Noise-Generating-Mechanism-Driven Unsupervised Learning for Low-Dose CT Sinogram Recovery

Wang, Lei; Geng, Mingrui; Li, Sui; Deng, Yun; Xie, Qi; Li, Danyang; Zhang, Hua; Li, Yuanqing; Xu, Zongben; Meng, Deyu; Ma, Jianhua

doi:10.1109/trpms.2021.3083361

Cited by 15 publications

(16 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, the modification of network architecture was used in our experiments to better combine high-dimensional information for LDCT reconstruction. On the other hand, it is different from the NGM-driven DL mode in [30], which considers the noise generation mechanism in CT imaging and combines data-driven and model-driven, EASEL uses a generative model where samples were produced via Langevin dynamics using gradients of the data distribution estimated with DSM was utilized for LDCT reconstruction. In addition, the distance between the reconstructed images and the learned manifold was minimized along with the data fidelity by the SQS algorithm during iterative reconstruction.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

“…For example, Adler and Öktem [29] employed a Wasserstein GAN to draw samples from the conditional distribution. Zeng et al [30] considered the noise-generating mechanism in CT imaging and using pairs of noisy images to reduce the noise. Regretfully, despite the success of generative models in tackling natural images, there have been few studies in the field of medical imaging, especially in LDCT.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Iterative Reconstruction for Low-Dose CT Using Deep Gradient Priors of Generative Model

He¹,

Zhang²,

Guan³

et al. 2022

IEEE Trans. Radiat. Plasma Med. Sci.

View full text Add to dashboard Cite

Dose reduction in computed tomography (CT) is essential for decreasing radiation risk in clinical applications. Iterative reconstruction is one of the most promising ways to compensate for the increased noise due to the reduction of photon flux. Rather than most existing prior-driven algorithms that benefit from manually designed prior functions or supervised learning schemes, in this work, we integrate the data consistency as a conditional term into the iterative generative model for lowdose CT. At the stage of prior learning, the gradient of data density is directly learned from normal-dose CT images as a prior. Then, at the iterative reconstruction stage, the stochastic gradient descent is employed to update the trained prior with annealed and conditional schemes. The distance between the reconstructed image and the manifold is minimized along with data fidelity during reconstruction. Experimental comparisons demonstrated the noise reduction and detail preservation abilities of the proposed method.

show abstract

Section: Conclusion and Discussionmentioning

confidence: 99%

mentioning

confidence: 99%

Iterative Reconstruction for Low-Dose CT Using Deep Gradient Priors of Generative Model

He¹,

Zhang²,

Guan³

et al. 2022

IEEE Trans. Radiat. Plasma Med. Sci.

View full text Add to dashboard Cite

show abstract

“…Bai et al [59] used probabilistic characteristics of denoising and Zhang et al [60] utilized adjacent CT slices. Zeng et al [61] proposed a LDCT sinogram recovery strategy based on unsupervised learning using the noise generation mechanism of CT measurement in the sinogram. Kim et al [62] also proposed a two-step-based LDCT denoising framework using multi-slice input to predict a center slice with improved objective and perceptual quality.…”

Section: Introductionmentioning

confidence: 99%

“…Two studies [59], [60] removed noise using only CT images; however, Bai et al [59] depended on the Gaussian noise distribution, and Zhang et al [60] focused more on noise reduction by training one noise realization to map to its two adjacent LDCT images. In Zeng et al's work [61], the noise generation mechanism of CT measurement relies on a specific noise model, and denoising performance may not be guaranteed for systems with different noise characteristics. Kim et al [62] needed unpaired clean NDCT images and relied on heavy self-attention [63].…”

Section: Introductionmentioning

confidence: 99%

MM-Net: Multiframe and Multimask-Based Unsupervised Deep Denoising for Low-Dose Computed Tomography

Jeon

Kim

Choi

2023

IEEE Trans. Radiat. Plasma Med. Sci.

