Yanbo Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Yanbo Zhang has a total of 57 co-authors. Here are the top co-authors by number of shared publications:
- Xuanqin Mou
Institute of Image Processing and Pattern Recognition, School of Information and Communications Engineering, Xi’an Jiaotong University, Xi’an 710049, China
13 shared publications - Hengyong Yu
University of Massachusetts Lowell
11 shared publications Shaojie Tang
Xi'an University of Posts and Telecommunications, School of Automation, Xi'an, Shaanxi, 710121, China and Xi'an Jiaotong University, Institute of Image Processing and Pattern Recognition, Xi'an, Shaanxi, 710049, China
5 shared publicationsQiong Xu
Xi'an Jiaotong Univ. (China)
4 shared publications- Fenglin Liu
Chongqing University
3 shared publications - Ge Wang
Rensselaer Polytechnic Institute
3 shared publications
Latest Publications
- Open Access
- 2 January 2026
- Published by Oxford University Press (OUP) in Radiology Advances
- Vol. 3 (1)
Deep learning-based pulmonary nodule risk assessment outperforms established malignancy risk scores in lung cancer screening
Background: Pulmonary nodules are commonly encountered in lung cancer screening. The risk of malignancy varies widely and is generally estimated using expert consensus guidelines (Lung CT Imaging Reporting and Data Systems [Lung-RADS]). Purpose: To assess the performance of a deep learning algorithm (Deep Pulmonary Nodule Profiler [DeepPNP]) for pulmonary nodule malignancy risk estimation in a lung cancer screening dataset and the effect of data enrichment in model training. Materials and Methods: A retrospective analysis was conducted using 3 datasets. DeepPNP is a 3D convolutional network (EfficientNet-B0-based) operating on nodule-centered 3D patches. For the DeepPNP model training and validation, the National Lung Screening Trial (NLST) dataset was combined with 2 independent malignant nodule-only datasets, resulting in a merged dataset of 28 057 nodules, including 2362 malignant nodules. An ablation model (DeepPNP-NLST) was trained on NLST only. The testing was conducted on a held-out dataset from the NLST dataset. Performance metrics, including sensitivity, specificity, precision, F1 score, and accuracy, were analyzed across 3 operating thresholds selected based on specificities of 0.80, 0.85, and 0.90 (selected on the validation set). Benchmarks included Lung-RADS v2022 and the PanCan model. Results: On the NLST test set (including 2597 nodules from 1243 CT scans), DeepPNP achieved an area under the receiver operating characteristic curve (ROC AUC) of 0.96 (95% confidence interval [CI], 0.95-0.97), outperforming Lung-RADS AUC = 0.91 (95% CI, 0.89-0.94; P < .001) and PanCan AUC = 0.93 (95% CI, 0.91-0.95; P < .001). DeepPNP-NLST had an AUC of 0.95 (95% CI, 0.93-0.97; P = .045 vs DeepPNP), indicating a modest gain from positive-only supplementation. Subgroup analyses showed consistent outperformance across nodule sizes and types. Operating-point metrics at 0.80/0.85/0.90 specificity are reported; at 0.80 specificity, DeepPNP achieved sensitivity of 0.94 (100/107; 95% CI, 0.88-0.98) and specificity of 0.88 (2196/2490; 95% CI, 0.87-0.90). Conclusion: DeepPNP outperformed established malignancy risk models in lung cancer screening. The inclusion of biopsy-confirmed malignant nodules from 2 external datasets provided a measurable performance gain, underscoring the importance of data enrichment during model training.
