch-radon.

conda create -n spener python=3.10
conda activate spener
pip install -r requirements.txt

Training and Checkpoints

To train Spener from scratch, navigate to the project directory and run the following command in your terminal:

python train.py

The code will train the Spener to reconstruct a demo slice ./demo_data/demo_AAPM_slice.nii.gz from the AAPM dataset under the following conditions:

• 90 projection views
• Normal dose settings ($I_0=10^6$)

The training config file is ./config/demo_config.json. The code will save the corresponding model weights and reconstruction results in each iteration in ./output.

Others

NIFTI files (.nii) can be viewed using the ITK-SNAP software, available for free download at: http://www.itksnap.org/pmwiki/pmwiki.php?n=Downloads.SNAP4

Citation

@article{Tian_Chen_Wu_Du_Shi_Wei_Zhang_2025, 
title={Unsupervised Self-Prior Embedding Neural Representation for Iterative Sparse-View CT Reconstruction}, 
volume={39}, 
url={https://ojs.aaai.org/index.php/AAAI/article/view/32794}, 
DOI={10.1609/aaai.v39i7.32794}, 
number={7}, 
journal={Proceedings of the AAAI Conference on Artificial Intelligence}, 
author={Tian, Xuanyu and Chen, Lixuan and Wu, Qing and Du, Chenhe and Shi, Jingjing and Wei, Hongjiang and Zhang, Yuyao}, 
year={2025}, 
month={Apr.}, 
pages={7383-7391} 
}