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.

scholarly journals Three-dimensional assessment of the anterior and inferior loop of the inferior alveolar nerve using computed tomography images in patients with and without mandibular asymmetry

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Jae-Young Kim ◽  
Michael D. Han ◽  
Kug Jin Jeon ◽  
Jong-Ki Huh ◽  
Kwang-Ho Park
Keyword(s):  
Computed Tomography ◽  
Symmetry Group ◽  
Three Dimensional ◽  
Inferior Alveolar Nerve ◽  
Mental Foramen ◽  
3D Analysis ◽  
Mandibular Asymmetry ◽  
Computed Tomography Images ◽  
Significant Difference ◽  
Mandibular Body

Abstract Background The purpose of this study was to investigate the differences in configuration and dimensions of the anterior loop of the inferior alveolar nerve (ALIAN) in patients with and without mandibular asymmetry. Method Preoperative computed tomography images of patients who had undergone orthognathic surgery from January 2016 to December 2018 at a single institution were analyzed. Subjects were classified into two groups as “Asymmetry group” and “Symmetry group”. The distance from the most anterior and most inferior points of the ALIAN (IANant and IANinf) to the vertical and horizontal reference planes were measured (dAnt and dInf). The distance from IANant and IANinf to the mental foramen were also calculated (dAnt_MF and dInf_MF). The length of the mandibular body and symphysis area were measured. All measurements were analyzed using 3D analysis software. Results There were 57 total eligible subjects. In the Asymmetry group, dAnt and dAnt_MF on the non-deviated side were significantly longer than the deviated side (p < 0.001). dInf_MF on the non-deviated side was also significantly longer than the deviated side (p = 0.001). Mandibular body length was significantly longer on the non-deviated side (p < 0.001). There was no significant difference in length in the symphysis area (p = 0.623). In the Symmetry group, there was no difference between the left and right sides for all variables. Conclusion In asymmetric patients, there is a difference tendency in the ALIAN between the deviated and non-deviated sides. In patients with mandibular asymmetry, this should be considered during surgery in the anterior mandible.

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 PREVALENCE OF MANDIBULAR ASYMMETRY IN SKELETAL CLASS I ADULT PATIENTS

2014 ◽  
Vol 2 (2) ◽  
pp. 189 ◽  
Author(s):  
Bruno Frazão Gribel ◽  
Guilherme Thiesen ◽  
Tássia Silvana Borges ◽  
Maria Perpétua Mota Freitas
Keyword(s):  
Computed Tomography ◽  
Calculation Method ◽  
Significant Error ◽  
Adult Patients ◽  
Class I ◽  
Cone Beam ◽  
Skeletal Class ◽  
Mandibular Asymmetry ◽  
Computed Tomography Images ◽  
Significant Difference

AIM: The objective of this study was evaluating the prevalence of mandibular asymmetry in skeletal Class I adult patients. MATERIALS AND METHODS: The sample was composed by cone-beam computed tomography images of 250 skeletal Class I patients with age from 18 to 70 years old. The side deviation of mandibular asymmetry was evaluated (right and left), as well as the intensity of this asymmetry. People with gnathic deviation until 2mm were considered as patients with slight asymmetry; deviation between 2 and 5mm was considered moderate asymmetry and those higher than 5mm as severe asymmetries. The error calculation method was performed and there was not significant error in the measurements. To verify the association between the prevalence of mandibular asymmetry and the gender of individuals, the Χ2 was carried out and the significant level adopted was 5% (p<0.05). RESULTS: The results showed that mandibular deviation occurred more frequently to the left side in both genders (male 56.5% e female 58.2%), without statistically significant difference between them (p=0.792). Besides, the prevalence of slight, moderate and severe asymmetries were, respectively 61%, 27% and 12% for males and 54%, 28% and 18% for female; and there was not statistically significant difference (p=0.366) between the gender of the sample evaluated. CONCLUSION: In this study, both deviation and intensity of mandibular asymmetry did not present preference regarding to the gender.

scholarly journals Evaluation of the Prevalence of Concha Bullosa in Cone-Beam Computed Tomography Images

2020 ◽  
Vol 12 (3) ◽  
pp. 93-96
Author(s):  
Nasim Shams ◽  
Bahareh Shams ◽  
Zahra Sajadi
Keyword(s):  
Computed Tomography ◽  
Cone Beam Computed Tomography ◽  
Middle Turbinate ◽  
Cone Beam ◽  
P Value ◽  
Concha Bullosa ◽  
Computed Tomography Images ◽  
Significant Difference ◽  
And Gender ◽  
Sinus Ostium

