Dataset Of Computed Tomography Scans Research Articles

A novel approach for detection of coronavirus disease from computed tomography scan images using the pivot distribution count method

ABSTRACT Detecting a person infected with coronavirus disease is a challenging task. In some cases, the infection is asymptomatic in nature. A computed tomography scan of images provides information about lungs and helps detect coronavirus disease infected lungs. An accurate algorithm helps the specialist to detect corona virus-infected lungs easily. In this article, a new technique called the pivot distribution count method has been proposed to extract the texture features of computed tomography scanned images of lungs and apply it for the detection and analysis of coronavirus disease. The technique is compared with a recently developed methodology called pixel range calculation method and some state-of-art methods like local binary pattern and gliding box method. The ‘severe acute respiratory syndrome coronavirus-2 computed tomography scan dataset’ was used for our experiments. The experimental results show that the pivot distribution count method produces better accuracy for detecting the infection of coronavirus disease with less computational time. It is also observed that the detection accuracy obtained from coronavirus disease infected images is 98% and from non-infected images is 94% using the pivot distribution count method, which is much higher as compared to other methods.

What is the Orbital Volume of African-Americans?

PurposeMeasurements of orbital volume (OV) are important for reconstructive operations of orbits such as congenital deformities, trauma, and pathology. Successful orbital reconstruction relies on an accurate restoration of OV. The purpose of this study was to identify the OV in the African-American (AA) population. The specific aims were to compare OV of AAs to Caucasians and right-to-left OV in the same racial group. This is the first project in available literature that calculates these OVs. Materials and methodsThe authors implemented a retrospective observational study of successive subjects who received a maxillofacial computed tomography (CT) scan at Grady Memorial Hospital between 2017 and 2020. The primary predictor variable was race. The primary outcome was orbital volume. Two independent examiners calculated OV with an open access OsiriX MD software (version 10.0.5; Pixmeo, Switzerland). All CT scans were acquired using a departmental-based max-face CT scan protocol; 0.6mm slice thickness, 0.5mm slice increment, 100-120kV, 80-440mA, 200-220 FOV, 0.656 Pitch, and a 512 × 512 image matrix. All CT scan data sets were anonymized. The CT images was exported in the digital imaging and communication in medicine (DICOM) format file from Philips Intellispace PACS Enterprise 4.4 software to an external hard drive WD My Passport for Mac. Computer-assisted manual segmentation of OV was performed using an open-access OsiriX MD (software version 10.0.5 [Pixmeo, Switzerland]). The anterior and posterior boundary of the orbit was defined by posterior lacrimal crest and anterior aspect of the optic canal, respectively. Inter-rater reliability was calculated. Differences in means between races, genders, and sides were tested using an independent samples T-test. ResultsSixty subjects (120 orbits) with an average age of 36.7 years (range 22 to 78) met the inclusion criteria. Male-to-female ratio was 1:1. Racial distribution represented 30 AA (50%) and 30 Caucasian patients (50%). Inter-examiner reliability was 0.973. The mean OV of AA and Caucasians was 22.375 cm3 and 23.225 cm3, respectively (P = .07). The mean OV in AA and Caucasian males was 23.919 cm3 and 24.168 cm3, respectively (P = .71). The mean OV in AA and Caucasian females was 20.832 cm3 and 22.282cm3, respectively (P = .013). Gender analysis revealed that AA males had a statistically significant larger OV than AA females (P = <.0001). ConclusionIn conclusion, this study showed that the mean OV for AAs was 22.38 cm3 (male 23.919 cm3; female 20.832 cm3). Caucasian males and females had a larger OV than AA males and females. Caucasian and AA males had a larger OV than Caucasian and AA females. The study findings provide normative values for race and gender between AA and Caucasians. Understanding the average value for OV based on race and gender will assist in preoperative surgical planning of orbital trauma repair and therefore will improve surgical outcome.