- 2 January 2026
- Published by Oxford University Press (OUP) in Radiology Advances
- Vol. 3 (1)
Open AccessDeep learning-based pulmonary nodule risk assessment outperforms established malignancy risk scores in lung cancer screening
Background: Pulmonary nodules are commonly encountered in lung cancer screening. The risk of malignancy varies widely and is generally estimated using expert consensus guidelines (Lung CT Imaging Reporting and Data Systems [Lung-RADS]). Purpose: To assess the performance of a deep learning algorithm (Deep Pulmonary Nodule Profiler [DeepPNP]) for pulmonary nodule malignancy risk estimation in a lung cancer screening dataset and the effect of data enrichment in model training. Materials and Methods: A retrospective analysis was conducted using 3 datasets. DeepPNP is a 3D convolutional network (EfficientNet-B0-based) operating on nodule-centered 3D patches. For the DeepPNP model training and validation, the National Lung Screening Trial (NLST) dataset was combined with 2 independent malignant nodule-only datasets, resulting in a merged dataset of 28 057 nodules, including 2362 malignant nodules. An ablation model (DeepPNP-NLST) was trained on NLST only. The testing was conducted on a held-out dataset from the NLST dataset. Performance metrics, including sensitivity, specificity, precision, F1 score, and accuracy, were analyzed across 3 operating thresholds selected based on specificities of 0.80, 0.85, and 0.90 (selected on the validation set). Benchmarks included Lung-RADS v2022 and the PanCan model. Results: On the NLST test set (including 2597 nodules from 1243 CT scans), DeepPNP achieved an area under the receiver operating characteristic curve (ROC AUC) of 0.96 (95% confidence interval [CI], 0.95-0.97), outperforming Lung-RADS AUC = 0.91 (95% CI, 0.89-0.94; P < .001) and PanCan AUC = 0.93 (95% CI, 0.91-0.95; P < .001). DeepPNP-NLST had an AUC of 0.95 (95% CI, 0.93-0.97; P = .045 vs DeepPNP), indicating a modest gain from positive-only supplementation. Subgroup analyses showed consistent outperformance across nodule sizes and types. Operating-point metrics at 0.80/0.85/0.90 specificity are reported; at 0.80 specificity, DeepPNP achieved sensitivity of 0.94 (100/107; 95% CI, 0.88-0.98) and specificity of 0.88 (2196/2490; 95% CI, 0.87-0.90). Conclusion: DeepPNP outperformed established malignancy risk models in lung cancer screening. The inclusion of biopsy-confirmed malignant nodules from 2 external datasets provided a measurable performance gain, underscoring the importance of data enrichment during model training.
- Open Access
Tensor Dictionary Learning with an Enhanced Sparsity Constraint for Sparse-View Spectral CT Reconstruction
Spectral computed tomography (CT) can divide collected photons into multi-energy channels and gain multi-channel projections synchronously by using photon-counting detectors. However, reconstructed images usually contain severe noise due to the limited number of photons in the corresponding energy channel. Tensor dictionary learning (TDL)-based methods have achieved better performance, but usually lose image edge information and details, especially from an under-sampling dataset. To address this problem, this paper proposes a method termed TDL with an enhanced sparsity constraint for spectral CT reconstruction. The proposed algorithm inherits the superiority of TDL by exploring the correlation of spectral CT images. Moreover, the method designs a regularization using the L0-norm of the image gradient to constrain images and the difference between images and a prior image in each energy channel simultaneously, further improving the ability to preserve edge information and subtle image details. The split-Bregman algorithm has been applied to address the proposed objective minimization model. Several numerical simulations and realistic preclinical mice are studied to assess the effectiveness of the proposed algorithm. The results demonstrate that the proposed method improves the quality of spectral CT images in terms of noise elimination, edge preservation, and image detail recovery compared to the several existing better methods.
Open AccessTensor Dictionary Learning with an Enhanced Sparsity Constraint for Sparse-View Spectral CT Reconstruction
Spectral computed tomography (CT) can divide collected photons into multi-energy channels and gain multi-channel projections synchronously by using photon-counting detectors. However, reconstructed images usually contain severe noise due to the limited number of photons in the corresponding energy channel. Tensor dictionary learning (TDL)-based methods have achieved better performance, but usually lose image edge information and details, especially from an under-sampling dataset. To address this problem, this paper proposes a method termed TDL with an enhanced sparsity constraint for spectral CT reconstruction. The proposed algorithm inherits the superiority of TDL by exploring the correlation of spectral CT images. Moreover, the method designs a regularization using the L0-norm of the image gradient to constrain images and the difference between images and a prior image in each energy channel simultaneously, further improving the ability to preserve edge information and subtle image details. The split-Bregman algorithm has been applied to address the proposed objective minimization model. Several numerical simulations and realistic preclinical mice are studied to assess the effectiveness of the proposed algorithm. The results demonstrate that the proposed method improves the quality of spectral CT images in terms of noise elimination, edge preservation, and image detail recovery compared to the several existing better methods.