Background: The ostiomeatal complex (OMC) is not a separate anatomical structure although it is a functional unit of structures, including the middle meatus, uncinate process, infundibulum, maxillary sinus ostium, ethmoidal bulla, anterior ethmoid sinus ostium, and frontal recess. Concha bullosa is the pneumatization of the concha, which is one of the most common anatomical variations in the middle turbinate. Methods: This study was conducted using the cone-beam computed tomography (CBCT) images of 172 patients in the archives of the Department of Oral and Maxillofacial Radiology, Dentistry School, Ahvaz Jundishapur. Patient information including age and gender, presence or absence of concha bullosa, the involved side (left or right), and its type (i.e., extensive, lamellar, and bulbous) were collected in the information form. Finally, the chi-square test (with SPSS, version 22) was used to analyze the data, and P value less than 0.05 was considered statistically significant. Results: Patients with and without concha bullosa were 39.1 and 41.7 years, respectively, but it was no significant difference in terms of age (P = 0.321). Out of 52 patients with concha bullosa, 19 (36.5%) cases were males and 33 (63.5%) of them were females. The prevalence of concha bullosa was higher for the bilateral side (20 patients, 38.5%, P = 0.000). The prevalence of bulbulsand lamellar-shape was nearly the same (32.7% and 30.8%, respectively). Eventually, the extensive shape with 36.5% was more frequent for the shape of concha bullosa (P = 0.000). Conclusions: The prevalence of concha bullosa was high. There was no significant difference in terms of age (P = 0.321) and gender (P = 0.058) of patients with concha bullosa. The extensive type and the bilateral appearance of concha bullosa were more significant (P = 0.000).

scholarly journals A Comparative Study of Auto-contouring Softwares in Delineation of Organs at Risk in Lung Cancer and Rectal Cancer

2021 ◽  
Author(s):  
weijun chen ◽  
Cheng Wang ◽  
Wenming Zhan ◽  
Yongshi Jia ◽  
Fangfang Ruan ◽  
...  
Keyword(s):  
Lung Cancer ◽  
At Risk ◽  
Rectal Cancer ◽  
Organs At Risk ◽  
Similarity Index ◽  
Slight Modification ◽  
Ct Images ◽  
Target Area ◽  
Ct Image ◽  
Significant Difference

Abstract Background:Radiotherapy requires the target area and the organs at risk to be contoured on the CT image of the patient. During the process of organs-at-Risk (OAR) of the chest and abdomen, the doctor needs to contour at each CT image. The delineations of large and varied shapes are time-consuming and laborious.This study aims to evaluate the results of two automatic contouring software on OAR definition of CT images of lung cancer and rectal cancer patients. Methods: The CT images of 15 patients with rectal cancer and 15 patients with lung cancer were selected separately, and the organs at risk were outlined by the same experienced doctor as references, and then the same datasets were automatically contoured based on AiContour®© (Manufactured by Linking MED, China) and Raystation®© (Manufactured by Raysearch, Sweden) respectively. Overlap index (OI), Dice similarity index (DSC) and Volume difference (DV) were evaluated based on the auto-contours, and independent-sample t-test analysis is applied to the results. Results: The results of AiContour®© on OI and DSC were better than that of Raystation®© with statistical difference. There was no significant difference in DV between the results of two software. Conclusions: With AiContour®©, auto-contouring results of most organs in the chest and abdomen are good, and with slight modification, it can meet the clinical requirements for planning. With Raystation®©, auto-contouring results in most OAR is not as good as AiContour®©, and only the auto-contouring results of some organs can be used clinically after modification.

scholarly journals Virtual segmentectomy based on high-quality three-dimensional lung modelling from computed tomography images

2013 ◽  
Vol 17 (2) ◽  
pp. 227-232 ◽  
Author(s):  
Hisashi Saji ◽  
Tatsuya Inoue ◽  
Yasufumi Kato ◽  
Yoshihisa Shimada ◽  
Masaru Hagiwara ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Three Dimensional ◽  
High Quality ◽  
Computed Tomography Images

