Detection of COVID-19 Using Deep Transfer Learning-Based Approach from X-Ray and Computed Tomography(CT) 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.

Download Full-text

Vulnerability in Deep Transfer Learning Models to Adversarial Fast Gradient Sign Attack for COVID-19 Prediction from Chest Radiography Images

Applied Sciences ◽

10.3390/app11094233 ◽

2021 ◽

Vol 11 (9) ◽

pp. 4233

Author(s):

Biprodip Pal ◽

Debashis Gupta ◽

Md. Rashed-Al-Mahfuz ◽

Salem A. Alyami ◽

Mohammad Ali Moni

Keyword(s):

Deep Learning ◽

Transfer Learning ◽

Chest Radiography ◽

High Sensitivity ◽

Ct Images ◽

Classification Performance ◽

Learning Models ◽

X Ray ◽

Image Capturing ◽

Fast Gradient

The COVID-19 pandemic requires the rapid isolation of infected patients. Thus, high-sensitivity radiology images could be a key technique to diagnose patients besides the polymerase chain reaction approach. Deep learning algorithms are proposed in several studies to detect COVID-19 symptoms due to the success in chest radiography image classification, cost efficiency, lack of expert radiologists, and the need for faster processing in the pandemic area. Most of the promising algorithms proposed in different studies are based on pre-trained deep learning models. Such open-source models and lack of variation in the radiology image-capturing environment make the diagnosis system vulnerable to adversarial attacks such as fast gradient sign method (FGSM) attack. This study therefore explored the potential vulnerability of pre-trained convolutional neural network algorithms to the FGSM attack in terms of two frequently used models, VGG16 and Inception-v3. Firstly, we developed two transfer learning models for X-ray and CT image-based COVID-19 classification and analyzed the performance extensively in terms of accuracy, precision, recall, and AUC. Secondly, our study illustrates that misclassification can occur with a very minor perturbation magnitude, such as 0.009 and 0.003 for the FGSM attack in these models for X-ray and CT images, respectively, without any effect on the visual perceptibility of the perturbation. In addition, we demonstrated that successful FGSM attack can decrease the classification performance to 16.67% and 55.56% for X-ray images, as well as 36% and 40% in the case of CT images for VGG16 and Inception-v3, respectively, without any human-recognizable perturbation effects in the adversarial images. Finally, we analyzed that correct class probability of any test image which is supposed to be 1, can drop for both considered models and with increased perturbation; it can drop to 0.24 and 0.17 for the VGG16 model in cases of X-ray and CT images, respectively. Thus, despite the need for data sharing and automated diagnosis, practical deployment of such program requires more robustness.

Download Full-text

Evaluation of Microstructure and Transport Properties of Deteriorated Cementitious Materials from Their X-ray Computed Tomography (CT) Images

Materials ◽

10.3390/ma9050388 ◽

2016 ◽

Vol 9 (5) ◽

pp. 388 ◽

Cited By ~ 18

Author(s):

Michael Promentilla ◽

Shermaine Cortez ◽

Regina Papel ◽

Bernadette Tablada ◽

Takafumi Sugiyama

Keyword(s):

Computed Tomography ◽

Transport Properties ◽

Cementitious Materials ◽

Ct Images ◽

X Ray ◽

X Ray Computed

Download Full-text

Improving grazing-incidence small-angle X-ray scattering–computed tomography images by total variation minimization

Journal of Applied Crystallography ◽

10.1107/s1600576719016558 ◽

2020 ◽

Vol 53 (1) ◽

pp. 140-147

Author(s):

Hiroki Ogawa ◽

Shunsuke Ono ◽

Yukihiro Nishikawa ◽

Akihiko Fujiwara ◽

Taizo Kabe ◽

...

