Top Download
1
2022, 31(1): 117-134.
DOI: 10.15953/j.1004-4140.2022.31.01.14
Abstract：
Computed tomography (CT) is widely used in clinical diagnosis because of its fast imaging speed and high resolution. However, higher doses of radiation will cause damages to human tissues and organs, while lower doses will lead to serious deterioration of imaging quality. In order to solve the above contradiction, researchers have focused on the low-dose CT imaging technology to study how to reduce the harm caused by radiation to the human body to the greatest extent under the condition of ensuring the imaging quality to meet the needs of clinical diagnosis. In recent years, deep learning has developed rapidly in the field of artificial intelligence, and has been widely used in image processing, pattern recognition, signal processing fields. Driven by big data, LDCT imaging algorithms based on deep learning have made great progress. This paper studies the development of low-dose CT imaging algorithms in recent years in terms of three aspects: the process of CT imaging, the noise modeling of low-dose CT, and the design of imaging algorithms. In particular, the imaging algorithms in the field of deep learning are systematically elaborated and analyzed. Finally, future developments in the field of LDCT image artifact suppression are also prospected.
3
2023, 32(4): 480-486.
DOI: 10.15953/j.ctta.2022.109
Abstract：
Objective: To explore the influence of different CT acquisition and reconstruction parameters on the CT number of the chest in vivo. Methods: The CT number of the trachea, blood vessels, lungs, vertebral bodies, and muscles of the human chest were measured under different CT scanning parameters. Six groups of different scanning parameters and reconstruction algorithms were set respectively: slice thickness 5 mm, 50% multi-model adaptive statistical iterative reconstruction Veo (ASIR-V) and low-dose for S1; slice thickness 5 mm, filtered back projection (FBP) and standard-dose for S2; slice thickness 1.25 mm, 50% ASIR-V and low-dose for S3; slice thickness 1.25 mm, 50% ASIR-V and standard-dose for S4; slice thickness 1.25 mm, FBP, low-dose for S5; slice thickness 1.25 mm, FBP, standard-dose for S6. The radiation dose of the scan was controlled using two noise indexes (NI), including low-dose (NI=40) and standard-dose (NI=10). Differences in CT number between two groups were compared using t-test or rank-sum test. Results: Significant differences of CT number of the trachea were detected between low-dose and standard-dose, but no significant differences of CT number of other tissues were detected between low-dose and standard-dose. No significant differences of CT number of chest tissues were detected between either 5 mm thickness and 1.25 mm thickness or 50% ASIR-V and FBP. Conclusion: The CT number of human chest tissues showed well stability which was scarcely influenced by slices thickness, reconstruction algorithm and scan dose.
Objective: To explore the influence of different CT acquisition and reconstruction parameters on the CT number of the chest in vivo. Methods: The CT number of the trachea, blood vessels, lungs, vertebral bodies, and muscles of the human chest were measured under different CT scanning parameters. Six groups of different scanning parameters and reconstruction algorithms were set respectively: slice thickness 5 mm, 50% multi-model adaptive statistical iterative reconstruction Veo (ASIR-V) and low-dose for S1; slice thickness 5 mm, filtered back projection (FBP) and standard-dose for S2; slice thickness 1.25 mm, 50% ASIR-V and low-dose for S3; slice thickness 1.25 mm, 50% ASIR-V and standard-dose for S4; slice thickness 1.25 mm, FBP, low-dose for S5; slice thickness 1.25 mm, FBP, standard-dose for S6. The radiation dose of the scan was controlled using two noise indexes (NI), including low-dose (NI=40) and standard-dose (NI=10). Differences in CT number between two groups were compared using t-test or rank-sum test. Results: Significant differences of CT number of the trachea were detected between low-dose and standard-dose, but no significant differences of CT number of other tissues were detected between low-dose and standard-dose. No significant differences of CT number of chest tissues were detected between either 5 mm thickness and 1.25 mm thickness or 50% ASIR-V and FBP. Conclusion: The CT number of human chest tissues showed well stability which was scarcely influenced by slices thickness, reconstruction algorithm and scan dose.
4
2022, 31(6): 697-707.
DOI: 10.15953/j.ctta.2022.114
Abstract：
Under the condition that the number of projection angles is constant, reducing the radiation dose under each angle is an effective way to realize low-dose CT. However, the reconstructed images obtained through this method can be very noisy. At present, the deep learning image denoising method represented by convolutional neural networks (CNN) has become a classical method for low-dose CT image denoising. Inspired by the good performance of transformer in computer vision tasks, this paper proposes a CNN transformer coupling network (CTC) to further improve the performance of CT image denoising. CTC network makes comprehensive use of local information association ability of CNN and global information capture ability of transformer, constructs eight core network blocks composed of CNN components and an improved transformer component, which are interconnected based on residual connection mechanism and information reuse mechanism. Compared with the existing four denoising networks, CTC network demonstrate better denoising ability and can realize high-precision low-dose CT image reconstruction.
