Search

Equine parvovirus-hepatitis (EqPV-H) and equine hepacivirus (EqHV) are etiologically associated with Theiler's disease (TD), causing fulminant equine hepatitis, but the transmission route and co-infection effect remain unclear. We determined EqPV-H and EqHV prevalence and coinfection rate in 160 serum and 114 faecal samples using nested polymerase chain reaction. Amino acid and nucleotide analyses were performed and phylogenetic trees were constructed. By measuring liver-specific parameters (AST, GGT, TBIL and A/G ratio), hepatopathological changes in viremia status were compared. We found a high prevalence (EqPV-H: 10.6% in serum, 5.3% in faeces; EqHV: 8.1% in serum) and coinfection rate (35.3% in EqPV-H) of TD-causing agents. The newly identified EqPV-H genomes showed high nucleotide and amino acid similarities with previously reported strains in the USA, China and Austria. In phylogenetic tree and recombination analysis, a natural recombination event was confirmed between Chinese and Korean strains. In the EqPV-H or EqHV viremic horses, AST was significantly elevated and at least two liver-specific parameters were outside the reference intervals in 43.5% (10/23) of horses. To our knowledge, this is the first prevalence field study of EqPV-H and EqHV using both serum and faeces, providing further evidence of faecal-oral transmission of TD. These epidemiologic and clinicopathologic analyses specify the risk factors of TD infection and promote disease prevention strategy.

Equine parvovirus-hepatitis (EqPV-H) is one of the etiological agents of Theiler's disease, causing fulminant hepatitis; however, its transmission route and pathogenesis remain unclear. In the present study, we aimed to determine EqPV-H shedding in oral/nasal/vaginal swabs or semen samples from horses living in Korea using nested polymerase chain reaction. We then used the data obtained to investigate various risk factors associated with EqPV-H including viral shedding, hepatopathological changes, and genetic diversity. Our data revealed occurrence of EqPV-H shedding in these animals (oral: 3/102 [2.9%]; nasal: 3/102 [2.9%]; semen: 1/9 [11.1%]) and identified that both age and country of foaling were significantly associated with EqPV-H shedding (p .05). In addition, we noted that one of the newly isolated strains clustered separately from the other strains in the phylogenetic tree, revealing unique nucleotide and amino acid substitutions. This is a field surveillance study providing evidence of natural and venereal shedding of EqPV-H and describing its presence in both oral/nasal fluids and semen. This epidemiological and clinical analysis may help specify the clinicopathological features of EqPV-H and facilitate the development of novel disease prevention strategies.

In this study, we provide a novel third-order weighted essentially non-oscillatory (WENO) method to solve Hamilton-Jacobi equations. The key idea is to incorporate exponential polynomials to construct numerical fluxes and smoothness indicators. First, the new smoothness indicators are designed by using the finite difference operator annihilating exponential polynomials such that singular regions can be distinguished from smooth regions more efficiently. Moreover, to construct numerical flux, we employ an interpolation method based on exponential polynomials which yields improved results around steep gradients. The proposed scheme retains the optimal order of accuracy (i.e., three) in smooth areas, even near the critical points. To illustrate the ability of the new scheme, some numerical results are provided along with comparisons with other WENO schemes.

The aim of this study is to develop a novel WENO scheme that improves the performance of the well-known fifth-order WENO methods. The approximation space consists of exponential polynomials with a tension parameter that may be optimized to fit the the specific feature of the data, yielding better results compared to the polynomial approximation space. However, finding an optimal tension parameter is a very important and difficult problem, indeed a topic of active research. In this regard, this study introduces a practical approach to determine an optimal tension parameter by taking into account the relationship between the tension parameter and the accuracy of the exponential polynomial interpolation under the setting of the fifth-order WENO scheme. As a result, the proposed WENO scheme attains an improved order of accuracy (that is, sixth-order) better than other fifth-order WENO methods without loss of accuracy at critical points. A detailed analysis is provided to verify the improved convergence rate. Further, we present modified nonlinear weights based on $L^1$-norm approach along with a new global smoothness indicator. The proposed nonlinear weights reduce numerical dissipation significantly, while attaining better resolution in smooth regions. Some experimental results for various benchmark test problems are presented to demonstrate the ability of the new scheme.