Keyword(s):

Computed Tomography ◽

Image Quality ◽

Total Variation ◽

Small Angle ◽

Grazing Incidence ◽

Ct Images ◽

Measurement Time ◽

X Ray ◽

X Ray Scattering ◽

Ray Scattering

Grazing-incidence small-angle X-ray scattering (GISAXS) coupled with computed tomography (CT) has enabled the visualization of the spatial distribution of nanostructures in thin films. 2D GISAXS images are obtained by scanning along the direction perpendicular to the X-ray beam at each rotation angle. Because the intensities at the q positions contain nanostructural information, the reconstructed CT images individually represent the spatial distributions of this information (e.g. size, shape, surface, characteristic length). These images are reconstructed from the intensities acquired at angular intervals over 180°, but the total measurement time is prolonged. This increase in the radiation dosage can cause damage to the sample. One way to reduce the overall measurement time is to perform a scanning GISAXS measurement along the direction perpendicular to the X-ray beam with a limited interval angle. Using filtered back-projection (FBP), CT images are reconstructed from sinograms with limited interval angles from 3 to 48° (FBP-CT images). However, these images are blurred and have a low image quality. In this study, to optimize the CT image quality, total variation (TV) regularization is introduced to minimize sinogram image noise and artifacts. It is proposed that the TV method can be applied to downsampling of sinograms in order to improve the CT images in comparison with the FBP-CT images.

Download Full-text

Using X-Ray CT Scanning to Study the Failure Mechanism of Concrete under Static and Dynamic Loadings

Advances in Materials Science and Engineering ◽

10.1155/2018/3019158 ◽

2018 ◽

Vol 2018 ◽

pp. 1-9 ◽

Cited By ~ 1

Author(s):

Guangyu Lei ◽

Jichang Han ◽

Faning Dang

Keyword(s):

Computed Tomography ◽

Failure Mechanism ◽

Static Load ◽

Ct Scanning ◽

Ct Images ◽

X Ray ◽

Dynamic Loadings ◽

Ultimate Failure ◽

X Ray Computed ◽

Straight Lines

X-ray images can be used to nondestructively monitor the initiation, extension, and combination of cracks in concrete. In this study, real-time X-ray computed tomography (CT) scanning of concrete specimens under static and dynamic loadings was done. The CT images showed the growth, propagation, and penetration of the cracks and showed the ultimate failure of the concrete samples. Analysis of the CT images and CT numbers showed that the failure followed the structure’s areas of weakness under the static load, but for dynamic loading, the cracks formed very rapidly along straight lines through the aggregate.

Download Full-text

X-Ray Microfocus Radioscopy and Computed Tomography for Failure Analysis

ISTFA 1996: Conference Proceedings from the 22nd International Symposium for Testing and Failure Analysis ◽

10.31399/asm.cp.istfa1996p0077 ◽

1996 ◽

Author(s):

R.H. Bossi ◽

D.A. Cross ◽

R.A. Mickelsen

Keyword(s):

Computed Tomography ◽

Failure Analysis ◽

Conventional Radiography ◽

Ct Images ◽

Ct Image ◽

Video Monitor ◽

Cross Sectional ◽

X Ray ◽

Technical Accuracy ◽

Conventional Film

Abstract X-ray microfocus radioscopy and computed tomography (CT) offer detailed information on the internal assembly and material condition of objects under failure analysis investigation. Using advanced systems for the acquisition of radioscopic and CT images, failure analysis investigations are improved in technical accuracy at a reduced schedule and cost over alternative approaches. A versatile microfocus radioscopic system with CT capability has been successfully implemented as a standard tool in the Boeing Defense & Space Group Failure Analysis Laboratory. Using this tool, studies of electronic, electromechanical and composite material items have been performed. Such a system can pay for itself within two years through higher productivity of the laboratory, increased laboratory value to the company and resolution of critical problems whose worth far exceeds the value of the equipment. The microfocus X-ray source provides projection magnification images that exceed the sensitivity to fine detail that can be obtained with conventional film radiography. Radioscopy, which provides real-time images on a video monitor, allows objects to be readily manipulated and oriented for optimum x-ray evaluation, or monitored during dynamic processes to check performance. Combined with an accurate manipulating stage and data acquisition system x-ray measurements can be used for CT image reconstruction. The CT image provides a cross sectional view of the interior of an object without the interference of superposition of features found in conventional radiography. Accurate dimensional measurements and material constituent identification are possible from the CT images. By taking multiple, contiguous CT slices entire three dimensional data files can be generated of objects.

