Computed Tomography Image Analysis Research Articles

Gray and White Matters Segmentation in Brain CT Images using Multi-Task Learning from Paired CT and MR Images

Computer tomography (CT) has been routinely used for decades in clinical neuroimaging. Compared to Magnetic resonance imaging (MR), CT is more readily available and more cost-effective, but the soft tissue contrast is much lower. In brain CT images, the unclear soft-tissue boundaries and the high noise level hamper accurate segmentation of the gray and white matter, leaving obstacles for the subsequent geometrical quantification of brain structures. To address this challenge, this paper specifically acquires same-patient CT and MR image pairs and proposes a multi-task learning model for simultaneous tissue segmentation and modality transfer. We aim to use the modality transfer task to learn corresponding MR and CT features which assists the segmentation task to achieve more accurate results than single modality learning. Moreover, we add a Shannon entropy loss function to the training loss to further combat the high noise influence and reduce the fragmentation problem of the segmentation results. Experimental results proved that our multi-task framework achieves more accurate segmentation than the single segmentation task, and the Shannon entropy loss results in much fewer broken brain regions than the state-of-the-art (SOTA) U-net method. Our study provides a useful tool for clinical neural CT image analysis.

Accuracy of virtual rhinomanometry

Abstract Introduction: This paper describes the results of research aimed at developing a method of otolaryngological diagnosis based on computational fluid dynamics, which has been called Virtual Rhinomanometry. Material and methods: Laboratory studies of airflows through a 3D printed model of nasal cavities based on computed tomography image analysis have been performed. The CFD results have been compared with those of an examination of airflow through nasal cavities (rhinomanometry) of a group of 25 patients. Results: The possibilities of simplifying model geometry for CFD calculations have been described, the impact of CT image segmentation on geometric model accuracy and CFD simulation errors have been analysed, and recommendations for future research have been described. Conclusions: The measurement uncertainty of the nasal cavities’ walls has a significant impact on CFD simulations. The CFD simulations better approximate RMM results of patients after anemization, as the influence of the nasal mucosa on airflow is then reduced. A minor change in the geometry of the nasal cavities (within the range of reconstruction errors by CT image segmentation) has a major impact on the results of CFD simulations.

Deep learning-based approach for the automatic segmentation of adult and pediatric temporal bone computed tomography images.

Automatic segmentation of temporal bone computed tomography (CT) images is fundamental to image-guided otologic surgery and the intelligent analysis of CT images in the field of otology. This study was conducted to test a convolutional neural network (CNN) model that can automatically segment almost all temporal bone anatomy structures in adult and pediatric CT images. A dataset comprising 80 annotated CT volumes was collected, of which 40 samples were obtained from adults and 40 from children. A further 60 annotated CT volumes (30 from adults and 30 from children) were used to train the model. The remaining 20 annotated CT volumes were employed to determine the model's generalizability for automatic segmentation. Finally, the Dice coefficient (DC) and average symmetric surface distance (ASSD) were utilized as metrics to evaluate the performance of the CNN model. Two independent-sample t-tests were used to compare the test set results of adults and children. In the adult test set, the mean DC values of all the structures ranged from 0.714 to 0.912, and the ASSD values were less than 0.24 mm for 11 structures. In the pediatric test set, the mean DC values of all the structures ranged from 0.658 to 0.915, and the ASSD values were less than 0.18 mm for 11 structures. There was no statistically significant difference between the adult and child test sets in most temporal bone structures. Our CNN model shows excellent automatic segmentation performance and good generalizability for both adult and pediatric temporal bone CT images, which can help to advance otologist education, intelligent imaging diagnosis, surgery simulation, application of augmented reality, and preoperative planning for image-guided otology surgery.

A comparative study of two automated solutions for cross-sectional skeletal muscle measurement from abdominal computed tomography images.