Equine parvovirus-cerebrospinal fluid (EqPV-CSF) and eqcopivirus (EqCoPV) are new parvovirus species (EqPVs) identified from various tissues (CSF, blood, and respiratory swabs) in horses with neurologic and respiratory diseases. In this study, we described the prevalence rate of EqPV-CSF and EqCoPV in 133 and 77 serum and fecal samples, respectively, using polymerase chain reaction. Further, we analyzed the potential risk factors for infection. We calculated the nucleotide and amino acid similarity and constructed phylogenetic trees. There was a moderate-to-high prevalence rate (EqPV-CSF: 3.8%, EqCoPV 9.8%) of each virus in serum, moreover, age, country of foaling, and clinical colic signs were significantly associated with the EqPVs infection. The newly identified EqPV-CSF/EqCoPV genomes had high nucleotide and amino acid identities with previously isolated strains in the USA. In phylogenetic analysis, they clustered and formed a new subgroup in the genus Copiparvovirus. To our knowledge, this is the first field epidemiologic study on EqPV-CSF and EqCoPV using both serum and fecal samples. Our findings demonstrate the risk factors for infection and could facilitate the development of disease prevention strategies.

This paper describes how the Big Data Research Center of Kyung Hee University and Benple Inc. developed and deployed an artificial intelligence system to automate the quality management process for Frontec, an SME company that manufactures automobile parts. Various constraints, such as response time requirements and the limited computing resources available, needed to be considered in this project. Defect finders using large-scale images are expected to classify weld nuts within 0.2 s with an accuracy rate of over 95%. Our system uses Circular Hough Transform for preprocessing as well as an adjusted VGG (Visual Geometry Group) model. Our convolutional neural network (CNN) system shows an accuracy of over 99% and a response time of about 0.14 s. To embed the CNN model into the factory, we reimplemented the preprocessing modules using LabVIEW and had the classification model server communicate with an existing vision inspector. We share our lessons from this experience by explain-ing the procedure and real-world issues developing and embedding a deep learn-ing framework in an existing manufacturing environment without implementing any hardware changes.

Hepatitis E virus (HEV) is an emerging zoonotic pathogen causing hepatitis worldwide. Despite the prevalent evidence of interspecies HEV infection in various animal species, the role of horses in HEV epidemiology remains unclear. In this study, we investigated the prevalence of HEV infection in 283 blood and 114 fecal samples from 397 horses using sandwich enzyme-linked immunosorbent assay and nested reverse transcription-polymerase chain reaction. Among the 283 serum samples, 35 were positive for anti-HEV antibodies (12.4%; 95% confidence interval: 8.8-16.8), and four of the five sampling regions (80%) had these seropositive individuals. Analyses of the potential risk factors for HEV infection revealed that racing horses had a significantly higher risk of infection (P = 0.01). However, HEV RNA was not detected in any of the tested serum and fecal samples. To the best of our knowledge, this is the first epidemiological HEV study on horses in Republic of Korea, thereby providing evidence of HEV exposure in the horse population in Korea and specifying the risk factors for HEV infection.

(1) Background: The intestinal microbiota plays an essential role in maintaining the host’s health. Dysbiosis of the equine hindgut microbiota can alter the fermentation patterns and cause metabolic disorders. (2) Methods: This study compared the fecal microbiota composition of horses with intestinal disease and their healthy counterparts living in Korea using 16S rRNA sequencing from fecal samples. A total of 52 fecal samples were collected and divided into three groups: horses with large intestinal disease (n = 20), horses with small intestinal disease (n = 8), and healthy horses (n = 24). (3) Results: Horses with intestinal diseases had fewer species and a less diverse bacterial population than healthy horses. Lactic acid bacteria, Lachnospiraceae, and Lactobacillaceae were overgrown in horses with large intestinal colic. The Firmicutes to Bacteroidetes ratio (F/B), which is a relevant marker of gut dysbiosis, was 1.94, 2.37, and 1.74 for horses with large intestinal colic, small intestinal colic, and healthy horses, respectively. (4) Conclusions: The overgrowth of two lactic acid bacteria families, Lachnospiraceae and Lactobacillaceae, led to a decrease in hindgut pH that interfered with normal fermentation, which might cause large intestinal colic. The overgrowth of Streptococcus also led to a decrease in pH in the hindgut, which suppressed the proliferation of the methanogen and reduced methanogenesis in horses with small intestinal colic.