- 5 December 2020
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 65 (24) , 245006
Dictionary learning based image-domain material decomposition for spectral CT
The potential huge advantage of spectral computed tomography (CT) is that it can provide accurate material identification and quantitative tissue information by material decomposition. However, material decomposition is a typical inverse problem, where the noise can be magnified. To address this issue, we develop a dictionary learning based image-domain material decomposition (DLIMD) method for spectral CT to achieve accurate material components with better image quality. Specifically, a set of image patches are extracted from the mode-1 unfolding of normalized material images decomposed by direct inversion to train a unified dictionary using the K-SVD technique. Then, the DLIMD model is established to explore the redundant similarities of the material images, where the split-Bregman is employed to optimize the model. Finally, more constraints (i.e. volume conservation and the bounds of each pixel within material maps) are integrated into the DLIMD model. Numerical phantom, physical phantom and preclinical experiments are performed to evaluate the performance of the proposed DLIMD in material decomposition accuracy, material image edge preservation and feature recovery.
- 5 December 2020
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 65 (24) , 245006
Dictionary learning based image-domain material decomposition for spectral CT
The potential huge advantage of spectral computed tomography (CT) is that it can provide accurate material identification and quantitative tissue information by material decomposition. However, material decomposition is a typical inverse problem, where the noise can be magnified. To address this issue, we develop a dictionary learning based image-domain material decomposition (DLIMD) method for spectral CT to achieve accurate material components with better image quality. Specifically, a set of image patches are extracted from the mode-1 unfolding of normalized material images decomposed by direct inversion to train a unified dictionary using the K-SVD technique. Then, the DLIMD model is established to explore the redundant similarities of the material images, where the split-Bregman is employed to optimize the model. Finally, more constraints (i.e. volume conservation and the bounds of each pixel within material maps) are integrated into the DLIMD model. Numerical phantom, physical phantom and preclinical experiments are performed to evaluate the performance of the proposed DLIMD in material decomposition accuracy, material image edge preservation and feature recovery.
- Open Access
- 20 July 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 133367-133376
Spectral CT Reconstruction Based on PICCS and Dictionary Learning
Photon-counting detector based spectral computed tomography (CT) can obtain energy-discriminative attenuation map of an object in different energy channels, extending the conventional volumetric image along a spectral dimension. However, compared with the full spectrum data, the noise in a narrower energy channel is significantly increased. In order to improve image quality of spectral CT images, this paper proposes an iterative reconstruction algorithm based on the prior image constrained compressed sensing (PICCS) and dictionary learning (DL) theories, which is called PICCS-DL. The PICCS-DL utilizes the correlation of the images reconstructed from different energy channels by taking the broad spectrum image as a prior constraint, and it utilizes the sparse of the images by taking the total variation (TV) and DL as prior constraints. The alternating minimization, Split-Bregman and the steepest descent (SD) methods are used to solve the objective function. The effectiveness of the proposed method is validated with numerical simulations and preclinical applications. The results demonstrate that the proposed algorithm generally produces superior image quality, especially for noisy and sparse projection data.
- 20 July 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 133367-133376
Open AccessSpectral CT Reconstruction Based on PICCS and Dictionary Learning
Photon-counting detector based spectral computed tomography (CT) can obtain energy-discriminative attenuation map of an object in different energy channels, extending the conventional volumetric image along a spectral dimension. However, compared with the full spectrum data, the noise in a narrower energy channel is significantly increased. In order to improve image quality of spectral CT images, this paper proposes an iterative reconstruction algorithm based on the prior image constrained compressed sensing (PICCS) and dictionary learning (DL) theories, which is called PICCS-DL. The PICCS-DL utilizes the correlation of the images reconstructed from different energy channels by taking the broad spectrum image as a prior constraint, and it utilizes the sparse of the images by taking the total variation (TV) and DL as prior constraints. The alternating minimization, Split-Bregman and the steepest descent (SD) methods are used to solve the objective function. The effectiveness of the proposed method is validated with numerical simulations and preclinical applications. The results demonstrate that the proposed algorithm generally produces superior image quality, especially for noisy and sparse projection data.