Classification of lung nodule malignancy in computed tomography imaging utilising generative adversarial networks and semi-supervised transfer learning

The pulmonary nodules' malignancy rating is commonly confined in patient follow-up; examining the nodule's activity is estimated with the Positron Emission Tomography (PET) system or biopsy. However, these strategies are usually after the initial detection of the malignant nodules acquired from the Computed Tomography (CT) scan. In this study, a Deep Learning methodology to address the challenge of the automatic characterisation of Solitary Pulmonary Nodules (SPN) detected in CT scans is proposed.The research methodology is based on Convolutional Neural Networks, which have proven to be excellent automatic feature extractors for medical images. The publicly available CT dataset, called Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI), and a small CT scan dataset derived from a PET/CT system, is considered the classification target. New, realistic nodule representations are generated employing Deep Convolutional Generative Adversarial Networks to circumvent the shortage of large-scale data to train robust CNNs. Besides, a hierarchical CNN called Feature Fusion VGG19 (FF-VGG19) was developed to enhance feature extraction of the CNN proposed by the Visual Geometry Group (VGG). Moreover, the generated nodule images are separated into two classes by utilising a semi-supervised approach, called self-training, to tackle weak labelling due to DC-GAN inefficiencies.The DC-GAN can generate realistic SPNs, as the experts could only distinguish 23 % of the synthetic nodule images. As a result, the classification accuracy of FF-VGG19 on the LIDC-IDRI dataset increases by +7%, reaching 92.07 %, while the classification accuracy on the CT dataset is increased by 5 %, reaching 84,3 %.

COVID-FACT: A Fully-Automated Capsule Network-Based Framework for Identification of COVID-19 Cases from Chest CT Scans.

The newly discovered Coronavirus Disease 2019 (COVID-19) has been globally spreading and causing hundreds of thousands of deaths around the world as of its first emergence in late 2019. The rapid outbreak of this disease has overwhelmed health care infrastructures and arises the need to allocate medical equipment and resources more efficiently. The early diagnosis of this disease will lead to the rapid separation of COVID-19 and non-COVID cases, which will be helpful for health care authorities to optimize resource allocation plans and early prevention of the disease. In this regard, a growing number of studies are investigating the capability of deep learning for early diagnosis of COVID-19. Computed tomography (CT) scans have shown distinctive features and higher sensitivity compared to other diagnostic tests, in particular the current gold standard, i.e., the Reverse Transcription Polymerase Chain Reaction (RT-PCR) test. Current deep learning-based algorithms are mainly developed based on Convolutional Neural Networks (CNNs) to identify COVID-19 pneumonia cases. CNNs, however, require extensive data augmentation and large datasets to identify detailed spatial relations between image instances. Furthermore, existing algorithms utilizing CT scans, either extend slice-level predictions to patient-level ones using a simple thresholding mechanism or rely on a sophisticated infection segmentation to identify the disease. In this paper, we propose a two-stage fully automated CT-based framework for identification of COVID-19 positive cases referred to as the “COVID-FACT”. COVID-FACT utilizes Capsule Networks, as its main building blocks and is, therefore, capable of capturing spatial information. In particular, to make the proposed COVID-FACT independent from sophisticated segmentations of the area of infection, slices demonstrating infection are detected at the first stage and the second stage is responsible for classifying patients into COVID and non-COVID cases. COVID-FACT detects slices with infection, and identifies positive COVID-19 cases using an in-house CT scan dataset, containing COVID-19, community acquired pneumonia, and normal cases. Based on our experiments, COVID-FACT achieves an accuracy of , a sensitivity of , a specificity of , and an Area Under the Curve (AUC) of 0.98, while depending on far less supervision and annotation, in comparison to its counterparts.

COVID-CT-MD, COVID-19 computed tomography scan dataset applicable in machine learning and deep learning

Novel Coronavirus (COVID-19) has drastically overwhelmed more than 200 countries affecting millions and claiming almost 2 million lives, since its emergence in late 2019. This highly contagious disease can easily spread, and if not controlled in a timely fashion, can rapidly incapacitate healthcare systems. The current standard diagnosis method, the Reverse Transcription Polymerase Chain Reaction (RT- PCR), is time consuming, and subject to low sensitivity. Chest Radiograph (CXR), the first imaging modality to be used, is readily available and gives immediate results. However, it has notoriously lower sensitivity than Computed Tomography (CT), which can be used efficiently to complement other diagnostic methods. This paper introduces a new COVID-19 CT scan dataset, referred to as COVID-CT-MD, consisting of not only COVID-19 cases, but also healthy and participants infected by Community Acquired Pneumonia (CAP). COVID-CT-MD dataset, which is accompanied with lobe-level, slice-level and patient-level labels, has the potential to facilitate the COVID-19 research, in particular COVID-CT-MD can assist in development of advanced Machine Learning (ML) and Deep Neural Network (DNN) based solutions.