Under the condition that the number of projection angles is constant, reducing the radiation dose under each angle is an effective way to realize low-dose CT. However, the reconstructed images obtained through this method can be very noisy. At present, the deep learning image denoising method represented by convolutional neural networks (CNN) has become a classical method for low-dose CT image denoising. Inspired by the good performance of transformer in computer vision tasks, this paper proposes a CNN transformer coupling network (CTC) to further improve the performance of CT image denoising. CTC network makes comprehensive use of local information association ability of CNN and global information capture ability of transformer, constructs eight core network blocks composed of CNN components and an improved transformer component, which are interconnected based on residual connection mechanism and information reuse mechanism. Compared with the existing four denoising networks, CTC network demonstrate better denoising ability and can realize high-precision low-dose CT image reconstruction.
5
2022, 31(6): 709-720.
DOI: 10.15953/j.ctta.2021.077
Abstract：
Reducing the dose of computed tomography (CT) is essential for reducing the radiation risk in clinical applications. With the rapid development and wide application of deep learning, it has brought new directions for the development of low-dose CT imaging algorithms. Unlike most existing prior-driven algorithms that benefit from manually designed prior functions or supervised learning schemes, in this paper, we use an energy-based deep model to learn the prior knowledge of normal-dose CT, and then in the iterative reconstruction phase, we integrate data consistency as a conditional item into the iterative generation model of low-dose CT, and realize the low-dose CT reconstruction through the prior experience of iterative updating training of Langevin dynamics. The experimental results show that the proposed method hold excellent noise reduction and detail retention capabilities.
Reducing the dose of computed tomography (CT) is essential for reducing the radiation risk in clinical applications. With the rapid development and wide application of deep learning, it has brought new directions for the development of low-dose CT imaging algorithms. Unlike most existing prior-driven algorithms that benefit from manually designed prior functions or supervised learning schemes, in this paper, we use an energy-based deep model to learn the prior knowledge of normal-dose CT, and then in the iterative reconstruction phase, we integrate data consistency as a conditional item into the iterative generation model of low-dose CT, and realize the low-dose CT reconstruction through the prior experience of iterative updating training of Langevin dynamics. The experimental results show that the proposed method hold excellent noise reduction and detail retention capabilities.
6
2023, 32(4): 553-558.
DOI: 10.15953/j.ctta.2022.161
Abstract：
Objective: To analyze the image performance of colon innervation defect with multilayer spiral CT (MSCT) and clinical manifestations, summarize its image characteristics and make correct diagnosis. Methods: The clinical features of colonic innervations deficiency present with prolonged constipation and incomplete ileus. MSCT imaging data using GE gem energy spectrum, CT 750 HD, and Philips MSCT. 1 mm layer thickness, 1 mm layer spacing, tube voltage, 120 kV, automatic tube current from the diaphragm to the bilateral pubic joint. Scan in the natural state of the intestine (No bowel preparation, no cleansing enema and bowel cleansing), after scanning, conduct MPR 3D reconstruction at the CT workstation, and the reconstructed data were archived and analyzed in the PACS system. Clarify the intestinal location of the diseased segment, measure intestinal wall thickness of dilated segment and narrow segment respectively; measure intestinal tube length of diseased segment (narrow segment); observe intestinal peristalsis with multiple-stage MSCT; and observe intestinal blood transport through enhancement. Results: The clinical features of colon innervations defect was constipation and incomplete ileus. In this study group, there were 5 adult patients with colon innervations defect, and the lesion site was located in the spleen and descending colon respectively, among which the diseased segment was located in 3 cases and the spleen was located in 2 cases of colon; MSCT shows relative narrowing of the colon and expansion of the proximal colon; The intestinal wall thickness was normal in the diseased area, and the intestinal wall thickness of the dilated colon section was normal or somewhat thickened, and the thickened intestinal wall in this group is less than 0.9 cm; The intestinal length of the diseased segment in this group was somewhere between 4.3~8.6 cm. The MSCT enhancement scan of mesangic vessels and mesangial density showed no abnormal changes, and no abnormal enhancement of the colon wall in the diseased section, suggesting normal blood supply; MSCT enhanced scan showed rigidity and no peristalsis in the diseased segment, suggesting loss of peristaltic function in the diseased segment. Conclusion: Colonic innervations defect has imaging findings of characteristic post dilating stenosis and clinical features of prolonged constipation and incomplete obstruction in adults, The MSCT combined with clinical data was able to indicate the diagnosis of colonic innervations defect before surgery.