- 26 March 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 39 (10) , 2996-3007
Spectral CT Reconstruction via Low-Rank Representation and Region-Specific Texture Preserving Markov Random Field Regularization
Photon-counting spectral computed tomography (CT) is capable of material characterization and can improve diagnostic performance over traditional clinical CT. However, it suffers from photon count starving for each individual energy channel which may cause severe artifacts in the reconstructed images. Furthermore, since the images in different energy channels describe the same object, there are high correlations among different channels. To make full use of the inter-channel correlations and minimize the count starving effect while maintaining clinically meaningful texture information, this paper combines a region-specific texture model with a low-rank correlation descriptor as an a priori regularization to explore a superior texture preserving Bayesian reconstruction of spectral CT. Specifically, the inter-channel correlations are characterized by the low-rank representation, and the inner-channel regional textures are modeled by a texture preserving Markov random field. In other words, this paper integrates the spectral and spatial information into a unified Bayesian reconstruction framework. The widely-used Split-Bregman algorithm is employed to minimize the objective function because of the non-differentiable property of the low-rank representation. To evaluate the tissue texture preserving performance of the proposed method for each channel, three references are built for comparison: one is the traditional CT image from energy integration detection. The second one is spectral images from dual-energy CT. The third one is individual channels images from custom-made photon-counting spectral CT. As expected, the proposed method produced promising results in terms of not only preserving texture features but also suppressing image noise in each channel, comparing to existing methods of total variation (TV), low-rank TV and tensor dictionary learning, by both visual inspection and quantitative indexes of root mean square error, peak signal to noise ratio, structural similarity and feature similarity.
- 26 March 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 39 (10) , 2996-3007
Spectral CT Reconstruction via Low-Rank Representation and Region-Specific Texture Preserving Markov Random Field Regularization
Photon-counting spectral computed tomography (CT) is capable of material characterization and can improve diagnostic performance over traditional clinical CT. However, it suffers from photon count starving for each individual energy channel which may cause severe artifacts in the reconstructed images. Furthermore, since the images in different energy channels describe the same object, there are high correlations among different channels. To make full use of the inter-channel correlations and minimize the count starving effect while maintaining clinically meaningful texture information, this paper combines a region-specific texture model with a low-rank correlation descriptor as an a priori regularization to explore a superior texture preserving Bayesian reconstruction of spectral CT. Specifically, the inter-channel correlations are characterized by the low-rank representation, and the inner-channel regional textures are modeled by a texture preserving Markov random field. In other words, this paper integrates the spectral and spatial information into a unified Bayesian reconstruction framework. The widely-used Split-Bregman algorithm is employed to minimize the objective function because of the non-differentiable property of the low-rank representation. To evaluate the tissue texture preserving performance of the proposed method for each channel, three references are built for comparison: one is the traditional CT image from energy integration detection. The second one is spectral images from dual-energy CT. The third one is individual channels images from custom-made photon-counting spectral CT. As expected, the proposed method produced promising results in terms of not only preserving texture features but also suppressing image noise in each channel, comparing to existing methods of total variation (TV), low-rank TV and tensor dictionary learning, by both visual inspection and quantitative indexes of root mean square error, peak signal to noise ratio, structural similarity and feature similarity.
- 26 October 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 38 (4) , 1079-1093
Non-Local Low-Rank Cube-Based Tensor Factorization for Spectral CT Reconstruction
Spectral computed tomography (CT) reconstructs material-dependent attenuation images from the projections of multiple narrow energy windows which is meaningful for material identification and decomposition. Unfortunately, the multi-energy projection datasets usually have lower signal-noise-ratios (SNR). Very recently, a spatial-spectral cube matching frame (SSCMF) was proposed to explore the non-local spatial-spectral similarities for spectral CT. This method constructs a group by clustering up a series of non-local spatial-spectral cubes. The small size of spatial patches for such a group makes the SSCMF fail to fully encode the sparsity and low-rank properties. The hard-thresholding and collaboration filtering in the SSCMF also cause difficulty in recovering the image features and spatial edges. While all the steps are operated on 4-D group, the huge computational cost and memory load might not be affordable in practice. To avoid the above limitations and further improve image quality, we first formulate a non-local cube-based tensor instead of group to encode the sparsity and low-rank properties. Then, as a new regularizer, the Kronecker- Basis-Representation (KBR) tensor factorization is employed into a basic spectral CT reconstruction model to enhance the capability of image feature extraction and spatial edge preservation, generating a non-local low-rank cube-based tensor factorization (NLCTF) method. Finally, the split-Bregman method is adopted to solve the NLCTF model. Both numerical simulations and preclinical mouse studies are performed to validate and evaluate the NLCTF algorithm. The results show that the NLCTF method outperforms other state-of-the-art competing algorithms.
