Paper: Low-rank Angular Prior Guided Multi-diffusion Model for Few-shot Low-dose CT Reconstruction
IEEE Transactions on Computational Imaging, Vol. 10, pp. 1763-1774, 2024.
Authors: Wenhao Zhang, Bin Huang, Shuyue Chen, Xiaoling Xu, Weiwen Wu, Qiegen Liu
https://ieeexplore.ieee.org/abstract/document/10776993
The code and the algorithm are for non-comercial use only. Copyright 2024, School of Information Engineering, Nanchang University.
Low-dose computed tomography (LDCT) is essential in clinical settings to minimize radiation exposure; however, reducing the dose often leads to a significant decline in image quality. Additionally, conventional deep learning approaches typically require large datasets, raising con-cerns about privacy, costs, and time constraints. To ad-dress these challenges, a few-shot low-dose CT reconstruc-tion method is proposed, utilizing low-Rank Angular Pri-or (RAP) multi-diffusion model. In the prior learning phase, projection data is transformed into multiple con-secutive views organized by angular segmentation, allow-ing for the extraction of rich prior information through low-rank processing. This structured approach enhances the learning capacity of the multi-diffusion model. Dur-ing the iterative reconstruction phase, a stochastic differ-ential equation solver is employed alongside data con-sistency constraints to iteratively refine the acquired pro-jection data. Furthermore, penalized weighted least-squares and total variation techniques are integrated to improve image quality. Results demonstrate that the re-constructed images closely resemble those obtained from normal-dose CT, validating the RAP model as an effec-tive and practical solution for artifact and noise reduc-tion while preserving image fidelity in low-dose situation.
(a) The reference image, (b) FBP, (c) SART-TV, (d) CNN, (e) NCSN++, (f) U-ViT, (g) OSDM, (h) RAP (1).
python main_up.py --config=aapm_sin_ncsnpp_up.py --workdir=exp1_up --mode=train --eval_folder=result1_up
python main_middle.py --config=aapm_sin_ncsnpp_middle.py --workdir=exp1_middle --mode=train --eval_folder=result1_middle
python main_down.py --config=aapm_sin_ncsnpp_down.py --workdir=exp1_down --mode=train --eval_folder=result1_down
python PCsampling_demo.py
-
Generative Modeling in Sinogram Domain for Sparse-view CT Reconstruction
[Paper] [Code] -
One Sample Diffusion Model in Projection Domain for Low-Dose CT Imaging
[Paper] [Code] -
Iterative Reconstruction for Low-Dose CT using Deep Gradient Priors of Generative Model
[Paper] [Code] [PPT] -
REDAEP: Robust and Enhanced Denoising Autoencoding Prior for Sparse-View CT Reconstruction
[Paper] [Code] [PPT] -
Wavelet-improved score-based generative model for medical imaging
[Paper] -
Stage-by-stage Wavelet Optimization Refinement Diffusion Model for Sparse-view CT Reconstruction
[Paper] [Code] -
Dual-Domain Collaborative Diffusion Sampling for Multi-Source Stationary Computed Tomography Reconstruction
[Paper] [Code] -
Physics-informed DeepCT: Sinogram Wavelet Decomposition Meets Masked Diffusion
[Paper] [Code] -
MSDiff: Multi-Scale Diffusion Model for Ultra-Sparse View CT Reconstruction
[Paper] [Code] -
Ordered-subsets Multi-diffusion Model for Sparse-view CT Reconstruction
[Paper] -
Virtual-mask Informed Prior for Sparse-view Dual-Energy CT Reconstruction
[Paper] [Code] -
Raw_data_generation [Code]
-
PRO: Projection Domain Synthesis for CT Imaging [Paper] [Code]
-
UniSino: Physics-Driven Foundational Model for Universal CT Sinogram Standardization[Paper] [Code]