Patch-based low rank matrix approximation has shown great potential in image denoising. Among state-of-the-art methods in this topic, the weighted nuclear norm minimization (WNNM) has been attracting significant attention due to its competitive denoising performance. For each local patch in an image, the WNNM method groups nonlocal similar patches by block matching to formulate a low-rank matrix. However, the WNNM often chooses irrelevant patches such that it may lose fine details of the image, resulting in undesirable artifacts in the final reconstruction. In this regards, this paper aims to provide a denoising algorithm which further improves the performance of the WNNM method. For this purpose, we develop a new nonlocal similarity measure by exploiting both pixel intensities and gradients and present a filter that enhances edge information in a patch to improve the performance of low rank approximation. The experimental results on widely used test images demonstrate that the proposed denoising algorithm performs better than other state-of-the-art denoising algorithms in terms of PSNR and SSIM indices as well as visual quality.

The aim of this study is to present an improved third-order weighted essentially non-oscillatory (WENO) scheme for solving hyperbolic conservation laws. We first present a novel smoothness indicator by using discrete differential operator which annihilates exponential polynomials. The new smoothness indicator can vanish to zero in smooth regions with higher rates than the classical methods such that it can distinguish the smooth region and discontinuity more efficiently. The proposed scheme achieves the maximal approximation order without loss of accuracy at critical points. A detailed analysis is provided to verify the third-order accuracy. The proposed scheme attains better resolution in smooth regions, while reducing numerical dissipation significantly near singularities. The advantages are more pronounced in two-dimensional model problems. Some numerical experiments are provided to illustrate the performance of the proposed WENO scheme.

Structure-preserving image decomposition separates a given image into structure and texture by smoothing the image, simultaneously preserving or enhancing image edges. The well-studied problem of image decomposition is applied to various areas, such as image smoothing, detail enhancement, non-photorealistic rendering, image artistic rendering, and high-dynamic-range compression. In this paper, we propose a fast algorithm for structure-preserving image decomposition that adopts total variation (TV) minimization to the moving least squares (MLS) method with non-local weights, called structure adaptive TV (SATV) minimization. MLS with non-local weights provides high accuracy approximation that is robust to noise, and allows a fast convergence with TV regularization term. As a result, our proposed SATV preserves the dominant structure while flattening fine-scale details. The experimental results show that the SATV minimization algorithm provides faster and more robust image decomposition than the well-known previous approaches. We demonstrate the usefulness of our algorithm by presenting successful applications in image smoothing and detail enhancement.

Equine parvovirus-hepatitis (EqPV-H) is a newly identified etiologic agent of Theiler’s disease (TD). We present a case of EqPV-H-related fulminant hepatitis in a 14-year-old thoroughbred mare in Korea. The mare had acute hepatopathy and gastrointestinal symptoms, with abnormal liver-related blood parameters. The horse was born in the USA and imported to Korea in 2017, with no history of administration of equine biological products after entry into Korea. The horse was diagnosed with EqPV-H-associated hepatitis after abdominal ultrasonography, laparotomy, and nested polymerase chain reaction (PCR) and in situ hybridization (ISH) assays. The serum, nasal swab, oral swab, and liver biopsy were positive for EqPV-H according to the PCR assay. Genetic analysis of the partial NS1 gene of EqPV-H showed a unique nucleotide substitution, distinct from that in previously deposited strains. EqPV-H DNA was found not only in hepatocytes but also in bile duct epithelium and Kupffer cells, particularly via ISH. To the best of our knowledge, this is the first case of EqPV-H-associated TD in Asia, providing the first clinical evidence for viral shedding from the mouth and nose, and identification of EqPV-H in the liver. This study contributes to a better understanding of the pathological features of EqPV-H-associated TD.

In this study, we present a new sixth-order finite difference weighted essentially non-oscillatory (WENO) scheme for solving Hamilton–Jacobi equations. The proposed scheme recovers the maximal approximation order in smooth regions without loss of accuracy at critical points. We incorporate exponential polynomials into the scheme to obtain better approximation near steep gradients without spurious oscillations. In order to design nonlinear weights based on exponential polynomials, we suggest an alternative approach to construct Lagrange-type exponential functions reproducing the cell-average values of exponential basis functions. Using the Lagrange-type exponential functions, we provide a detailed analysis of the approximation order of the proposed WENO scheme. Compared to other WENO schemes, the proposed scheme is simpler to implement, yielding better approximations with lower computational costs. A number of numerical experiments are presented to demonstrate the performance of the proposed scheme.