Automatic diagnosis of coronavirus (COVID-19) using shape and texture characteristics extracted from X-Ray and CT-Scan images

Automatic diagnosis of coronavirus (COVID-19) is studied in this research. Deep learning methods especially convolutional neural networks (CNNs) have shown great success in COVID-19 diagnosis in recent works. But they are efficient when the depth of network is high enough. However, the use of a deep network requires a sufficiently large training set, which is not available in practice. From the other hand, the use of a shallow CNN may not provide superior results because it is not able to rich feature extraction due to lacking enough convolutional layers. To deal with this difficulty, the contextual features reduced by convolutional filters (CFRCF) is proposed in this work. CFRCF extracts shape and textural features as contextual feature maps from the chest X-ray radiographs and abdominal computed tomography (CT) images. Morphological operators, Gabor filter banks and attribute filters are used for contextual feature extraction. Then, two convolutional filters are applied to the contextual feature cube to extract the nonlinear sub-features and hidden relationships among the contextual features. Finally, a fully connected layer is used to produce a reduced feature vector which is fed to a classifier. Support vector machine and random forest are used as classifier. The experimental results show the superior performance of the proposed method from the recognition accuracy and running time point of view using limited training samples. More than 76% and 94% overall classification accuracy is obtained by the proposed method in CT scan and X-ray images datasets, respectively.

Current Concepts of Contemporary Expandable Lumbar Interbody Fusion Cage Designs, Part 2: Feasibility Assessment of an Endplate Conforming Bidirectional Expandable Interbody Cage.

Expandable cages that allow for bidirectional expansion, in both height and width, may offer benefits over traditional expandable cages or static cages. Effective stiffness must also be considered, as implants with exceedingly high stiffness may increase subsidence risk and reduce graft loading. A retrospective case series of 7 patients were assessed with computed tomography (CT) scan at the final 1-year follow-up to evaluate the interbody fusion and configuration of the expandable cage related to the endplates within the intervertebral space. CT scans were reformatted using cage's tantalum markers as fiducials for single-plane orientation for each intervertebral cage. Device height and width at maximum in situ expansion was measured at its anterior and posterior aspects to evaluate implant deformation. The new bone volume within each cage was measured from the same CT scan data sets and by the Bridwell classification of interbody fusion. The average difference between medial and lateral height measurements was 1.82 mm (±1.08) at the device's anterior aspect and 1.41 mm (±0.98) at the posterior aspect. The average difference between medial and lateral heights was 18.55% (±9.34) anteriorly and 15.49% (±9.24) posteriorly. There was a successful fusion in all 7 patients, as evidenced by measurable bone volume in the center of each interbody cage with an average of 586.42 mm3 (±237.06). The authors demonstrated the feasibility of successfully using bidirectionally expandable multimaterial cages to achieve interbody fusion. These composite open-architecture cages were found to conform to each patient's endplate configuration. The authors' observations support the concept of material selection impacting the effective construct stiffness. The design investigated by the authors provided sufficient anterior column support and successful fusion in all patients. 4.

Applying 3D U-Net Architecture to the Task of Multi-Organ Segmentation in Computed Tomography

Abstract The achievement of high-precision segmentation in medical image analysis has been an active direction of research over the past decade. Significant success in medical imaging tasks has been feasible due to the employment of deep learning methods, including convolutional neural networks (CNNs). Convolutional architectures have been mostly applied to homogeneous medical datasets with separate organs. Nevertheless, the segmentation of volumetric medical images of several organs remains an open question. In this paper, we investigate fully convolutional neural networks (FCNs) and propose a modified 3D U-Net architecture devoted to the processing of computed tomography (CT) volumetric images in the automatic semantic segmentation tasks. To benchmark the architecture, we utilised the differentiable Sørensen-Dice similarity coefficient (SDSC) as a validation metric and optimised it on the training data by minimising the loss function. Our hand-crafted architecture was trained and tested on the manually compiled dataset of CT scans. The improved 3D UNet architecture achieved the average SDSC score of 84.8 % on testing subset among multiple abdominal organs. We also compared our architecture with recognised state-of-the-art results and demonstrated that 3D U-Net based architectures could achieve competitive performance and efficiency in the multi-organ segmentation task.