Objective: To analyze the image performance of colon innervation defect with multilayer spiral CT (MSCT) and clinical manifestations, summarize its image characteristics and make correct diagnosis. Methods: The clinical features of colonic innervations deficiency present with prolonged constipation and incomplete ileus. MSCT imaging data using GE gem energy spectrum, CT 750 HD, and Philips MSCT. 1 mm layer thickness, 1 mm layer spacing, tube voltage, 120 kV, automatic tube current from the diaphragm to the bilateral pubic joint. Scan in the natural state of the intestine (No bowel preparation, no cleansing enema and bowel cleansing), after scanning, conduct MPR 3D reconstruction at the CT workstation, and the reconstructed data were archived and analyzed in the PACS system. Clarify the intestinal location of the diseased segment, measure intestinal wall thickness of dilated segment and narrow segment respectively; measure intestinal tube length of diseased segment (narrow segment); observe intestinal peristalsis with multiple-stage MSCT; and observe intestinal blood transport through enhancement. Results: The clinical features of colon innervations defect was constipation and incomplete ileus. In this study group, there were 5 adult patients with colon innervations defect, and the lesion site was located in the spleen and descending colon respectively, among which the diseased segment was located in 3 cases and the spleen was located in 2 cases of colon; MSCT shows relative narrowing of the colon and expansion of the proximal colon; The intestinal wall thickness was normal in the diseased area, and the intestinal wall thickness of the dilated colon section was normal or somewhat thickened, and the thickened intestinal wall in this group is less than 0.9 cm; The intestinal length of the diseased segment in this group was somewhere between 4.3~8.6 cm. The MSCT enhancement scan of mesangic vessels and mesangial density showed no abnormal changes, and no abnormal enhancement of the colon wall in the diseased section, suggesting normal blood supply; MSCT enhanced scan showed rigidity and no peristalsis in the diseased segment, suggesting loss of peristaltic function in the diseased segment. Conclusion: Colonic innervations defect has imaging findings of characteristic post dilating stenosis and clinical features of prolonged constipation and incomplete obstruction in adults, The MSCT combined with clinical data was able to indicate the diagnosis of colonic innervations defect before surgery.
7
2023, 32(4): 545-552.
DOI: 10.15953/j.ctta.2022.135
Abstract：
Objective: To explore the film printing technology for displaying the true size of costal cartilage on medical film. Methods: The CT images of costal cartilage of 31 patients were selected and processed 14×17 inches (35×43 cm), calculate the main image display area and 14×17 the scale between the actual image display areas on the film shall be made into physical scales of 10 cm and 5 cm in length. Print the film in 2×2. Four images are arranged in the grid, and the transverse or sagittal images of single rib cartilage are printed on a 5 cm physical scale; with 1×1 The grid is a single image layout, and the 3D images of all the front rib arches (including costal cartilage) are printed on a 10 cm physical scale. The film measurement values of six indicators, including the length of the ascending and transverse parts of the right sixth costal cartilage, the width and thickness of the transverse junction of the ascending part, the thickness of the sternal end of the costal cartilage and the rib end of the costal cartilage, were compared with the solid measurement values during the operation, and statistical analysis was made. Results: (1) The printed film was measured, and the 10 cm and 5 cm scales shown on the film were equal to the actual size of the ruler. (2) There was no significant difference between the measured values of six groups of costal cartilage images on film and the measured values of costal cartilage entities during surgery. Conclusion: The printing technology based on DICOM protocol can realize the real size printing of costal cartilage CT image on film. The morphological data of the target tissue obtained by the operator from the film are reliable.
Objective: To explore the film printing technology for displaying the true size of costal cartilage on medical film. Methods: The CT images of costal cartilage of 31 patients were selected and processed 14×17 inches (35×43 cm), calculate the main image display area and 14×17 the scale between the actual image display areas on the film shall be made into physical scales of 10 cm and 5 cm in length. Print the film in 2×2. Four images are arranged in the grid, and the transverse or sagittal images of single rib cartilage are printed on a 5 cm physical scale; with 1×1 The grid is a single image layout, and the 3D images of all the front rib arches (including costal cartilage) are printed on a 10 cm physical scale. The film measurement values of six indicators, including the length of the ascending and transverse parts of the right sixth costal cartilage, the width and thickness of the transverse junction of the ascending part, the thickness of the sternal end of the costal cartilage and the rib end of the costal cartilage, were compared with the solid measurement values during the operation, and statistical analysis was made. Results: (1) The printed film was measured, and the 10 cm and 5 cm scales shown on the film were equal to the actual size of the ruler. (2) There was no significant difference between the measured values of six groups of costal cartilage images on film and the measured values of costal cartilage entities during surgery. Conclusion: The printing technology based on DICOM protocol can realize the real size printing of costal cartilage CT image on film. The morphological data of the target tissue obtained by the operator from the film are reliable.
8
Abstract：
Objective: To investigate the applicability of DR Bedside chest radiography and chest computed tomography (CT) scanning for the imaging and clinical diagnosis of severe novel coronavirus pneumonia. Methods: Imaging and clinical data of 43 patients with novel coronavirus pneumonia who were admitted to Beijing Shijitan Hospital Affiliated to Capital Medical University from December 10, 2022 to January 10, 2023 were retrospectively analyzed. Joint diagnoses by multidisciplinary experts according to the clinical and laboratory diagnosis criteria confirmed that all the 43 enrolled patients were severely infected with novel coronavirus pneumonia. All 43 patients underwent digital DR Bedside chest radiography; of these, 6 patients underwent chest CT scanning as they had relatively mild symptoms. The time interval for chest X-ray reexamination was 1~5 days, with 1~4 reexaminations. The chest CT scan review interval was 1~4 days, with 2 reviews. The imaging findings of chest radiography and chest CT scanning were observed and analyzed. Results: Among 43 patients with severe novel coronavirus pneumonia, 25 cases showed the first digital DR Bedside chest radiographs showing plaques and solid shadows in both lungs, 6 cases of pleural effusion in both lungs, 18 cases of unilateral lung exudation and solid shadows, 4 cases of unilateral pleural effusion, 7 cases of cardiac enlargement combined with pulmonary edema, and 2 cases of suspected pulmonary tumor mass shadows. There were multiple ground glass shadows in both lungs in 6 patients who underwent chest CT scanning. Spot shadows, grid shadows, thickened interlobular septum, and thickened pulmonary vessels and subbronchus were identified in the lesion area. Double lung spot density increased in the chest radiographs; the shadow area expanded in 5 cases and pleural effusion increased in 2 cases. Reexamination of chest CT images showed that 4 cases of increased ground-glass shadow transformed into irregular patchy high CT value shadow, 1 case of new atelectasis and 1 case of new pleural effusion. Conclusion: Digital DR Bedside chest radiography and chest CT scanning are the primary imaging methods for the diagnosis of novel coronavirus pneumonia. In particular, the former plays an important auxiliary diagnostic role when chest CT scan cannot be performed in patients with severe disease, and can greatly assist in the later review and clinical evaluation of the disease during active treatment.
