scholarly journals Novel Computational Approaches for Understanding Computed Tomography (CT) Images and Their Applications

10.5772/15156 ◽  
2011 ◽  
Author(s):  
Oyeon Kum
Keyword(s):  
Computed Tomography ◽  
Ct Images ◽  
Computational Approaches

scholarly journals Multi-Cycle-Consistent Adversarial Networks for Edge Denoising of Computed Tomography Images

2021 ◽  
Vol 17 (4) ◽  
pp. 1-16
Author(s):  
Xiaowe Xu ◽  
Jiawei Zhang ◽  
Jinglan Liu ◽  
Yukun Ding ◽  
Tianchen Wang ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Image Denoising ◽  
Ct Images ◽  
High Dose ◽  
Ct Image ◽  
Paired Data ◽  
High Quality ◽  
Computed Tomography Images ◽  
Significant Difference ◽  
Global Cycle

As one of the most commonly ordered imaging tests, the computed tomography (CT) scan comes with inevitable radiation exposure that increases cancer risk to patients. However, CT image quality is directly related to radiation dose, and thus it is desirable to obtain high-quality CT images with as little dose as possible. CT image denoising tries to obtain high-dose-like high-quality CT images (domain Y ) from low dose low-quality CT images (domain X ), which can be treated as an image-to-image translation task where the goal is to learn the transform between a source domain X (noisy images) and a target domain Y (clean images). Recently, the cycle-consistent adversarial denoising network (CCADN) has achieved state-of-the-art results by enforcing cycle-consistent loss without the need of paired training data, since the paired data is hard to collect due to patients’ interests and cardiac motion. However, out of concerns on patients’ privacy and data security, protocols typically require clinics to perform medical image processing tasks including CT image denoising locally (i.e., edge denoising). Therefore, the network models need to achieve high performance under various computation resource constraints including memory and performance. Our detailed analysis of CCADN raises a number of interesting questions that point to potential ways to further improve its performance using the same or even fewer computation resources. For example, if the noise is large leading to a significant difference between domain X and domain Y , can we bridge X and Y with a intermediate domain Z such that both the denoising process between X and Z and that between Z and Y are easier to learn? As such intermediate domains lead to multiple cycles, how do we best enforce cycle- consistency? Driven by these questions, we propose a multi-cycle-consistent adversarial network (MCCAN) that builds intermediate domains and enforces both local and global cycle-consistency for edge denoising of CT images. The global cycle-consistency couples all generators together to model the whole denoising process, whereas the local cycle-consistency imposes effective supervision on the process between adjacent domains. Experiments show that both local and global cycle-consistency are important for the success of MCCAN, which outperforms CCADN in terms of denoising quality with slightly less computation resource consumption.

Image restoration in computed tomography: restoration of experimental CT images

1992 ◽  
Vol 11 (4) ◽  
pp. 546-553 ◽  
Author(s):  
S. Rathee ◽  
Z.J. Koles ◽  
T.R. Overton
Keyword(s):  
Computed Tomography ◽  
Image Restoration ◽  
Ct Images

The Labyrinthine Portion of the Facial Canal in Patients with Bell's Palsy Investigated by Computed Tomography

1987 ◽  
Vol 28 (1) ◽  
pp. 25-30 ◽  
Author(s):  
K. Wadin ◽  
L. Thomander ◽  
H. Wilbrand
Keyword(s):  
Computed Tomography ◽  
Temporal Bone ◽  
No Value ◽  
Bell’S Palsy ◽  
Ct Images ◽  
Slice Thickness ◽  
Facial Canal ◽  
Bell's Palsy ◽  
Unreliable Method ◽  
Different Parts

The reproducibility of the labyrinthine portion of the facial canal by computed tomography was investigated in 22 patients with Bell's palsy. The CT images were compared with those obtained in 18 temporal bone specimens. Measurements of the diameters of different parts of the facial canal were made on these images and also microscopically in plastic casts of the temporal bone specimens. No marked difference was found between the dimensions of the labyrinthine portion of the facial canal of the involved and healthy temporal bone in the patient, nor did these differ from the dimensions in the specimens. CT of the slender, curved labyrinthine portion was found to be of doubtful value for metric estimation of small differences in width. The anatomic variations of the canal rendered the evaluation more difficult. CT with a slice thickness of 2 mm was of no value for assessment of this part of the canal. Measurement of the diameters of the labyrinthine portion on CT images is an inappropriate and unreliable method for clinical purposes.