- 26 October 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 38 (4) , 1079-1093
Non-Local Low-Rank Cube-Based Tensor Factorization for Spectral CT Reconstruction
Spectral computed tomography (CT) reconstructs material-dependent attenuation images from the projections of multiple narrow energy windows which is meaningful for material identification and decomposition. Unfortunately, the multi-energy projection datasets usually have lower signal-noise-ratios (SNR). Very recently, a spatial-spectral cube matching frame (SSCMF) was proposed to explore the non-local spatial-spectral similarities for spectral CT. This method constructs a group by clustering up a series of non-local spatial-spectral cubes. The small size of spatial patches for such a group makes the SSCMF fail to fully encode the sparsity and low-rank properties. The hard-thresholding and collaboration filtering in the SSCMF also cause difficulty in recovering the image features and spatial edges. While all the steps are operated on 4-D group, the huge computational cost and memory load might not be affordable in practice. To avoid the above limitations and further improve image quality, we first formulate a non-local cube-based tensor instead of group to encode the sparsity and low-rank properties. Then, as a new regularizer, the Kronecker- Basis-Representation (KBR) tensor factorization is employed into a basic spectral CT reconstruction model to enhance the capability of image feature extraction and spatial edge preservation, generating a non-local low-rank cube-based tensor factorization (NLCTF) method. Finally, the split-Bregman method is adopted to solve the NLCTF model. Both numerical simulations and preclinical mouse studies are performed to validate and evaluate the NLCTF algorithm. The results show that the NLCTF method outperforms other state-of-the-art competing algorithms.
- 14 August 2018
- Published by IOP Publishing in Inverse Problems
- Vol. 34 (10) , 104003
Spatial-spectral cube matching frame for spectral CT reconstruction
Spectral computed tomography (CT) reconstructs the same scanned object from projections of multiple narrow energy windows, and it can be used for material identification and decomposition. However, the multi-energy projection dataset has a lower signal-noise-ratio (SNR), resulting in poor reconstructed image quality. To address this thorny problem, we develop a spectral CT reconstruction method, namely spatial-spectral cube matching frame (SSCMF). This method is inspired by the following three facts: (i) human body usually consists of two or three basic materials implying that the reconstructed spectral images have a strong sparsity; (ii) the same basic material component in a single channel image has similar intensity and structures in local regions. Different material components within the same energy channel share similar structural information; (iii) multi-energy projection datasets are collected from the subject by using different narrow energy windows, which means images reconstructed from different energy-channels share similar structures. To explore those information, we first establish a tensor cube matching frame (CMF) for a BM4D denoising procedure. Then, as a new regularizer, the CMF is introduced into a basic spectral CT reconstruction model, generating the SSCMF method. Because the SSCMF model contains an L 0 -norm minimization of 4D transform coefficients, an effective strategy is employed for optimization. Both numerical simulations and realistic preclinical mouse studies are performed. The results show that the SSCMF method outperforms the state-of-the-art algorithms, including the simultaneous algebraic reconstruction technique, total variation minimization, total variation plus low rank, and tensor dictionary learning.
- 14 August 2018
- Published by IOP Publishing in Inverse Problems
- Vol. 34 (10) , 104003
Spatial-spectral cube matching frame for spectral CT reconstruction
Spectral computed tomography (CT) reconstructs the same scanned object from projections of multiple narrow energy windows, and it can be used for material identification and decomposition. However, the multi-energy projection dataset has a lower signal-noise-ratio (SNR), resulting in poor reconstructed image quality. To address this thorny problem, we develop a spectral CT reconstruction method, namely spatial-spectral cube matching frame (SSCMF). This method is inspired by the following three facts: (i) human body usually consists of two or three basic materials implying that the reconstructed spectral images have a strong sparsity; (ii) the same basic material component in a single channel image has similar intensity and structures in local regions. Different material components within the same energy channel share similar structural information; (iii) multi-energy projection datasets are collected from the subject by using different narrow energy windows, which means images reconstructed from different energy-channels share similar structures. To explore those information, we first establish a tensor cube matching frame (CMF) for a BM4D denoising procedure. Then, as a new regularizer, the CMF is introduced into a basic spectral CT reconstruction model, generating the SSCMF method. Because the SSCMF model contains an L 0 -norm minimization of 4D transform coefficients, an effective strategy is employed for optimization. Both numerical simulations and realistic preclinical mouse studies are performed. The results show that the SSCMF method outperforms the state-of-the-art algorithms, including the simultaneous algebraic reconstruction technique, total variation minimization, total variation plus low rank, and tensor dictionary learning.