Objective: To investigate the applicability of DR Bedside chest radiography and chest computed tomography (CT) scanning for the imaging and clinical diagnosis of severe novel coronavirus pneumonia. Methods: Imaging and clinical data of 43 patients with novel coronavirus pneumonia who were admitted to Beijing Shijitan Hospital Affiliated to Capital Medical University from December 10, 2022 to January 10, 2023 were retrospectively analyzed. Joint diagnoses by multidisciplinary experts according to the clinical and laboratory diagnosis criteria confirmed that all the 43 enrolled patients were severely infected with novel coronavirus pneumonia. All 43 patients underwent digital DR Bedside chest radiography; of these, 6 patients underwent chest CT scanning as they had relatively mild symptoms. The time interval for chest X-ray reexamination was 1~5 days, with 1~4 reexaminations. The chest CT scan review interval was 1~4 days, with 2 reviews. The imaging findings of chest radiography and chest CT scanning were observed and analyzed. Results: Among 43 patients with severe novel coronavirus pneumonia, 25 cases showed the first digital DR Bedside chest radiographs showing plaques and solid shadows in both lungs, 6 cases of pleural effusion in both lungs, 18 cases of unilateral lung exudation and solid shadows, 4 cases of unilateral pleural effusion, 7 cases of cardiac enlargement combined with pulmonary edema, and 2 cases of suspected pulmonary tumor mass shadows. There were multiple ground glass shadows in both lungs in 6 patients who underwent chest CT scanning. Spot shadows, grid shadows, thickened interlobular septum, and thickened pulmonary vessels and subbronchus were identified in the lesion area. Double lung spot density increased in the chest radiographs; the shadow area expanded in 5 cases and pleural effusion increased in 2 cases. Reexamination of chest CT images showed that 4 cases of increased ground-glass shadow transformed into irregular patchy high CT value shadow, 1 case of new atelectasis and 1 case of new pleural effusion. Conclusion: Digital DR Bedside chest radiography and chest CT scanning are the primary imaging methods for the diagnosis of novel coronavirus pneumonia. In particular, the former plays an important auxiliary diagnostic role when chest CT scan cannot be performed in patients with severe disease, and can greatly assist in the later review and clinical evaluation of the disease during active treatment.
9
Abstract：
Objective: This study aimed to investigate the correlation between the neutrophil-to-lymphocyte ratio (NLR) and chest high-resolution computed tomography (HRCT) findings of coronavirus disease 2019 (COVID-19). Materials and Methods: NLR and chest HRCT findings of 132 patients diagnosed with COVID-19 in the department of infectious diseases of Beijing Shijitan Hospital Capital Medical University from December 1, 2022 to February 1, 2023 were retrospectively analyzed. The patients were divided into two groups with NLR cut-off value of 3.0, and their HRCT characteristics and imaging manifestation patterns were analyzed. For the measurement data of normal distribution, the t-test of continuous variables was used between the groups. The data of non-normal distribution are expressed as median and quartile and compared using Mann-Whitney U test. The counting data are expressed as frequency, and the chi-squared or Fisher's exact test was used for comparison between the groups. P<0.05 indicates that the difference is statistically significant. Results: The number of lesions ≤5 and the proportion of lesions ≤10% were higher in the low NLR group than that in the high NLR group. The number of lesions >10 and the proportion of lesions >50% were higher in the high NLR group than that in the low NLR group. The high NLR group was prone to mixed density shadow, crazy-paving pattern, mosaic sign, anti-halo sign, subpleural black belt, arcade-like sign than that in the low NLR group. The high NLR group was most likely to have nonspecific interstitial pneumonia-like, organizing pneumonia-like, and diffuse alveolar damage-like patterns than that in the low NLR group. Conclusion: Different NLRs have different manifestations of COVID-19 chest HRCT. The high NLR group is more prone to mixed density shadow, crazy-paving pattern, mosaic sign, anti-halo sign, subpleural black belt, and arcade-like sign, as well as most likely to have radiologic patterns of nonspecific interstitial pneumonia, organizing pneumonia, diffuse alveolar damage.