Deep Learning Enabled Deblurring of Computed Tomography Images of Porous Media

2021 ◽  
Author(s):  
Khalid Labib Alsamadony ◽  
Ertugrul Umut Yildirim ◽  
Guenther Glatz ◽  
Umair bin Waheed ◽  
Sherif M. Hanafy
Keyword(s):  
Computed Tomography ◽  
Image Quality ◽  
Transfer Learning ◽  
Temporal Resolution ◽  
Exposure Time ◽  
Ct Images ◽  
Loss Functions ◽  
High Quality ◽  
X Ray ◽  
Training Images

Abstract Computed tomography (CT) is an important tool to characterize rock samples allowing quantification of physical properties in 3D and 4D. The accuracy of a property delineated from CT data is strongly correlated with the CT image quality. In general, high-quality, lower noise CT Images mandate greater exposure times. With increasing exposure time, however, more wear is put on the X-Ray tube and longer cooldown periods are required, inevitably limiting the temporal resolution of the particular phenomena under investigation. In this work, we propose a deep convolutional neural network (DCNN) based approach to improve the quality of images collected during reduced exposure time scans. First, we convolve long exposure time images from medical CT scanner with a blur kernel to mimic the degradation caused because of reduced exposure time scanning. Subsequently, utilizing the high- and low-quality scan stacks, we train a DCNN. The trained network enables us to restore any low-quality scan for which high-quality reference is not available. Furthermore, we investigate several factors affecting the DCNN performance such as the number of training images, transfer learning strategies, and loss functions. The results indicate that the number of training images is an important factor since the predictive capability of the DCNN improves as the number of training images increases. We illustrate, however, that the requirement for a large training dataset can be reduced by exploiting transfer learning. In addition, training the DCNN on mean squared error (MSE) as a loss function outperforms both mean absolute error (MAE) and Peak signal-to-noise ratio (PSNR) loss functions with respect to image quality metrics. The presented approach enables the prediction of high-quality images from low exposure CT images. Consequently, this allows for continued scanning without the need for X-Ray tube to cool down, thereby maximizing the temporal resolution. This is of particular value for any core flood experiment seeking to capture the underlying dynamics.

scholarly journals Machine Learning Models for Sarcopenia Identification Based on Radiomic Features of Muscles in Computed Tomography

2021 ◽  
Vol 18 (16) ◽  
pp. 8710
Author(s):  
Young Jae Kim
Keyword(s):  
Machine Learning ◽  
Computed Tomography ◽  
Ct Images ◽  
Erector Spinae ◽  
Support Vector ◽  
Gray Level ◽  
Learning Models ◽  
Extreme Gradient Boosting ◽  
Nsclc Patients ◽  
Machine Learning Models

The diagnosis of sarcopenia requires accurate muscle quantification. As an alternative to manual muscle mass measurement through computed tomography (CT), artificial intelligence can be leveraged for the automation of these measurements. Although generally difficult to identify with the naked eye, the radiomic features in CT images are informative. In this study, the radiomic features were extracted from L3 CT images of the entire muscle area and partial areas of the erector spinae collected from non-small cell lung carcinoma (NSCLC) patients. The first-order statistics and gray-level co-occurrence, gray-level size zone, gray-level run length, neighboring gray-tone difference, and gray-level dependence matrices were the radiomic features analyzed. The identification performances of the following machine learning models were evaluated: logistic regression, support vector machine (SVM), random forest, and extreme gradient boosting (XGB). Sex, coarseness, skewness, and cluster prominence were selected as the relevant features effectively identifying sarcopenia. The XGB model demonstrated the best performance for the entire muscle, whereas the SVM was the worst-performing model. Overall, the models demonstrated improved performance for the entire muscle compared to the erector spinae. Although further validation is required, the radiomic features presented here could become reliable indicators for quantifying the phenomena observed in the muscles of NSCLC patients, thus facilitating the diagnosis of sarcopenia.

scholarly journals Computed Tomography Imaging Results of Single Band Anterior Cruciate Ligament Reconstruction: A Retrospective Study Comparing Fixed- and Adjustable-Length Loop Cortical Fixation on the First Day After Surgery

2021 ◽  
Author(s):  
Lifeng Yin ◽  
Hua Zhang ◽  
Junbo Liu ◽  
xingyu zhang ◽  
zhengxing wen ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Anterior Cruciate Ligament ◽  
Acl Reconstruction ◽  
Femoral Tunnel ◽  
Cruciate Ligament ◽  
Ct Images ◽  
Graft Fixation ◽  
Bone Preservation ◽  
Anterior Cruciate ◽  
Graft Insertion

Abstract Background: Cortical suspensory femoral fixation is commonly performed for graft fixation of the femur in anterior cruciate ligament (ACL) reconstruction using hamstring tendons. This study aimed to compare the morphology of femoral tunnel and graft insertion between fixed-length loop devices (FLD) and adjustable-length loop devices (ALD) using computed tomography (CT) images on the first day after hamstring ACL reconstruction. Methods: Overall, 94 patients who underwent ACL reconstruction from January 2016 to January 2021 were included. For femoral graft fixation, FLD (Smith & Nephew, ENDOBUTTON) and ALD (DePuy Synthes, Mitek sports medicine, RIGIDLOOP Adjustable cortical system) were used in 56 and 38 patients, respectively (FLD and ALD groups). For evaluation of the morphology of the humeral tunnel and graft depth, CT scans were performed immediately on the first postoperative day. The gap distance between the top of the graft and the socket tunnel end, the length of lateral bone preservation, and the depth of graft insertion were measured on the CT images. Results: The gap distance and bone preservation significantly differed between the two groups (1.90±1.81 mm and 14.35±4.67 mm in ALD groups; 7.08±2.63 mm and 7.35±3.62 mm in FLD groups, respectively; both P values < 0.01). The graft insertion depth did not significantly differ between the groups. Conclusion: The ALD group had a smaller gap distance, better bone preservation, and a similar graft insertion length in the femoral tunnel when compared to the FLD group. Based on these findings, ALD might be better for bone preservation and tunnel utilization in patients with short femoral tunnels. Trial registration: retrospectively registered