Measurement of cross-sectional muscle area (CSMA) at the mid third lumbar vertebra (L3) level from computed tomography (CT) images is becoming one of the reference methods for sarcopenia diagnosis. However, manual skeletal muscle segmentation is tedious and is thus restricted to research. Automated solutions are required for use in clinical practice. The aim of this study was to compare the reliability of two automated solutions for the measurement of CSMA. We conducted a retrospective analysis of CT images in our hospital database. We included consecutive individuals hospitalized at the Grenoble University Hospital in France between January and May 2018 with abdominal CT images and sagittal reconstruction. We used two types of software to automatically segment skeletal muscle: ABACS, a module of the SliceOmatic software solution "ABACS-SliceOmatic," and a deep learning-based solution called "AutoMATiCA." Manual segmentation was performed by a medical expert to generate reference data using "SliceOmatic." The Dice similarity coefficient (DSC) was used to measure overlap between the results of the manual and the automated segmentations. The DSC value for each method was compared with the Mann-Whitney U test. A total of 676 hospitalized individuals was retrospectively included (365 males [53.8%] and 312 females [46.2%]). The median DSC for SliceOmatic vs AutoMATiCA (0.969 [5th percentile: 0.909]) was greater than the median DSC for SliceOmatic vs. ABACS-SliceOmatic (0.949 [5th percentile: 0.836]) (p<0.001). AutoMATiCA, which used artificial intelligence, was more reliable than ABACS-SliceOmatic for skeletal muscle segmentation at the L3 level in a cohort of hospitalized individuals. The next step is to develop and validate a neural network that can identify L3 slices, which is currently a fastidious process.

Evaluation of transport properties of deteriorated concrete due to calcium leaching with coupled CT image analysis and random walk simulation

Concrete may deteriorate when chronically in contact with water due to the leaching of calcium ions. This study conducted a natural diffusion test for calcium ions with a minute cylindrical specimen of 3 mm in diameter and 6 mm in height to determine leaching at 20 and 80 °C in 33 days. One piece of limestone as aggregate was inserted inside each specimen to investigate the effect of the interfacial transition zone (ITZ). Changes in mass transfer in the leaching and non-leaching regions in bulk cement paste were evaluated using synchrotron X-ray computed tomography (CT) in a micrometer order. A deeper dissolution front of Ca(OH)2 inside the specimen was observed for a higher temperature leaching test. Random walk simulation was performed to obtain diffusion parameters. Results showed that the calcium diffusion coefficient of non-deteriorated ITZ is about ten times as high as that of the non-leaching region. For the leaching region, the diffusion coefficient is about 50 times and 100 times as high as that of the non-leaching region at 20 °C and 80 °C, respectively. In addition, the Ca(OH)2 dissolution front appeared deeper inside the specimen in the direction where aggregate is close or exposed to the dissolution front because of ITZ.

Determination of patient-specific airway stent fit using novel 3D reconstruction measurement techniques: a 4-year follow-up of a patient.

Managing complex benign airway disease is a major challenge in interventional pulmonology. With the introduction of additive manufacturing in the medical field, patient-specific (PS) implants are an innovate prospect for airway management. Historically, stents were oversized to resist migration. However, the optimal degree and impact of stent oversizing remains unclear. The ability to design stents based on computed tomography (CT) invites opportunity to understand sizing. Here, we report a novel three-dimensional (3D) image reconstruction tool to quantify fit repeatedly over time. Analysis of CT imaging before and after successive stent implants in a single patient with different areas of stenosis and malacia was done. Nine PS airway stents over 4 years (five left mainstem and four right mainstem) were studied. The distance between the airway model and stent was calculated. The CT images were correlated to stent designs in CloudCompare software (v2.10-alpha) for novel analysis. Heat map was exported depicting the distances between the airway and the stent to the clinician's prescribed stent model. Corresponding histograms containing distances, mean, and standard deviation were reported. It is possible to measure stent fit based on heat map quantification on patient imaging. Observation of the airway over time and stent change suggests that the airway became more open over time requiring increased stent diameters. The ability to design and measure stent fit over time can help quantify the utility and impact of PS silicone airway stent. The airway appears to display plasticity such that there is notable change in stent prescription over time.

Anatomical Consideration for Anterior Approach of Cervicothoracic Junction: A Computed Tomography Image Analysis.

