Paper: One-sample Diffusion Modeling in Projection Domain for Low-dose CT Imaging

Authors: Bin Huang, Shiyu Lu, Liu Zhang, Boyu Lin, Weiwen Wu, Member, IEEE Qiegen Liu, Senior Member, IEEE IEEE Transactions on Radiation and Plasma Medical Sciences, vol. 8, no. 8, pp. 902-915, Nov. 2024.
https://ieeexplore.ieee.org/document/10506793

Low-dose computed tomography (CT) is crucial in clinical applications for reducing radiation risks. However, lowering the radiation dose will significantly degrade the image quality. In the meanwhile, common deep learning methods require large data, which are short for privacy leaking, expensive, and time-consuming. Therefore, we propose a fully unsupervised one-sample diffusion modeling (OSDM) in projection domain for low-dose CT reconstruction. To extract sufficient prior information from a single sample, the Hankel matrix formulation is employed. Besides, the penalized weighted least-squares and total variation are introduced to achieve superior image quality. Firstly, we train a score-based diffusion model on one sinogram to capture the prior distribution with input tensors extracted from the structural-Hankel matrix. Then, at inference, we perform iterative stochastic differential equation solver and data-consistency steps to obtain sinogram data, followed by the filtered back-projection algorithm for image reconstruction. The results approach normal-dose counterparts, validating OSDM as an effective and practical model to reduce artifacts while preserving image quality.

The OSDM training process

图片e

The pipeline of iterative reconstruction procedure in OSDM

图片f

Reconstruction results from 1e5 noise level using different methods.

图片g

(a) GT (b) FBP (c) SART-TV (d) CNN (e) U-Net (f) NCSN++ (g) OSDM

train: python main.py --config=aapm_sin_ncsnpp_gb.py --workdir=exp --mode=train --eval_folder=result

test: python PCsampling_demo.py

--workdir=exp_zl --mode=train --eval_folder=result --config=aapm_sin_ncsnpp_gb.py

CUDA_VISIBLE_DEVICES=1 python main.py --config=aapm_sin_ncsnpp_gb.py --workdir=exp_zl --mode=train --eval_folder=result

vali[vali < 0] = 0

Other Related Projects

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]
基于深度能量模型的低剂量CT重建
[Paper] [Code]
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]
Low-rank Angular Prior Guided Multi-diffusion Model for Few-shot Low-dose CT 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]
Diffusion Models for Medical Imaging [Paper] [Code] [PPT]

OSDM

Install / Use

README

The OSDM training process

The pipeline of iterative reconstruction procedure in OSDM

Reconstruction results from 1e5 noise level using different methods.

(a) GT (b) FBP (c) SART-TV (d) CNN (e) U-Net (f) NCSN++ (g) OSDM

Other Related Projects