Semi-supervised learned sinogram restoration network for low-dose CT image reconstruction

Meng, Min; Sui, Li; Yao, Lisha; Li, Danyang; Zhu, Manman; Gao, Qi; Xie, Qi; Zhao, Qinglin; Bian, Zhaoying; Huang, Jing; Meng, Deyu; Ma, Jun

doi:10.1117/12.2548985

Cited by 17 publications

(9 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The second approach is sinogram domain data inpainting, which preprocesses a neural network in a few-view sinogram domain and synthesizes it into a complete view sinogram [58][59][60][61][62][63][64][65][66][67][68][69][70][71][72][73][74][75][76]. Applying analytical image reconstruction algorithms such as filtered back projection (FBP) directly to sparse view data will result in poor image quality and serious fringe artifacts.…”

Section: Applications In Different Domainsmentioning

confidence: 99%

“…Semi-supervised learning falls between unsupervised learning (with no labeled training data) and supervised learning (with only labeled training data). Some of semi-supervised and unsupervised studies reviewed are [39,66,74,[127][128][129][130][131][132][133][134][135][136][137][138][139][140]. For example, [129] proposed an unsupervised model-based deep learning (MBDL) for LDCT reconstruction.…”

Section: Applications In Semi-supervised/unsupervised Mannermentioning

confidence: 99%

See 1 more Smart Citation

The use of deep learning methods in low-dose computed tomography image reconstruction: a systematic review

Zhang

Yang

Shi

2022

Complex Intell. Syst.

View full text Add to dashboard Cite

Conventional reconstruction techniques, such as filtered back projection (FBP) and iterative reconstruction (IR), which have been utilised widely in the image reconstruction process of computed tomography (CT) are not suitable in the case of low-dose CT applications, because of the unsatisfying quality of the reconstructed image and inefficient reconstruction time. Therefore, as the demand for CT radiation dose reduction continues to increase, the use of artificial intelligence (AI) in image reconstruction has become a trend that attracts more and more attention. This systematic review examined various deep learning methods to determine their characteristics, availability, intended use and expected outputs concerning low-dose CT image reconstruction. Utilising the methodology of Kitchenham and Charter, we performed a systematic search of the literature from 2016 to 2021 in Springer, Science Direct, arXiv, PubMed, ACM, IEEE, and Scopus. This review showed that algorithms using deep learning technology are superior to traditional IR methods in noise suppression, artifact reduction and structure preservation, in terms of improving the image quality of low-dose reconstructed images. In conclusion, we provided an overview of the use of deep learning approaches in low-dose CT image reconstruction together with their benefits, limitations, and opportunities for improvement.

show abstract

Section: Applications In Different Domainsmentioning

confidence: 99%

Section: Applications In Semi-supervised/unsupervised Mannermentioning

confidence: 99%

The use of deep learning methods in low-dose computed tomography image reconstruction: a systematic review

Zhang

Yang

Shi

2022

Complex Intell. Syst.

View full text Add to dashboard Cite

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%

“…[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. For example, Meng et al 7 proposed a semi-supervised learned sinogram restoration network for low-dose CT image reconstruction. The image-domain DL-based methods are directly applied to the reconstructed CT images.…”

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

“…The sinogram restoration network, which is also utilised for feature distribution and the distribution of sinograms, is employed in the instruction of this topic. A wide range of techniques, including feature classification and high-fidelity sonograms [2,3], may be used to acquire these paired sinograms. The plugand-play network structure has allowed the convolution neural network to demonstrate outstanding performance in image denoising [3].…”

Section: Introductionmentioning

confidence: 99%

Security Protection Mechanism in Cloud Computing Authorization Model Using Machine Learning Techniques

Tiwari

Kannan²,

Veeraiah

et al. 2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

Because of the vast number of applications and the ambiguity in application methods, handwritten character recognition has garnered widespread recognition and increased prominence in the community of pattern recognition researchers ever since it was first developed. This is due to the fact that application methods can be quite ambiguous. Computer in the cloud, on the other hand, allows for suitable network access on demand to a shared pool of customizable computing resources and digital devices. According to those knowledgeable in the subject, the standard filtering techniques are not enough when it comes to the process of denoising images. In many different approaches to machine learning, information is lost not just during the filtering process itself but also at other points during the process. When a convolutional neural network is going through its pooling operation, the internal data representation either becomes misaligned or entirely vanishes (CNN). The reconstruction of low-intensity digital photographs, which takes place during repetitive filtering, breaks away the artefacts that remain after each filtering function, which results in an image that is more uniform. The multilayer wavelet transform, or MLWT, is a method for processing features that comprises of many filter bands and is used in cloud computing authorization that is protected securely. In this scenario, a significant quantity of information gets obliterated from digital photographs during the process of feature extraction and processing. These issues are investigated by the deep learning algorithms that make use of autoencoder, and the methods also handle the novel windowing blocks that are being introduced to the layers. In this section, the magnitude and phase information is considered in order to construct a deep learning framework that will provide good denoising of digital images. The proposed architecture is equipped with the capability of accurately identifying, in real time, the noise level and type that was employed in the training of the network. The method that we have proposed, which is centred on the distribution of noise, may be used to determine the kind of noise. In order to categorise the various types of noise, we investigated nine distinct noise distributions. Dilated convolutional filtering will be used as the method of choice in order to ascertain the specific nature of the noise that can be found in the digital images. An autoencoder-based deep learning algorithm is able to accomplish numerous experimental results in digital image denoising operations that are superior to those achieved by a typical deep learning algorithm even when the intensity of the scenario is low. The performance of the method that is produced by combining the autoencoder and the dilated convolutional filtering techniques is enhanced over the performance of the technology that is currently in use. Using the method that we have outlined in this study, it is possible to recreate the low-intensity images in their entirety. We were able to show that our proposed method beat other existing algorithms for low-density digital photos by comparing several metrics, such as the peak signal-to-noise ratio (PSNR), the structural similarity index, and others (SSIM).

show abstract

Semi-supervised learned sinogram restoration network for low-dose CT image reconstruction

Cited by 17 publications

References 0 publications

The use of deep learning methods in low-dose computed tomography image reconstruction: a systematic review

The use of deep learning methods in low-dose computed tomography image reconstruction: a systematic review

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

Security Protection Mechanism in Cloud Computing Authorization Model Using Machine Learning Techniques

Contact Info

Product

Resources

About