In the cervicothoracic junction (CTJ), there is limited working space to perform the posterior-only approach. Therefore, a combined anterior approach is required in some cases. However, the great vessels and sternum obstruct the anterior corridor and make the anterior approach difficult. We analyzed relevant anatomical structures encountered during the anterior approach in the CTJ and evaluated the feasibility of previously reported surgical corridors. We retrospectively examined 49 patients who underwent neck computed tomography angiography between January 2015 and May 2020. Using the coronal images, we measured the intercarotid artery angle (ICAA), intercarotid artery distance (ICAD), shape of the brachiocephalic trunk (BCT), and position of the BCT base. We then measured the most cranial level requiring manubriotomy for the anterior approach (ML), the most caudal level accessible through the superior corridor (SC), and the most caudal level through the inferior corridor (IC) according to the surgeon's line of sight using the sagittal axis image. The mean ICAA and ICAD were 50.83° ± 15.23° and 33.38 ± 12.11 mm, respectively. Notably, BCT shape was of the convex type in most cases (42.9%), followed by the straight type (36.7%). In addition, the base of BCT was most commonly located inside the body (49%). Moreover, ICAA and ICAD were significantly greater in men. Although men mostly had the BCT base inside the body (64.3%), female mostly had it on the edge of the body (47.6%). Notably, ML showed the highest frequency (16.3%) in the T1 lower and upper bodies. Furthermore, through SC and IC, it was possible to approach the T4 lower body and T6 midbody, respectively. SC showed the highest frequency (16.3%) in the T3 lower body, and IC showed the highest frequency (20.4%) in the T5 midbody. ICAA and ICAD were larger and higher in men. BCT was convex and located inside the body in most cases. The accessible level of ML, SC, and IC were T1, T3, and T5, respectively. For the anterior approach in the CTJ, preoperative vascular and accessible level analysis of corridors is essential to decide on the appropriate corridor and reduce complications.

Deep learning for diagnosis of malign pleural effusion on computed tomography images

BackgroundThe pleura is a serous membrane that surrounds the lungs. The visceral surface secretes fluid into the serous cavity and the parietal surface ensures a regular absorption of this fluid. If this balance is disturbed, fluid accumulation occurs in the pleural space called “Pleural Effusion”. Today, accurate diagnosis of pleural diseases is becoming more critical, as advances in treatment protocols have contributed positively to prognosis. Our aim is to perform computer-aided numerical analysis of Computed Tomography (CT) images from patients showing pleural effusion images on CT and to examine the prediction of malignant/benign distinction using deep learning by comparing with the cytology results. MethodsThe authors classified 408 CT images from 64 patients whose etiology of pleural effusion was investigated using the deep learning method. 378 of the images were used for the training of the system; 15 malignant and 15 benign CT images, which were not included in the training group, were used as the test. ResultsAmong the 30 test images evaluated in the system; 14 of 15 malignant patients and 13 of 15 benign patients were estimated with correct diagnosis (PPD: 93.3%, NPD: 86.67%, Sensitivity: 87.5%, Specificity: 92.86%). ConclusionAdvances in computer-aided diagnostic analysis of CT images and obtaining a pre-diagnosis of pleural fluid may reduce the need for interventional procedures by guiding physicians about which patients may have malignancies. Thus, it is cost and time-saving in patient management, allowing earlier diagnosis and treatment.

The Strength and Elastic Modulus of Pervious Concrete Considering Pore and Fiber during Freeze–Thaw Cycles

The interface between aggregate and cement matrix and the strength of the cementation layer between aggregates are the key factors affecting the strength of pervious concrete. The purpose of this paper is to evaluate the effects of porosity, compressive strength and elastic modulus of pervious concrete before and after freeze–thaw cycles. The effective porosity and total porosity were obtained by the underwater weighing and CT (computed tomography) image analysis methods. Uniaxial and triaxle compression tests were carried out to obtain the strength and elastic modulus of pervious concrete considering pore and fiber. The results indicated that the effective modulus and effective stress were closely related to the porosity, and the continuous fracture of cementation points between aggregates caused damage to pervious concrete. Inclined shear failure of pervious concrete occurred under uniaxial pressure, and the strength and elastic modulus increased with increases in confining pressure. With the increase in freeze–thaw cycles, the porosity increased linearly, the strength and elastic modulus decreased and a mutational point appeared between 40 and 50 times during the freeze–thaw cycles. The fiber enhanced the strength of the cementation layer and increased the connection between aggregates, thus improving the strength and integrity of pervious concrete. This work is needed to serve as a reference for the fracture mechanism of pervious concrete and the effect of freeze–thaw cycles considering pore and fiber.