Objective: This study aimed to investigate the correlation between the neutrophil-to-lymphocyte ratio (NLR) and chest high-resolution computed tomography (HRCT) findings of coronavirus disease 2019 (COVID-19). Materials and Methods: NLR and chest HRCT findings of 132 patients diagnosed with COVID-19 in the department of infectious diseases of Beijing Shijitan Hospital Capital Medical University from December 1, 2022 to February 1, 2023 were retrospectively analyzed. The patients were divided into two groups with NLR cut-off value of 3.0, and their HRCT characteristics and imaging manifestation patterns were analyzed. For the measurement data of normal distribution, the t-test of continuous variables was used between the groups. The data of non-normal distribution are expressed as median and quartile and compared using Mann-Whitney U test. The counting data are expressed as frequency, and the chi-squared or Fisher's exact test was used for comparison between the groups. P<0.05 indicates that the difference is statistically significant. Results: The number of lesions ≤5 and the proportion of lesions ≤10% were higher in the low NLR group than that in the high NLR group. The number of lesions >10 and the proportion of lesions >50% were higher in the high NLR group than that in the low NLR group. The high NLR group was prone to mixed density shadow, crazy-paving pattern, mosaic sign, anti-halo sign, subpleural black belt, arcade-like sign than that in the low NLR group. The high NLR group was most likely to have nonspecific interstitial pneumonia-like, organizing pneumonia-like, and diffuse alveolar damage-like patterns than that in the low NLR group. Conclusion: Different NLRs have different manifestations of COVID-19 chest HRCT. The high NLR group is more prone to mixed density shadow, crazy-paving pattern, mosaic sign, anti-halo sign, subpleural black belt, and arcade-like sign, as well as most likely to have radiologic patterns of nonspecific interstitial pneumonia, organizing pneumonia, diffuse alveolar damage.
10
Abstract：
The cone-beam CT (CBCT) system based on the two-dimensional flat-panel detector technology is widely applied in patient location verification before radiotherapy. However, during the application of intraperitoneal tumor radiotherapy, severe shading and streaking artifacts caused by respiratory movement and intestinal peristalsis make it difficult to distinguish tumor areas from the CBCT images. Due to the non-rigid deformation of flexible organs such as the pancreas under the action of respiratory motion, it is hard to quantify deviation between the body surface motion monitoring results and the actual organ motion, and it is also difficult to monitor irregular motion represented by intestinal peristalsis. There is no effective solution to motion artifact correction in CBCT. Based on theory of biodynamics and common knowledge of human physiology, in this paper we propose a brand new radiotherapy image-guided cone-beam CT motion artifact correction method without motion monitoring or implantation of in-vivo markers. The proposed artifact correction strategy is designed based on the features of the artifact images and fusion of various CT image domain processing algorithms. The results suggest that the image quality of cone beam CT has been significantly improved after the application of this strategy in the clinical abdominal CBCT image processing. The average CT number error in typical soft tissue areas reduces from 90 HU to 30 HU, and the boundary of the intestinal cavity and surrounding soft tissue information are partially recovered. The proposed artifact correction strategy does not require respiratory gating or increase of projections, which can be integrated into existing workflows without marker implantation surgery. The motion-artifact-corrected CBCT images provide more accurate tumor localization information for image-guided radiotherapy of pancreatic carcinoma. The proposed method is proved practical and efficient for clinical applications
11
2022, 31(6): 721-729.
DOI: 10.15953/j.ctta.2022.108
Abstract：
Low-dose computed tomography (CT) is a relatively safe method for disease screening. But low-dose CT images often contain severe noise and artifacts, which seriously affect the subsequent diagnosis. To solve this problem, this paper proposes a subspace projection and edge enhancement network (SPEENet). SPEENet hold an architecture of autoencoder, including two main modules: dual stream encoder and decoder. The dual stream encoder can be divided into two parts: noise image coding stream and edge information coding stream. The noise image coding stream removes the noise and artifacts in low-dose CT images by using the image features extracted from the low-dose CT images. The edge information coding stream mainly focuses on the edge information of low-dose CT images and fully utilize the edge information to preserve the structures. In order to make full use of the encoder features, this paper introduces the noise basis projection module to establish a basis based on the features of encoder and decoder, and uses this basis to project the features extracted by the encoder into the corresponding subspace to obtain better feature representation. In this paper, experiments are conducted on the public database to verify the effectiveness. The experimental results show that SPEENet can achieve better denoising performance than other low-dose CT denoising networks.
Low-dose computed tomography (CT) is a relatively safe method for disease screening. But low-dose CT images often contain severe noise and artifacts, which seriously affect the subsequent diagnosis. To solve this problem, this paper proposes a subspace projection and edge enhancement network (SPEENet). SPEENet hold an architecture of autoencoder, including two main modules: dual stream encoder and decoder. The dual stream encoder can be divided into two parts: noise image coding stream and edge information coding stream. The noise image coding stream removes the noise and artifacts in low-dose CT images by using the image features extracted from the low-dose CT images. The edge information coding stream mainly focuses on the edge information of low-dose CT images and fully utilize the edge information to preserve the structures. In order to make full use of the encoder features, this paper introduces the noise basis projection module to establish a basis based on the features of encoder and decoder, and uses this basis to project the features extracted by the encoder into the corresponding subspace to obtain better feature representation. In this paper, experiments are conducted on the public database to verify the effectiveness. The experimental results show that SPEENet can achieve better denoising performance than other low-dose CT denoising networks.