- 1 August 2018
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 63 (15) , 155021
Iterative spectral CT reconstruction based on low rank and average-image-incorporated BM3D
In a photon counting detector integrated spectral CT scanner, the received photons are counted in several energy channels to generate the corresponding projections. Since the projection in each energy channel is generated using part of the received photons, the reconstructed channel image suffers from severe noise. Therefore, image reconstruction in spectral CT is considered to be a big challenge. Because the inter-channel images are all from the same object but in different energy bins, there exists a strong correlation among these images. Moreover, it is suggested that there are similarities among various patches of CT images in the spatial domain. In this work, we propose average-image-incorporated block-matching and 3D (aiiBM3D) filtering along with low rank regularization for iterative spectral CT reconstruction. The aiiBM3D method is based on filtered 3D data arrays formed by similar 2D blocks using the mapped version of the average image obtained from linear regression. The reconstruction procedure consists of two main steps. First, the alternating direction method of multipliers is employed to solve the problem with low rank regularization where the goal is to exploit the correlation in inter-channel images. Second, our proposed BM3D-based algorithm is applied to all the channel images to make use of the redundant information in the spatial domain and inter-channel. The two steps repeat until the stopping criteria are satisfied. The proposed method is validated on numerically simulated and preclinical datasets. Our results confirm its high performance in terms of signal to noise ratio and structural preservation.
- 1 August 2018
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 63 (15) , 155021
Iterative spectral CT reconstruction based on low rank and average-image-incorporated BM3D
In a photon counting detector integrated spectral CT scanner, the received photons are counted in several energy channels to generate the corresponding projections. Since the projection in each energy channel is generated using part of the received photons, the reconstructed channel image suffers from severe noise. Therefore, image reconstruction in spectral CT is considered to be a big challenge. Because the inter-channel images are all from the same object but in different energy bins, there exists a strong correlation among these images. Moreover, it is suggested that there are similarities among various patches of CT images in the spatial domain. In this work, we propose average-image-incorporated block-matching and 3D (aiiBM3D) filtering along with low rank regularization for iterative spectral CT reconstruction. The aiiBM3D method is based on filtered 3D data arrays formed by similar 2D blocks using the mapped version of the average image obtained from linear regression. The reconstruction procedure consists of two main steps. First, the alternating direction method of multipliers is employed to solve the problem with low rank regularization where the goal is to exploit the correlation in inter-channel images. Second, our proposed BM3D-based algorithm is applied to all the channel images to make use of the redundant information in the spatial domain and inter-channel. The two steps repeat until the stopping criteria are satisfied. The proposed method is validated on numerically simulated and preclinical datasets. Our results confirm its high performance in terms of signal to noise ratio and structural preservation.
- Top Cited Papers
- 17 April 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 37 (6) , 1348-1357
Low-Dose CT Image Denoising Using a Generative Adversarial Network With Wasserstein Distance and Perceptual Loss
The continuous development and extensive use of computed tomography (CT) in medical practice has raised a public concern over the associated radiation dose to the patient. Reducing the radiation dose may lead to increased noise and artifacts, which can adversely affect the radiologists' judgment and confidence. Hence, advanced image reconstruction from low-dose CT data is needed to improve the diagnostic performance, which is a challenging problem due to its ill-posed nature. Over the past years, various low-dose CT methods have produced impressive results. However, most of the algorithms developed for this application, including the recently popularized deep learning techniques, aim for minimizing the mean-squared error (MSE) between a denoised CT image and the ground truth under generic penalties. Although the peak signal-to-noise ratio is improved, MSE- or weighted-MSE-based methods can compromise the visibility of important structural details after aggressive denoising. This paper introduces a new CT image denoising method based on the generative adversarial network (GAN) with Wasserstein distance and perceptual similarity. The Wasserstein distance is a key concept of the optimal transport theory and promises to improve the performance of GAN. The perceptual loss suppresses noise by comparing the perceptual features of a denoised output against those of the ground truth in an established feature space, while the GAN focuses more on migrating the data noise distribution from strong to weak statistically. Therefore, our proposed method transfers our knowledge of visual perception to the image denoising task and is capable of not only reducing the image noise level but also trying to keep the critical information at the same time. Promising results have been obtained in our experiments with clinical CT images.