Download Full-text

Medical Image Analysis and Correlation Between Ankle Fracture Classification and Ankle Computed Tomography

Journal of Medical Imaging and Health Informatics ◽

10.1166/jmihi.2020.3235 ◽

2020 ◽

Vol 10 (12) ◽

pp. 2935-2939

Author(s):

Yugang Teng ◽

Yuanzhen Zhang ◽

Zhenyu Wang

Keyword(s):

Computed Tomography ◽

Cross Section ◽

Ankle Fracture ◽

Medical Image Analysis ◽

Ct Images ◽

Ankle Fractures ◽

Fracture Classification ◽

Horizontal Cross Section ◽

X Ray ◽

The Relationship

Objective: In this paper, we summarize computed tomography (CT) manifestations and characteristics of ankle fractures, and analyze the relationship between CT images and common ankle fracture classifications. Methods: A retrospective survey of 369 adult ankle fractures was performed. CT images of 1 cm horizontal cross-section above the ankle points and their characteristics were analyzed. Ankle fracture X-ray classification was performed, and the relationship between CT images and fracture X-ray classification was analyzed. Results: There is a correlation between CT images and Danis-Weber classification. The incidence of IOL fractures varies with the severity of Danis-Weber classification. After rank correlation test, the difference is statistically significant (Spearman R = 0.781,P < 0.001). CT images can detect IOL fractures that cannot be judged by X-ray fracture classification, and the incidence rate is 5.9%. Conclusions: The 1 cm horizontal cross-section CT image on the ankle point can clearly determine the combined tibiofibular IOL injury before surgery, and it has a good correlation with the Danis-Weber fracture classification, and can detect unexplainable IOL fractures in some radiographs.

Download Full-text

On the use of X-ray computed tomography for determining wood properties: a review1This article is a contribution to the series The Role of Sensors in the New Forest Products Industry and Bioeconomy.

Canadian Journal of Forest Research ◽

10.1139/x11-111 ◽

2011 ◽

Vol 41 (11) ◽

pp. 2120-2140 ◽

Cited By ~ 50

Author(s):

Qiang Wei ◽

Brigitte Leblon ◽

Armand La Rocque

Keyword(s):

Computed Tomography ◽

Growth Ring ◽

Back Propagation ◽

Forest Products ◽

Ct Images ◽

Ct Scanner ◽

Forest Products Industry ◽

X Ray ◽

Feed Forward Back Propagation ◽

X Ray Computed

In several processes of the forest products industry, an in-depth knowledge of log and board internal features is required and their determination needs fast scanning systems. One of the possible technologies is X-ray computed tomography (CT) technology. Our paper reviews applications of this technology in wood density measurements, in wood moisture content monitoring, and in locating internal log features that include pith, sapwood, heartwood, knots, and other defects. Annual growth ring measurements are more problematic to be detected on CT images because of the low spatial resolution of the images used. For log feature identification, our review shows that the feed-forward back-propagation artificial neural network is the most efficient CT image processing method. There are also some studies attempting to reconstruct three-dimensional log or board images from two-dimensional CT images. Several industrial prototypes have been developed because medical CT scanners were shown to be inappropriate for the wood industry. Because of the high cost of X-ray CT scanner equipment, other types of inexpensive sensors should also be investigated, such as electric resistivity tomography and microwaves. It also appears that the best approach uses various different sensors, each of them having its own strengths and weaknesses.

Download Full-text

X-Ray Flow Visualization of a Circular Hydraulic Jump

Volume 1: Symposia, Parts A, B and C ◽

10.1115/fedsm2009-78035 ◽

2009 ◽

Author(s):

Megan R. DiVall ◽

Theodore J. Heindel

Keyword(s):

Computed Tomography ◽

Film Thickness ◽

Flow Visualization ◽

Hydraulic Jump ◽

Electrical Contact ◽

Ct Images ◽

3D Image ◽

X Ray ◽

Before And After ◽

X Ray Computed

The circular hydraulic jump is a product of the impingement of a vertical, circular jet upon a smooth horizontal surface. Previous studies of this phenomenon have used methods such as electrical contact probes, photography, and lasers to measure various features. This study utilizes X-ray computed tomography (CT) to visualize the circular hydraulic jump; analysis is then completed on the reconstructed 3D image. Time-averaged data of the film thickness before and after the jump and the jump radius, as measured from the X-ray CT images, compare well with available literature. Potential imaging improvements with the current equipment have been identified, particularly with respect to measuring film thickness.

Download Full-text