Role of post-mapping computed tomography in virtual-assisted lung mapping

2017 ◽  
Vol 25 (2) ◽  
pp. 123-130 ◽  
Author(s):  
Masaaki Sato ◽  
Kazuhiro Nagayama ◽  
Hideki Kuwano ◽  
Jun-ichi Nitadori ◽  
Masaki Anraku ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Image Guidance ◽  
Virtual Bronchoscopy ◽  
3 Dimensional ◽  
Computed Tomography Images ◽  
Significant Difference ◽  
Marking Technique ◽  
The Difference ◽  
Dye Marking

Background Virtual-assisted lung mapping is a novel bronchoscopic preoperative lung marking technique in which virtual bronchoscopy is used to predict the locations of multiple dye markings. Post-mapping computed tomography is performed to confirm the locations of the actual markings. This study aimed to examine the accuracy of marking locations predicted by virtual bronchoscopy and elucidate the role of post-mapping computed tomography. Methods Automated and manual virtual bronchoscopy was used to predict marking locations. After bronchoscopic dye marking under local anesthesia, computed tomography was performed to confirm the actual marking locations before surgery. Discrepancies between marking locations predicted by the different methods and the actual markings were examined on computed tomography images. Forty-three markings in 11 patients were analyzed. Results The average difference between the predicted and actual marking locations was 30 mm. There was no significant difference between the latest version of the automated virtual bronchoscopy system (30.7 ± 17.2 mm) and manual virtual bronchoscopy (29.8 ± 19.1 mm). The difference was significantly greater in the upper vs. lower lobes (37.1 ± 20.1 vs. 23.0 ± 6.8 mm, for automated virtual bronchoscopy; p < 0.01). Despite this discrepancy, all targeted lesions were successfully resected using 3-dimensional image guidance based on post-mapping computed tomography reflecting the actual marking locations. Conclusions Markings predicted by virtual bronchoscopy were dislocated from the actual markings by an average of 3 cm. However, surgery was accurately performed using post-mapping computed tomography guidance, demonstrating the indispensable role of post-mapping computed tomography in virtual-assisted lung mapping.

scholarly journals In-room computed tomography–based brachytherapy for uterine cervical cancer: results of a 5-year retrospective study

2016 ◽  
Vol 58 (4) ◽  
pp. 543-551 ◽  
Author(s):  
Tatsuya Ohno ◽  
Shin-Ei Noda ◽  
Noriyuki Okonogi ◽  
Kazutoshi Murata ◽  
Kei Shibuya ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Cervical Cancer ◽  
Overall Survival ◽  
Local Control ◽  
External Beam Radiotherapy ◽  
Clinical Stage ◽  
Stage Ii ◽  
High Dose ◽  
Significant Difference ◽  
Grade 3

Abstract Herein, we investigate the long-term clinical outcomes for cervical cancer patients treated with in-room computed tomography–based brachytherapy. Eighty patients with Stage IB1–IVA cervical cancer, who had undergone treatment with combined 3D high-dose rate brachytherapy and conformal radiotherapy between October 2008 and May 2011, were retrospectively analyzed. External beam radiotherapy (50 Gy) with central shielding after 20–40 Gy was performed for each patient. Cisplatin-based chemotherapy was administered concurrently to advanced-stage patients aged ≤75 years. Brachytherapy was delivered in four fractions of 6 Gy per week. In-room computed tomography imaging with applicator insertion was performed for treatment planning. Information from physical examinations at diagnosis, and brachytherapy and magnetic resonance imaging at diagnosis and just before the first brachytherapy session, were referred to for contouring of the high-risk clinical target volume. The median follow-up duration was 60 months. The 5-year local control, pelvic progression-free survival and overall survival rates were 94%, 90% and 86%, respectively. No significant differences in 5-year local control rates were observed between Stage I, Stage II and Stage III–IVA patients. Conversely, a significant difference in the 5-year overall survival rate was observed between Stage II and III–IVA patients (97% vs 72%; P = 0.006). One patient developed Grade 3 late bladder toxicity. No other Grade 3 or higher late toxicities were reported in the rectum or bladder. In conclusion, excellent local control rates were achieved with minimal late toxicities in the rectum or bladder, irrespective of clinical stage.

scholarly journals Measurement of the Endotracheal Tube Diameter Using Computed Tomography Images for Pediatric Patients: Comparison with Classic Formulas