12
2022, 31(6): 749-760.
DOI: 10.15953/j.ctta.2022.077
Abstract：
Spectral computed tomography (CT) based on photon counting detector can simultaneously collect projection data of multiple spectral channels and obtain absorption characteristics of material within corresponding energy ranges, so it can be effectively applied to material identification and material decomposition. Principal component analysis is an excellent multivariate analysis technique, which can be applied to process multi-energy spectral CT data. In this paper, principal component analysis was performed on spectral CT data in projection domain and image domain respectively, and the analysis results were compared systematically. Meanwhile, in order to reduce the influence of noise and improve the color characterization performance of spectral CT images, the method of combining double domain filtering with pixel value square was proposed to denoise the noisy principal component images, and then the selected principal component images were mapped to RGB color channels. The experimental results demonstrate that the principal component analysis can obtain clear CT images and identify the different components of the substance, whether in the projection domain or the image domain. However, compared with the principal component analysis method in the image domain, principal component analysis in the projection domain can retain more details of the substance and acquire clearer color CT images.
Spectral computed tomography (CT) based on photon counting detector can simultaneously collect projection data of multiple spectral channels and obtain absorption characteristics of material within corresponding energy ranges, so it can be effectively applied to material identification and material decomposition. Principal component analysis is an excellent multivariate analysis technique, which can be applied to process multi-energy spectral CT data. In this paper, principal component analysis was performed on spectral CT data in projection domain and image domain respectively, and the analysis results were compared systematically. Meanwhile, in order to reduce the influence of noise and improve the color characterization performance of spectral CT images, the method of combining double domain filtering with pixel value square was proposed to denoise the noisy principal component images, and then the selected principal component images were mapped to RGB color channels. The experimental results demonstrate that the principal component analysis can obtain clear CT images and identify the different components of the substance, whether in the projection domain or the image domain. However, compared with the principal component analysis method in the image domain, principal component analysis in the projection domain can retain more details of the substance and acquire clearer color CT images.
13
2022, 31(6): 731-747.
DOI: 10.15953/j.ctta.2022.111
Abstract：
Total variation (TV) minimum algorithm is an image reconstruction algorithm based on compressed sensing theory, which can realize the reconstruction of images with high accuracy from sparse projection or noisy projection data and has been widely used in computed tomography (CT), magnetic resonance imaging (MRI) and electronic paramagnetic resonance imaging (EPRI). Energy spectrum CT, T1 or T2 weighted MRI and EPRI both belong to multi-channel imaging. The channel-by-channel TV algorithm can achieve high-precision image reconstruction, but it ignores the similarity among the images of each channel while Nuclear TV algorithm is a TV algorithm that considers the image similarity among channels, and can realize high-precision image reconstruction. For multi-channel image reconstruction, taking CT reconstruction as an example, this paper proposes a nuclear TV multi-channel image reconstruction algorithm based on the framework of Chambolle-pock algorithm. Through the reconstruction experiments of simulated phantom and real CT image phantom, the accuracy of the algorithm is verified, the convergence of the algorithm is analyzed, the influence of algorithm parameters on the convergence rate is explored, and the sparse reconstruction ability and noisy projection reconstruction ability of the algorithm are evaluated. The experimental results show that the proposed algorithm can achieve higher reconstruction accuracy than the channel-by-channel TV algorithm. Nuclear TV algorithm is a high-precision multi-channel image reconstruction algorithm, which can be applied to multi-channel reconstruction of various imaging modes.
Total variation (TV) minimum algorithm is an image reconstruction algorithm based on compressed sensing theory, which can realize the reconstruction of images with high accuracy from sparse projection or noisy projection data and has been widely used in computed tomography (CT), magnetic resonance imaging (MRI) and electronic paramagnetic resonance imaging (EPRI). Energy spectrum CT, T1 or T2 weighted MRI and EPRI both belong to multi-channel imaging. The channel-by-channel TV algorithm can achieve high-precision image reconstruction, but it ignores the similarity among the images of each channel while Nuclear TV algorithm is a TV algorithm that considers the image similarity among channels, and can realize high-precision image reconstruction. For multi-channel image reconstruction, taking CT reconstruction as an example, this paper proposes a nuclear TV multi-channel image reconstruction algorithm based on the framework of Chambolle-pock algorithm. Through the reconstruction experiments of simulated phantom and real CT image phantom, the accuracy of the algorithm is verified, the convergence of the algorithm is analyzed, the influence of algorithm parameters on the convergence rate is explored, and the sparse reconstruction ability and noisy projection reconstruction ability of the algorithm are evaluated. The experimental results show that the proposed algorithm can achieve higher reconstruction accuracy than the channel-by-channel TV algorithm. Nuclear TV algorithm is a high-precision multi-channel image reconstruction algorithm, which can be applied to multi-channel reconstruction of various imaging modes.