View full text Add to dashboard Cite

Low-dose computed tomography (LDCT) is crucial due to the risk of radiation exposure to patients. However, the high noise level in LDCT images may reduce the image quality, leading to a less accurate diagnosis. Deep learning technology, especially supervised methods, has recently been widely accepted as a powerful tool for LDCT image denoising tasks. However, supervised methods require numerous paired datasets of LDCT and high-quality pristine CT images, which are rarely available in real-world clinical scenarios. This study presents an unsupervised learning-based framework called MM-Net, consisting of two training steps for a volumetric LDCT denoising task. In the two-step training approach, we first train the initial denoising network multi-scale attention U-Net (MSAU-Net) in a self-supervised manner to predict the noise-suppressed center slice with a neighboring multi-slice input. The second training step aims to train the U-Net-based final denoiser based on the pre-trained MSAU-Net to improve the image quality by introducing new multi-patch and multi-mask matching loss. Qualitative visual inspection and quantitative measures across texturally different domains of clinical and animal data reveal that the proposed MM-Net outperformed all competing state-ofthe-art unsupervised algorithms. The unsupervised method also achieved denoising performance comparable to the representative supervised methods trained with ground truth images.

show abstract

“…[2][3][4] Among them, the deep learning (DL)-based CT reconstruction methods have been widely developed, and achieved superior performance than the statistical iterative reconstruction methods. [5][6][7][8][9][10][11][12][13][14] The deep learning methods can be generally classified into categories, such as sinogram-domain DL-based methods, [5][6][7] image-domain DL-based methods [8][9][10] and dual-domain DL-based methods. [11][12][13][14] The sinogram-domain DL-based methods directly suppress noise in the sinogram domain, and then reconstruct the CT images from the filtered sinogram.…”

Section: Introductionmentioning

confidence: 99%

Dual-domain modulation for high-performance multi-geometry low-dose CT image reconstruction

Chen

et al. 2023

Medical Imaging 2023: Physics of Medical Imaging

View full text Add to dashboard Cite

Computed Tomography (CT) is one of the most used imaging modality in clinics. However, the associated high radiation is a major concern, and then the tradeoff between clinically accepted CT images and radiation dose is desired. In recent years, inspired by deep learning techniques, various deep learning-based methods have been developed for low-dose CT imaging, and they have potential to improve image quality of low dose CT. Meanwhile, most of these methods are trained on the CT datasets with the specific imaging geometry, and they could fail to successfully reconstruct the CT images from the other imaging geometries simultaneously, which might lead to poor generalization ability. In this work, to address this issue, we propose a dual-domain modulation for high-performance multi-geometry low-dose CT image reconstruction (DM-MG) method. Specific, the proposed DM-MG consists of two sub-networks, i.e., multilayer perceptron controlling the multiple imaging geometries, and inverse Radon transform approximation sub-network reconstructing the CT images from the different geometries. Moreover, the inverse Radon transform approximation sub-network contains sinogram-domain filtering module, back-projection module, and image-domain filtering module, which maps the Radon transformation to the network. Finally, the proposed DM-MG can reconstruct the CT images from the different imaging geometries and dose levels simultaneously. In this work, the CT data from 2016 NIHAAPM-Mayo Clinic Low Dose CT Grand Challenge are used to validate and evaluate the reconstruction performance of the proposed DM-MG. Experimental results demonstrate that the presented DM-MG method can produce high-quality CT images from multiple geometries simultaneously and outperform the competing method in the quantitative assessments and visual inspection show that the dual-domain modulated model we propose provides better reconstruction of low-dose CT images under different imaging geometries compared to other modulated models.

show abstract

Noise-Generating-Mechanism-Driven Unsupervised Learning for Low-Dose CT Sinogram Recovery

Cited by 15 publications

References 54 publications

Iterative Reconstruction for Low-Dose CT Using Deep Gradient Priors of Generative Model

Iterative Reconstruction for Low-Dose CT Using Deep Gradient Priors of Generative Model

MM-Net: Multiframe and Multimask-Based Unsupervised Deep Denoising for Low-Dose Computed Tomography

Dual-domain modulation for high-performance multi-geometry low-dose CT image reconstruction

Contact Info

Product

Resources

About