Hybrid resampling and multi-feature fusion for automatic recognition of cavity imaging sign in lung CT

Abstract The automatic recognition of cavity imaging signs in lung computed tomography (CT) images is of great importance for early diagnosis and possible cure of lung tuberculosis and cancers. The performance of existing recognition methods which adopt classical technology needs to be improved accordingly. In this paper, we propose an automatic recognition method based on hybrid resampling and multi-feature fusion strategies. The hybrid resampling includes multi-receptive-field and multi-window settings: the former reduces the risk of missing small or large cavities, the latter reserves context information of multiply CT windows more compactly. For multi-feature fusion, we extract features of convolutional neural networks (CNN) and classical methods (histograms of oriented gradients (HOG) and local binary pattern (LBP)). Then we compress CNN-HOG features by principal components analysis (PCA) algorithm and combine them with LBP feature. Finally, we use the fused feature to train a support vector machine (SVM) model for improving classification performance. We evaluate our method on the cavity samples from LIDC-IDRI and LISS publicly available dataset of chest CT scans, which contains 167 cavities in 164 CT images. The experimental results show that fused feature has better discriminative capability than any single feature, and has the highest FS score (0.1472 vs 0.1136) in the group with sensitivity greater than 0.8. The proposed method is compared with the latest methods for the automatic recognition of cavity imaging sign and enables higher sensitivity than the second-best method (85% vs 70%). The experiment shows that the fusion of CNN feature and classical hand-crafted feature makes full use of the complementary information, and improves classification performance when number of samples in our application is limited.

Cochlear implant phantom for evaluating computed tomography acquisition parameters.

Cochlear implants (CIs) are surgically implantable neuroprosthetic devices used to treat profound hearing loss. Recent literature indicates that there is a correlation between the final intracochlear positioning of the CI electrode arrays and the ultimate hearing outcome of the patient, indicating that further studies to better understand the relationship between electrode position and outcomes could have significant implications for future surgical techniques, array design, and processor programming methods. Postimplantation high-resolution computed tomography (CT) imaging is the best modality for localizing electrodes and provides the resolution necessary to visually identify electrode position, although with an unknown degree of accuracy depending on image acquisition parameters, like the hounsfield unit (HU) range of reconstruction, orientation, radiation dose, and image resolution. We report on the development of a phantom and on its use to study how four acquisition parameters, including image resolution and HU range of reconstruction, affect how accurately the true position of the electrodes can be found in a dataset of CT scans acquired from multiple helical and cone beam scanners. We also show how the phantom can be used to evaluate the effect of acquisition parameters on automatic electrode localization techniques.

A closer scientific look at foam bitumen

In the asphalt industry, a substantial interest is observed to find possibilities to reduce the production temperature of asphalt mixtures. In the context of this research, new methods for the visualisation of unstable bitumen foam, such as dynamic X-ray radiography, computed tomography (CT) and high-speed camera investigations, have been developed. Moreover, characterisation with empirical methods such as expansion ratio and half-life was determined accurately using ultrasonic measurements. This opens new possibilities to characterise bitumen foam (foaming process) for practical applications. Examination of the foam bitumen stream using a high-speed camera revealed that the foam bitumen contains fragmented pieces of bitumen, which resemble more a liquid than foam. This indicates that the foam is formed afterwards and not, as assumed, within the expansion chamber of the foam generator. In situ thermal imagery of the surface, during the hot foaming process, showed that the temperature distribution depends on the foaming water content (W.C.) and bubble size distribution. Higher W.C. results in more inhomogeneous temperature distribution as compared to lower W.C. (<2 wt%). The dynamic X-radiography results indicated that as the foam decays, the bubble size distribution becomes progressively larger with time for 160°C bitumen temperature. Furthermore, at the beginning of the foam formation, majority of the bubbles is small in cross-section size (0.2–10 mm2). At a later stage, the bubbles become polydisperse. Moreover, theoretical investigations based on the 3D X-ray CT scan data set of bubble merging show that the disjoining pressure increases as the foam film gets thinner with time and finally undergoes rupture. The speed of the bubbles also increases with time when the bubbles are getting closer to each other.