2021 ◽  
Author(s):  
Hatice Dilek Özcanoğlu ◽  
Berna Türkay ◽  
Neşe Kutlutürk Şahin ◽  
Zahide Özlem Ulubay ◽  
Ayşe Sevinç Revanlı ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Endotracheal Tube ◽  
Congenital Heart ◽  
Pediatric Patients ◽  
Transverse Diameter ◽  
External Diameter ◽  
Old Patients ◽  
Computed Tomography Images ◽  
Significant Difference ◽  
The Mean

Objective: In this retrospective study, we aimed to evaluate the compatibility of endotracheal tube (ETT) size used during surgery with tracheal diameter measured by computed tomography(CT) and classic formulas in children who underwent CT imaging preoperatively due to surgical indications. Methods: The study included preoperatively CT scanned, 0-3 year-old patients who were operated due to congenital heart diseases between June 1-October 1, 2018.Using the CT scans, transverse and anteroposterior tracheal diameters were measured from the subglottic level. As these diameters reflect the external diameter of the ETT, the inner diameter-which corresponds the ETT size-was calculated using a correction formula. Besides, Cole’s formula was used to calculate the ETT size for each child, and the tube sizes used during surgeries were obtained from anesthesia charts.ETT sizes were compared. Results: 43 patients (22 girls [51.2%],21 boys [48.8%]) were included.The mean age was 10.5±9.6 months. 18 patients (41.9%) had cyanotic, 25 patients (58.1%) had acyanotic heart disease.Mean corrected transverse and anteroposterior tracheal diameters at CT images were 4.35±0.69mm and 4.30±0.71 mm, respectively. The mean diameter calculated by Cole’s formula was 4.22±0.20 mm. The mean tube size used during surgeries was 4.37±0.60 mm. There was no statistically significant difference between the tube sizes used during the surgeries and the corrected transverse tracheal diameters from CT measurements (p>0.05). But the tube sizes used during the surgeries found significantly larger than the diameters obtained by Cole’s formula (p<0.05). Conclusion: In 0-3 years of age pediatric patients who undergo a congenital heart surgery,corrected tracheal transverse diameter measured by CT is more effective,reliable and less invasive than classic formulas for determining appropriate ETT size

scholarly journals Difference in glenoid retroversion between two-dimensional axial computed tomography and three-dimensional reconstructed images

2020 ◽  
Vol 23 (2) ◽  
pp. 71-79
Author(s):  
Hyungsuk Kim ◽  
Chang Hyun Yoo ◽  
Soo Bin Park ◽  
Hyun Seok Song
Keyword(s):  
Computed Tomography ◽  
Total Shoulder Arthroplasty ◽  
Three Dimensional ◽  
Ct Images ◽  
Body Axis ◽  
Reconstructed Image ◽  
Two Dimensional ◽  
Glenoid Version ◽  
Significant Difference ◽  
Scapular Body

Background: The glenoid version of the shoulder joint correlates with the stability of the glenohumeral joint and the clinical results of total shoulder arthroplasty. We sought to analyze and compare the glenoid version measured by traditional axial two-dimensional (2D) computed tomography (CT) and three-dimensional (3D) reconstructed images at different levels.Methods: A total of 30 cases, including 15 male and 15 female patients, who underwent 3D shoulder CT imaging was randomly selected and matched by sex consecutively at one hospital. The angular difference between the scapular body axis and 2D CT slice axis was measured. The glenoid version was assessed at three levels (midpoint, upper one-third, and center of the lower circle of the glenoid) using Friedman’s method in the axial plane with 2D CT images and at the same level of three different transverse planes using a 3D reconstructed image. Results: The mean difference between the scapular body axis on the 3D reconstructed image and the 2D CT slice axis was 38.4°. At the level of the midpoint of the glenoid, the measurements were 1.7° ± 4.9° on the 2D CT images and −1.8° ± 4.1° in the 3D reconstructed image. At the level of the center of the lower circle, the measurements were 2.7° ± 5.2° on the 2D CT images and −0.5° ± 4.8° in the 3D reconstructed image. A statistically significant difference was found between the 2D CT and 3D reconstructed images at all three levels. Conclusions: The glenoid version is measured differently between axial 2D CT and 3D reconstructed images at three levels. Use of 3D reconstructed imaging can provide a more accurate glenoid version profile relative to 2D CT. The glenoid version is measured differently at different levels.

Sign in / Sign up

Export Citation Format

Share Document