Evaluation of Articular Cartilage Injury Using Computed Tomography With Axial Traction in the Ankle Joint

2018 ◽  
Vol 39 (9) ◽  
pp. 1120-1127 ◽  
Author(s):  
Tomoyuki Nakasa ◽  
Yasunari Ikuta ◽  
Mikiya Sawa ◽  
Masahiro Yoshikawa ◽  
Yusuke Tsuyuguchi ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Articular Cartilage ◽  
Sensitivity And Specificity ◽  
Subchondral Bone ◽  
Ct Images ◽  
Cartilage Layer ◽  
Arthroscopic Findings ◽  
Cartilage Injuries ◽  
Axial Traction ◽  
Articular Cartilage Injuries

Background: Although chondral or osteochondral injuries are usually assessed by magnetic resonance imaging, its accuracy can be low, presumably related to the relatively thin cartilage layer and the close apposition of the cartilage of the talus and tibial plafond. We hypothesized that axial traction could provide a contrast between the articular cartilage and joint cavity, and it enabled the simultaneous evaluation of cartilage and subchondral bone. The purpose of this study was to assess the feasibility of using computed tomography (CT) imaging with axial traction for the diagnosis of articular cartilage injuries. Methods: Chondral lesions in 18 ankles were evaluated by CT with axial traction using a tensioning device and ankle strap for enlargement of the joint space of the ankle. CT was done in 3-mm slices and programmed for gray scale, and then CT images were allocated colors to make it easier to evaluate the cartilage layer. The International Cartilage Repair Society (ICRS) grades on CT were compared with those on arthroscopic findings. Results: The respective sensitivity and specificity of CT imaging with traction using ICRS grading were 74.4%, and 96.3%. The level of agreement of the ICRS grading between CT images and arthroscopic findings was moderate (kappa coefficient, 0.547). Adding axial traction to CT increased the delineation of the cartilage surface, including chondral thinning, chondral defect, and cartilage separation. Conclusions: CT with axial traction produced acceptable levels of sensitivity and specificity for the evaluation of articular cartilage injuries, in addition to the assessment of subchondral bone. Level of Evidence: Level III, comparative case series.

EVALUATION OF VARIATIONS IN SPHENOID SINUS PNEUMATIZATION USING MULTIDETECTOR COMPUTED TOMOGRAPHY IN PATIENTS ATTENDING TERTIARY CARE HOSPITAL, VISAKHAPATNAM

2021 ◽  
pp. 62-65
Author(s):  
Sonica Sharma ◽  
Bhamidipaty Kanaka Durgaprasad ◽  
Payala Vijayalakshmi
Keyword(s):  
Computed Tomography ◽  
Paranasal Sinuses ◽  
Sphenoid Sinus ◽  
Tertiary Care Hospital ◽  
Tertiary Care ◽  
Ct Images ◽  
Care Hospital ◽  
Anterior Clinoid Process ◽  
Anterior Extension ◽  
Sphenoid Sinuses

BACKGROUND: The purpose of our study was to assess the prevalence of different patterns of pneumatization in the sphenoid sinuses as detected on the computed tomography (CT) images of paranasal sinuses of the patients presenting with various pathologies. This is a retrospective radiological study of CT im MATERIALS AND METHODS: ages of paranasal sinuses, done at Radio diagnosis department of a Tertiary care hospital. The study comprised CT images of 500 patients in the age range of 18-75years who were referred for CT scan of paranasal sinuses for various pathologies between the period of July 2018 and July 2019. All images of paranasal sinuses had been acquired following a standardized protocol in axial plane. Their reconstructed images in axial, coronal and sagittal planes were evaluated, using Osirix software, for the extent and different patterns of sphenoid sinus pneumatization. The Images of sphenoid sinuses were assessed for the posterior, lateral and anterior extension of their pneumatization The sphenoid sinuses pneumatization patterns in the RESULTS: descending order of prevalence were complete sellar (75.0%), incomplete sellar (22.6%), presellar (2.4%) and conchal (0%). The clival extensions was seen in 75% of patients and lateral extension sides in 49.1% patients. Lateral recesses as assessed on coronal images was seen in 49.1 % of cases with the prevalence in descending order being extension into pterygoid process 59.8 %, greater wing of sphenoid 9.2 %, full lateral 41% and lesser wing (anterior clinoid process) 19.3%. The pure forms were relatively less common and combined forms being more common. A preoperative review of the sphenoid anatomy should allow for safer endo

Sign in / Sign up

Export Citation Format

Share Document