Effects of Benralizumab on Three-Dimensional Computed Tomography Analysis in Severe Eosinophilic Asthma

Introduction: Benralizumab, an anti-interleukin-5 receptor chain monoclonal antibody, is used to treat severe asthma and control asthma symptoms or exacerbations. The aim of this study was to examine the changes in airway morphology using computed tomography (CT) images in accordance with clinical efficacy following the administration of benralizumab. Methods: The clinical efficacy of benralizumab was evaluated in 11 patients with severe asthma by analyzing the changes in parameters, such as the asthma control test, asthma quality of life questionnaire, pulmonary function, and exacerbation count. We also investigated the airway wall thickness of the right bronchus (B1) and the total airway count (TAC) using CT images. Results: Most patients treated with benralizumab showed improvements in asthma symptoms and exacerbations. CT imaging analyses showed a decrease in the right B¹ airway wall thickness and an increase in the TAC. Correlations between blood eosinophil count and changes in CT imaging were observed. Discussion/Conclusion: The data suggested that benralizumab has the potential to improve airway wall thickening and ventilation by alleviating the obstruction and clearing an obstructed airway.

Computed tomography image analysis for Indonesian total hip arthroplasty designs

&lt;span lang="EN-US"&gt;Total hip arthroplasty purposes to replace a hip joint damaged by an artificial hip joint. However, the developed products that already exist in the market lead to the mismatch between the hip implant equipment and the patient’s bone morphometric. Besides causing complications, the mismatch also continues to the dislocation effects, fracture, osteolysis, and thigh pain. This paper aims to design a customized hip implant based on real patient data, particularly for Indonesian patient, limited to the acetabular components and stem parts. The computed images were analyzed to estimate the patient proximal femur morphometric; those are the femoral head diameter, neck-shaft angle, mediolateral width, anteroposterior width, neck length and neck width. The experiment has succeeded in designing the acetabular shell with the thickness of 3 mm, the acetabular liner with the thickness of 6 mm, the femoral head between 22.4 to 24.8 mm, the short stem in both the right for 110.656 mm and left femur bone for 111.49 mm; that fit the patient's femur bone. Overall, the proposed steps in designing the customized hip implant in this work, based on image analysis on medical imaging data, can be a standard to be applied for other patient-needs hip arthroplasty implants.&lt;/span&gt;

Improving tuberculosis severity assessment in computed tomography images using novel DAvoU-Net segmentation and deep learning framework

Tuberculosis is a bacterial infection that usually affects the lungs; however, it can be treated if diagnosed earlier. Effective analysis of computed tomography images of lungs captured from tuberculosis patients is one of the effective approaches to determining the disease's severity. Previous medical computed tomography image analysis algorithms have suffered many setbacks during the segmentation process. In this study, a novel approach to image segmentation is proposed. DAvoU-Net model comprises Multi-Scale Residual Block and Receptive Dense Connection to segment tuberculosis-affected areas from original CT images. The 2-dimensional weight parameters of seven different deep pre-trained neural networks were first converted into 3-dimensional weights using a rotation operation before being used to train a new 3-dimensional convolutional neural network. The extracted image features from the 3-dimensional convolutional neural network serve as input into a bi-directional long short-time memory for extracting high-level discriminative features. The proposed DAvoU-Net model outperformed the original U-Net model in segmenting ImageCLEF 2019 tuberculosis dataset. It generated a dice score of 98.34 % compared to 84.21 % produced by the original U-Net. Overall, the combination of DAvoU-Net + ResNet-50 along with a 3-dimensional convolutional neural network and bi-directional long short-time memory performed better than the six other deep pre-trained neural networks. It produced an overall Area Under the ROC Curve (AUC) and a Classification Accuracy (ACC) of 0.8513 and 0.8119 respectively.