The Utilization of Cranial Models Created Using Rapid Prototyping Techniques in the Development of Models for Navigation Training

Navigation in neurosurgery has expanded rapidly; however, suitable models to train end users to use the myriad software and hardware that come with these systems are lacking. Utilizing three-dimensional (3D) industrial rapid prototyping processes, we have been able to create models using actual computed tomography (CT) data from patients with pathology and use these models to simulate a variety of commonly performed neurosurgical procedures with navigation systems. To assess the possibility of utilizing models created from CT scan dataset obtained from patients with cranial pathology to simulate common neurosurgical procedures using navigation systems. Three patients with pathology were selected (hydrocephalus, right frontal cortical lesion, and midline clival meningioma). CT scan data following an image-guidance surgery protocol in DIACOM format and a Rapid Prototyping Machine were taken to create the necessary printed model with the corresponding pathology embedded. The ability in registration, planning, and navigation of two navigation systems using a variety of software and hardware provided by these platforms was assessed. We were able to register all models accurately using both navigation systems and perform the necessary simulations as planned. Models with pathology utilizing 3D rapid prototyping techniques accurately reflect data of actual patients and can be used in the simulation of neurosurgical operations using navigation systems.

Multi-scale analysis of lung computed tomography images

A computer-aided detection (CAD) system for the identification of lung internal nodules in low-dose multi-detector helical Computed Tomography (CT) images was developed in the framework of the MAGIC-5 project. The three modules of our lung CAD system, a segmentation algorithm for lung internal region identification, a multi-scale dot-enhancement filter for nodule candidate selection and a multi-scale neural technique for false positive finding reduction, are described. The results obtained on a dataset of low-dose and thin-slice CT scans are shown in terms of free response receiver operating characteristic (FROC) curves and discussed.

Po-topic IV-08: 4-dimensional CT imaging of free breathing motion

Modeling of breathing motion is important for the definition of conformal therapy radiation portal boundaries and will be critical to the development of intensity modulated radiation therapy of the thorax and upper abdomen. Development of motion models based on free breathing have been hampered by a lack of volumetric, high-resolution, free-breathing gated image data. We have developed a method for the acquisition of high spatial resolution, tidal volume gated computed tomography (CT) scans acquired during free breathing. A multislice CT scanner is used in ciné mode, where multiple CT scans are acquired without couch motion. After acquisition of a user-specified number of scans (15 scans using a 4-slice scanner in our current studies), the couch is advanced to the next abutting position and another set is acquired. This process is repeated until the entire region of interest is scanned. The patient undergoes quantitative spirometry monitoring during the scanning. The time stamps on the CT scans are used with the spirometry data to determine the patient’s tidal volume during scan acquisition. 3D CT scans are reconstructed for user-specified free breathing tidal volumes by selecting the CT scans at each couch position that were acquired at tidal volumes closest to the desired reconstruction volume. Movie loops can be created by reconstructing a sequence of tidal volumes. Because a finite number of scans are acquired, reconstruction is imperfect. Nevertheless, the desired tidal volume differs only slightly from the closest scanned volume. To date, three patients have been scanned. The breathing tidal volume ranged from 400 mL–600 mL and the standard deviation between desired and scanned tidal volumes ranged from 22 mL-33 mL. Comparisons between the total lung volume change of the reconstructed CT scan datasets and spirometry-measured tidal volume showed excellent agreement. Reconstructed sagittal and coronal images showed remarkably few motion artifacts, even in mid-inspiration tidal volumes. Motion movies of tumors showed nonrigid motion, with tumors deflecting, rotating, and stretching during breathing. This indicates that accurate definition of tumor motion will require nonrigid deformation modeling. Preliminary motion models were developed by manually localizing bronchial branches for a sequence of tidal volume-reconstructed CT scans. First- and second-order polynomial fits of the positions as a function of tidal volume were conducted and compared against the directly measured positions. The typical standard deviation of the fit and measured positions were 2.0 and 1.0 mm for first and second-order fits, respectively. Sources of these differences are the finite tidal volume resolution based on the 15-scan acquisitions, and errors in spirometry drift correction. In conclusion, we have developed a feasible and accurate method for gating computed CT scans to user-specified tidal volumes to construct 3D imaging datasets that reflect lung motion during free breathing. Future projects include automation of the deformable image registration process and IMRT dose calculations in the presence of breathing motion.