14
Abstract：
In medical computed tomography imaging systems, Compton scattered photons generated by the interaction between X-rays and objects have a serious impact on image quality, especially in cone-beam computed tomography and multi-layer detector systems. Currently, there are many scattering artifact correction methods, which can be classified into three categories: hardware, software, and hybrid software and hardware correction methods. However, with the advances in computing power and development of deep learning in medical image processing, new methods of scattering artifact correction have appeared in recent years. This study first introduces traditional correction methods. Then, a method of scattering artifact correction based on deep learning is described in detail, which is divided into the correction method based on image domain and the correction method based on projection domain. Various deep-learning neural networks for this method are also introduced in detail. Finally, the application prospects of the deep learning method in multi-source computed tomography imaging scattering artifacts were probed .
In medical computed tomography imaging systems, Compton scattered photons generated by the interaction between X-rays and objects have a serious impact on image quality, especially in cone-beam computed tomography and multi-layer detector systems. Currently, there are many scattering artifact correction methods, which can be classified into three categories: hardware, software, and hybrid software and hardware correction methods. However, with the advances in computing power and development of deep learning in medical image processing, new methods of scattering artifact correction have appeared in recent years. This study first introduces traditional correction methods. Then, a method of scattering artifact correction based on deep learning is described in detail, which is divided into the correction method based on image domain and the correction method based on projection domain. Various deep-learning neural networks for this method are also introduced in detail. Finally, the application prospects of the deep learning method in multi-source computed tomography imaging scattering artifacts were probed .
15
2023, 32(1): 139-146.
DOI: 10.15953/j.ctta.2021.067
Abstract：
Spectral CT can produce basis materials with different X-ray energies. Subsequently, the generated basis materials can be used for qualitative and quantitative evaluation of tissue components and contrast agent distribution. This approach presents a superior ability to separate and identify imaging materials compared to traditional single-energy CT. Dual-energy spectrum technology is one of the most commonly used modes in spectrum CT, which plays an important role in clinical application. In this study, the decomposition methods of a basis material in the image domain of dual-energy spectrum CT were classified into two categories: two-material decomposition and multi-material decomposition. Finally, these methods are summarized and trend of future development is addressed.
Spectral CT can produce basis materials with different X-ray energies. Subsequently, the generated basis materials can be used for qualitative and quantitative evaluation of tissue components and contrast agent distribution. This approach presents a superior ability to separate and identify imaging materials compared to traditional single-energy CT. Dual-energy spectrum technology is one of the most commonly used modes in spectrum CT, which plays an important role in clinical application. In this study, the decomposition methods of a basis material in the image domain of dual-energy spectrum CT were classified into two categories: two-material decomposition and multi-material decomposition. Finally, these methods are summarized and trend of future development is addressed.
16
Abstract：
Rock and mineral samples contain a large amount of high-density metal substances, which often lead to metal artifacts in CT images and seriously affect the parameters analysis accuracy of rock and mineral samples. In order to improve the quality of CT images and suppress metal artifacts, in this paper we propose a dual-domain adaptive network-based metal artifact reduction algorithm for CT images of rock samples (DDA-CNN-MAR). The metal artifacts are suppressed by the projection domain network and the image domain network successively, and the dual domain processing results are adaptively fused to realize the end-to-end mapping from images with artifacts to artifact-free images. The algorithm is based on the residual encoder-decoder network model (RED-CNN), which is easy to extract features and restore image details. The dual-domain structure can adaptively adjust the weights of the projection domain (artifact suppression) and image domain (detail inpainting) to obtain optimal reduction results. The experimental results show that, compared with the RED-CNN metal artifact denoising method in the image domain, the MSE of the image corrected by the DDA-CNN-MAR method is reduced by 2.570, while the PSNR and SSIM are increased by 1.218 dB and 0.018 respectively, which effectively improves the CT image quality.
Rock and mineral samples contain a large amount of high-density metal substances, which often lead to metal artifacts in CT images and seriously affect the parameters analysis accuracy of rock and mineral samples. In order to improve the quality of CT images and suppress metal artifacts, in this paper we propose a dual-domain adaptive network-based metal artifact reduction algorithm for CT images of rock samples (DDA-CNN-MAR). The metal artifacts are suppressed by the projection domain network and the image domain network successively, and the dual domain processing results are adaptively fused to realize the end-to-end mapping from images with artifacts to artifact-free images. The algorithm is based on the residual encoder-decoder network model (RED-CNN), which is easy to extract features and restore image details. The dual-domain structure can adaptively adjust the weights of the projection domain (artifact suppression) and image domain (detail inpainting) to obtain optimal reduction results. The experimental results show that, compared with the RED-CNN metal artifact denoising method in the image domain, the MSE of the image corrected by the DDA-CNN-MAR method is reduced by 2.570, while the PSNR and SSIM are increased by 1.218 dB and 0.018 respectively, which effectively improves the CT image quality.
17
2025, 34(3): 506-524.