Angioplasty or sacrifice is better in multiple renal artery grafts: a computed tomography image analysis study

Taiki Ogasa, Yuki Nakagawa, Shuji Isotani, Kenta Futamura, Manabu Okada, Takahisa Hiramitsu, Norihiko Goto, Shunji Narumi, Yoshihiko Watarai. Korean J Transplant 2022;36:281. https://doi.org/10.4285/ATW2022.F-4242

Development of the sphenoid sinus from newborn to age 18: A computed tomography imaging analysis

The sphenoid sinus (SS) is an important anatomical formation in terms of its location and relationship with important neurovascular structures. It is essential to evaluate the pneumatization and dimensions of the SS with Computed Tomography (CT) in the preoperative period of transsphenoidal interventions to be performed on tumors located in this region. The aim of this study was to evaluate the correlation of the volume and dimensions of the SS measured with CT with age in the period from birth to 18 years of age. In this retrospective study, a total of 360 brain, paranasal sinus, temporal bone and maxillofacial CT scans of individuals from birth to 18 years of age were evaluated. The cohort was divided into 18 groups with 20 (10 female, 10 male) patients in each age group. The height, width, length and volume values of the SS were measured on CT and their correlations with age and the differences between genders were evaluated statistically. It has been determined that the increases in the size and volume of the SS accelerate especially between 1 and 2 years of age and begins to reach adult dimensions after 12 years of age. Height and length of the sphenoid sinus showed a high correlation coefficient (r=0.717 and r=0.731 respectively) with age, while its width and volume showed a moderate correlation coefficient (r=0.662 and 0.543 respectively). There was no statistically significant difference between the genders in terms of SS height, width, length and volume. SS dimensions and volume showed a logaritmic growth pattern from birth to 18 years. The results of this study may help to better understand the normal development of SS in children and may serve as a reference for more comprehensive CT studies. In addition, it can help surgeons to evaluate the developmental characteristics of SS in surgical interventions for SS in children.

Multi-domain CT translation by a routable translation network

Objective. To unify the style of computed tomography (CT) images from multiple sources, we propose a novel multi-domain image translation network to convert CT images from different scan parameters and manufacturers by simply changing a routing vector. Approach. Unlike the existing multi-domain translation techniques, our method is based on a shared encoder and a routable decoder architecture to maximize the expressivity and conditioning power of the network. Main results. Experimental results show that the proposed CT image conversion can minimize the variation of image characteristics caused by imaging parameters, reconstruction algorithms, and hardware designs. Quantitative results and clinical evaluation from radiologists also show that our method can provide accurate translation results. Significance. Quantitative evaluation of CT images from multi-site or longitudinal studies has been a difficult problem due to the image variation depending on CT scan parameters and manufacturers. The proposed method can be utilized to address this for the quantitative analysis of multi-domain CT images.

Computed tomography features of discospondylitis in dogs.

Computed tomography (CT) findings of dogs with discospondylitis have not been widely described despite increased availability of this imaging modality. Describe the CT features of discospondylitis in a population of clinically affected dogs with discospondylitis diagnosed by magnetic resonance imaging (MRI). Forty-one dogs (63 affected discs) with MRI-identified discospondylitis presented to a single referral hospital between 2012 and 2022. Retrospective, single center, descriptive case series with analysis of MRI-identified discospondylitis sites and concomitant CT imaging. Computed tomographic features of MRI-affected sites including intervertebral disc space (IVDS), endplates, vertebral body, epidural space and paraspinal tissues were described. The most frequently found changes were: (1) endplate involvement (87.3%) most frequently bilateral (94.5%), with erosion (61.9%) and multifocal osteolysis (67.3%); (2) periosteal proliferation adjacent to the IVDS (73%) and spondylosis (66.7%); and (3) vertebral body involvement (66.7%) involving one-third of the vertebra (85.7%) with multifocal osteolysis (73.5%). Other less prevalent features included an abnormal IVDS (narrowed or collapsed), sclerosis of the adjacent vertebral body or endplates, presence of disseminated idiopathic skeletal hyperostosis or vacuum artifact. We determined that bilateral endplate erosion and periosteal proliferation were very common in dogs with discospondylitis. Careful evaluation of CT in all 3 planes (dorsal, sagittal, transverse) is necessary to identify an affected IVDS. These described CT features can aid in the diagnosis of discospondylitis in dogs but equivocal cases might still require MRI.