- 17 April 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 37 (6) , 1348-1357
Top Cited PapersLow-Dose CT Image Denoising Using a Generative Adversarial Network With Wasserstein Distance and Perceptual Loss
The continuous development and extensive use of computed tomography (CT) in medical practice has raised a public concern over the associated radiation dose to the patient. Reducing the radiation dose may lead to increased noise and artifacts, which can adversely affect the radiologists' judgment and confidence. Hence, advanced image reconstruction from low-dose CT data is needed to improve the diagnostic performance, which is a challenging problem due to its ill-posed nature. Over the past years, various low-dose CT methods have produced impressive results. However, most of the algorithms developed for this application, including the recently popularized deep learning techniques, aim for minimizing the mean-squared error (MSE) between a denoised CT image and the ground truth under generic penalties. Although the peak signal-to-noise ratio is improved, MSE- or weighted-MSE-based methods can compromise the visibility of important structural details after aggressive denoising. This paper introduces a new CT image denoising method based on the generative adversarial network (GAN) with Wasserstein distance and perceptual similarity. The Wasserstein distance is a key concept of the optimal transport theory and promises to improve the performance of GAN. The perceptual loss suppresses noise by comparing the perceptual features of a denoised output against those of the ground truth in an established feature space, while the GAN focuses more on migrating the data noise distribution from strong to weak statistically. Therefore, our proposed method transfers our knowledge of visual perception to the image denoising task and is capable of not only reducing the image noise level but also trying to keep the critical information at the same time. Promising results have been obtained in our experiments with clinical CT images.
- Top Cited Papers
- 4 April 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 37 (6) , 1370-1381
Convolutional Neural Network Based Metal Artifact Reduction in X-Ray Computed Tomography
In the presence of metal implants, metal artifacts are introduced to x-ray computed tomography CT images. Although a large number of metal artifact reduction (MAR) methods have been proposed in the past decades, MAR is still one of the major problems in clinical x-ray CT. In this paper, we develop a convolutional neural network (CNN)-based open MAR framework, which fuses the information from the original and corrected images to suppress artifacts. The proposed approach consists of two phases. In the CNN training phase, we build a database consisting of metal-free, metal-inserted and pre-corrected CT images, and image patches are extracted and used for CNN training. In the MAR phase, the uncorrected and pre-corrected images are used as the input of the trained CNN to generate a CNN image with reduced artifacts. To further reduce the remaining artifacts, water equivalent tissues in a CNN image are set to a uniform value to yield a CNN prior, whose forward projections are used to replace the metal-affected projections, followed by the FBP reconstruction. The effectiveness of the proposed method is validated on both simulated and real data. Experimental results demonstrate the superior MAR capability of the proposed method to its competitors in terms of artifact suppression and preservation of anatomical structures in the vicinity of metal implants.
- 4 April 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 37 (6) , 1370-1381
Top Cited PapersConvolutional Neural Network Based Metal Artifact Reduction in X-Ray Computed Tomography
In the presence of metal implants, metal artifacts are introduced to x-ray computed tomography CT images. Although a large number of metal artifact reduction (MAR) methods have been proposed in the past decades, MAR is still one of the major problems in clinical x-ray CT. In this paper, we develop a convolutional neural network (CNN)-based open MAR framework, which fuses the information from the original and corrected images to suppress artifacts. The proposed approach consists of two phases. In the CNN training phase, we build a database consisting of metal-free, metal-inserted and pre-corrected CT images, and image patches are extracted and used for CNN training. In the MAR phase, the uncorrected and pre-corrected images are used as the input of the trained CNN to generate a CNN image with reduced artifacts. To further reduce the remaining artifacts, water equivalent tissues in a CNN image are set to a uniform value to yield a CNN prior, whose forward projections are used to replace the metal-affected projections, followed by the FBP reconstruction. The effectiveness of the proposed method is validated on both simulated and real data. Experimental results demonstrate the superior MAR capability of the proposed method to its competitors in terms of artifact suppression and preservation of anatomical structures in the vicinity of metal implants.