DOI: 10.15953/j.ctta.2024.316
Abstract：
Generative artificial intelligence represented by diffusion models has significantly contributed to medical imaging reconstruction. To help researchers comprehensively understand the rich content of diffusion models, this review provides a detailed overview of diffusion models used in medical imaging reconstruction. The theoretical foundation and fundamental concepts underlying the diffusion modeling framework were first introduced, describing their origin and evolution. Second, a systematic characteristic-based taxonomy of diffusion models used in medical imaging reconstruction is provided, broadly covering their application to imaging modalities, including magnetic resonance imaging (MRI), computed tomography (CT), positron emission computed tomography (PET), and photoacoustic imaging (PAI). Finally, we discuss the limitations of current diffusion models and anticipate potential directions of future research, providing an intuitive starting point for subsequent exploratory research. Related codes are available at GitHub: https://github.com/yqx7150/Diffusion-Models-for-Medical-Imaging.
18
2022, 31(6): 773-781.
DOI: 10.15953/j.ctta.2022.085
Abstract：
In computed tomography (CT) imaging, artifacts will degrade the quality of reconstructed images. To solve this issue, in this paper we propose a method to remove concentric elliptical artifacts from CT images. This method is based on the idea of Directional total variation (DTV), which models the problem of elliptical artifact removal as an energy minimization problem, and establishes a variation-based model which is adapted the elliptical artifacts by analyzing the edge features of elliptical artifacts. Since the proposed model is a non-smooth convex optimization problem with a divisible structure, the alternating direction multiplier method (ADMM) is applied. Finally, the effectiveness of the model in removing elliptical artifacts is verified by simulation experiments.
In computed tomography (CT) imaging, artifacts will degrade the quality of reconstructed images. To solve this issue, in this paper we propose a method to remove concentric elliptical artifacts from CT images. This method is based on the idea of Directional total variation (DTV), which models the problem of elliptical artifact removal as an energy minimization problem, and establishes a variation-based model which is adapted the elliptical artifacts by analyzing the edge features of elliptical artifacts. Since the proposed model is a non-smooth convex optimization problem with a divisible structure, the alternating direction multiplier method (ADMM) is applied. Finally, the effectiveness of the model in removing elliptical artifacts is verified by simulation experiments.
19
2022, 31(6): 793-808.
DOI: 10.15953/j.ctta.2022.043
Abstract：
The traditional ultrasonic dynamic focusing imaging which simply uses the delay-and-sum method in beamforming shows low resolution and poor contrast. In this study, based on the coherent pixel-based beamforming (Coherent PB), a novel Wiener pre-filter with high computational efficiency is applied to correct the phase of the entire array signal by using the pulse-echo at the focal point. We also apply a Wiener post-filter calculating adaptive weights for the beamforming results of each sub-aperture to suppress the noise and artifacts. The proposed method's effectiveness is verified through simulation experiments, phantom experiments, and in vivo experiments. Compared with Coherent PB, the proposed method significantly improves the axial resolution and contrast of images while maintaining the high lateral resolution and computational efficiency, which shows certain clinical application value.
The traditional ultrasonic dynamic focusing imaging which simply uses the delay-and-sum method in beamforming shows low resolution and poor contrast. In this study, based on the coherent pixel-based beamforming (Coherent PB), a novel Wiener pre-filter with high computational efficiency is applied to correct the phase of the entire array signal by using the pulse-echo at the focal point. We also apply a Wiener post-filter calculating adaptive weights for the beamforming results of each sub-aperture to suppress the noise and artifacts. The proposed method's effectiveness is verified through simulation experiments, phantom experiments, and in vivo experiments. Compared with Coherent PB, the proposed method significantly improves the axial resolution and contrast of images while maintaining the high lateral resolution and computational efficiency, which shows certain clinical application value.
20
Abstract：
Since the first X-ray computed tomography(CT) scanning was performed by Hounsfield in 1971,CT has mainly walked through five generations and deeply influenced us in many fields,such as medical applications and security inspections.With the development in CT scanning configurations,their reconstruction theories have been growing rapidly.Compared with iterative methods,analytical algorithms can achieve faster reconstruction with simpler error analysis and fewer computing resources,which makes them rather popular in CT products.This review strives to illustrate analytical CT reconstruction methods following the development of scanning modes,from circular and helical trajectories to nonstandard trajectories.As a special case of nonstandard trajectories,recent progress on multi-source linear CT is also highlighted.Finally,we discuss some opportunities and challenges of CT scanning configurations in the future.
Since the first X-ray computed tomography(CT) scanning was performed by Hounsfield in 1971,CT has mainly walked through five generations and deeply influenced us in many fields,such as medical applications and security inspections.With the development in CT scanning configurations,their reconstruction theories have been growing rapidly.Compared with iterative methods,analytical algorithms can achieve faster reconstruction with simpler error analysis and fewer computing resources,which makes them rather popular in CT products.This review strives to illustrate analytical CT reconstruction methods following the development of scanning modes,from circular and helical trajectories to nonstandard trajectories.As a special case of nonstandard trajectories,recent progress on multi-source linear CT is also highlighted.Finally,we discuss some opportunities and challenges of CT scanning configurations in the future.
- First
- Prev
- 1
- 2
- 3
- 4
- 5
- Next
- Last
- Total 5 Pages
- To
- Go