T-SPOT with CT image analysis based on deep learning for early differential diagnosis of nontuberculous mycobacteria pulmonary disease and pulmonary tuberculosis

ObjectivesThis study aimed to establish a diagnostic algorithm combining T-SPOT with computed tomography image analysis based on deep learning (DL) for early differential diagnosis of nontuberculous mycobacteria pulmonary disease (NTM-PD) and pulmonary tuberculosis (PTB). MethodsA total of 1049 cases were enrolled, including 467 NTM-PD and 582 PTB cases. A total of 320 cases (160 NTM-PD and 160 PTB) were randomized as the testing set and were analyzed using T-SPOT combined with the DL model. The testing cases were first divided into T-SPOT-positive and -negative groups, and the DL model was then used to separate the cases into four subgroups further. ResultsThe precision was found to be 91.7% for the subgroup of T-SPOT-negative and DL classified as NTM-PD, and 89.8% for T-SPOT-positive and DL classified as PTB, which covered 66.9% of the total cases, compared with the accuracy rate of 80.3% of T-SPOT alone. In the other two remaining groups, where the T-SPOT prediction was inconsistent with the DL model, the accuracy was 73.0% and 52.2%, separately. ConclusionOur study shows that the new diagnostic system combining T-SPOT with DL based computed tomography image analysis can greatly improve the classification precision of NTM-PD and PTB when the two methods of prediction are consistent.

Method of evaluating cement-to-formation bond strength with computed tomography image analysis

Invention of new cements for effective zonal isolation requires bond strength evaluation of the cement to formation. We present a method to evaluate the cement-to-formation bond strength based on the flow of cement through a formation using image analysis. Most researchers perform image analysis on a single slice. This is the first known nondestructive method devised using a quantitative analysis on a sequence of slices from X-ray computed tomography (CT) scans. The number of CT scan slices in which the porosity changes is used to evaluate the bond strength of cement. Our findings demonstrate that the cement-to-formation bond strength is affected by permeability of a formation and viscosity of cement if other physical (sample preparation and measuring conditions) and chemical (cement additives) variables remain constant.

Correlation Analysis of Computed Tomography Features and Pathological Types of Multifocal Ground-Glass Nodular Lung Adenocarcinoma.

To investigate the correlation between computed tomography (CT) image characteristics of multiple lung ground-glass nodules (GGNs) and pathological classification, the CT image data of multiple lung GGN patients confirmed by pathology (n = 132) in our hospital were collected. The imaging features of GGNs were analyzed by qualified physicians, including lesion size (diameter, volume, and mass), location, CT values (mean and relative CT values), lesion morphology (round and irregular), marginal structure (pagination and burr), internal structure (bronchial inflation sign), and adjacent structure (pleural depression). CT imaging analysis was performed for the subtype of infiltrating adenocarcinoma (IAC). In CT findings, GGNs were greatly different from adenomatous hyperplasia (AAH), pure GGN adenocarcinoma in situ (AIS), and microinvasive adenocarcinoma (MIA) in terms of marginal structure, lesion morphology, internal structure, adjacent structure, and size (P < 0.05). The mean and relative CT values of mural adenocarcinoma, acinar adenocarcinoma, and papillary adenocarcinoma of IAC subtypes were greatly different from those of AAH/AIS/MIA (P < 0.05). In summary, the CT images of GGNs can be used as the basis for the differentiation of AAH, AIS, and MIA early noninvasive types and IAC invasive types, and the CT value of the IAC subtype can be used as the basis for the classification and differentiation of IAC pathological subtypes.

Integration of X-ray radiography and automated mineralogy data for the optimization of ore sorting routines

X-ray transmission sorting is arguably the most successful ore sorting technique used in the mining industry today. In order to establish the suitability of X-ray transmission sorting for a specific ore deposit or raw material type, the current state of the art involves time-consuming and costly empirical testing. In this paper, we show how the success of X-ray transmission sorting can be reliably predicted based on X-ray radiographs that are quick and inexpensive to obtain for a large number of samples. We document this novel approach using a large suite of samples of scheelite ores from the Mittersill deposit in Austria. Using only the X-ray radiographs at a spatial resolution of 0.12 mm we quantify the volume of high-density scheelite at different spatial resolution. The results are compared to and validated by data obtained by computer tomography and scanning electron microscopy-based image analysis. The X-ray radiograph data set is then down sampled to spatial resolutions of 0.8 mm, 3.2 mm, 12 mm and 24 mm in order to gauge the optimum spatial resolution prior to any empirical testing. Upgrading curves can be calculated, and in combination with quantitative data obtained by complementary analytical methods, the mineralogical and chemical composition of concentrate and waste can be predicted. The workflow can be transferred without any problems to many other ore types with a significant density difference between ore mineral(s) and gangue minerals.