In this paper, we present non-uniform subdivision schemes with variable parameter sequences. A locally different tension parameter is set at each edge of the initial control polygon to control locally the shape of the resulting curve such that the scheme becomes non-uniform. Due to the variable parameters, the scheme can reproduce locally different analytic curves such as conics, Lissajous, trigonometric and catenary curves. Hence blending curves including such analytic components can be successfully generated. We discuss the convergence and smoothness of the proposed non-uniform schemes and present some numerical results to demonstrate their advantages in geometric modeling. Furthermore, as an application, we propose a chamfering algorithm which can be used in designing automobile and mechanical products.

The aim of this study is to present a new class of quasi-interpolatory Hermite subdivision schemes of order two with tension parameters. This class extends and unifies some of well-known Hermite subdivision schemes, including the interpolatory Hermite schemes. Acting on a function and the associated first derivative values, each scheme in this class reproduces polynomials up to a certain degree depending on the size of stencil. This is desirable property since the reproduction of polynomials up to degree d leads to the approximation order d + 1 . The smoothness analysis has been performed by using the factorization framework of subdivision operators. Lastly, we present some numerical examples to demonstrate the performance of the proposed Hermite schemes.

(1) Background: The large intestine of horses is an anaerobic fermentative chamber filled with fibrolytic bacteria that play essential roles in digesting and absorbing nutrients for energy production. Although Jeju horses are a prominent local breed in Korea, few studies have investigated the gut microbiota of Jeju horses, (2) Methods: This study performed sequencing of V3 and V4 hypervariable regions of the partial 16S rRNA genes obtained from horse fecal samples and compared the gut microbiota between Jeju and Thoroughbred horses. Thirty and 24 fecal samples were obtained from Jeju and Thoroughbred horses, respectively, (3) Results: The gut microbiota belonged to 23 phyla and 159 families. Firmicutes and Bacteroidetes were the most abundant and predominant phyla, followed by Verrucomicrobia, Euryachaeota, and Spirochaete. The ratio of Firmicutes to Bacteroidetes (F/B), which is known as a relevant marker of gut dysbiosis, was 1.84 for Jeju horses, whereas it was 1.76 for Thoroughbred horses. Moreover, at the genus level, 21 genera were significantly different between the Jeju and Thoroughbred horses (p &lt, 0.05), (4) Conclusions: The Thoroughbred horse’s gut microbiotas had significantly higher diversity than the Jeju horses (p &lt, 0.05). In addition, beneficial commensal bacteria that produce short-chain fatty acids thus providing a significant source of energy are also more abundant in Thoroughbred horses. These results provide novel information on the horse gut microbiota and insights for further studies related to the horse gut microbiota.

Over the past decade, deep learning-based computer vision methods have been shown to surpass previous state-of-the-art computer vision techniques in various fields, and have made great progress in various computer vision problems, including object detection, object segmentation, face recognition, etc. Nowadays, major IT companies are adding new deep-learning-based computer technologies to edge devices such as smartphones. However, since the computational cost of deep learning-based models is still high for edge devices, research is being actively carried out to compress deep learning-based models while not sacrificing high performance. Recently, many lightweight architectures have been proposed for deep learning-based models which are based on low-rank approximation. In this paper, we propose an alternating tensor compose-decompose (ATCD) method for the training of low-rank convolutional neural networks. The proposed training method can better train a compressed low-rank deep learning model than the conventional fixed-structure based training method, so that a compressed deep learning model with higher performance can be obtained in the end of the training. As a representative and exemplary model to which the proposed training method can be applied, we propose a rank-1 convolutional neural network (CNN) which has a structure alternatively containing 3-D rank-1 filters and 1-D filters in the training stage and a 1-D structure in the testing stage. After being trained, the 3-D rank-1 filters can be permanently decomposed into 1-D filters to achieve a fast inference in the test time. The reason that the 1-D filters are not being trained directly in 1-D form in the training stage is that the training of the 3-D rank-1 filters is easier due to the better gradient flow, which makes the training possible even in the case when the fixed structured network with fixed consecutive 1-D filters cannot be trained at all. We also show that the same training method can be applied to the well-known MobileNet architecture so that better parameters can be obtained than with the conventional fixed-structure training method. Furthermore, we show that the 1-D filters in a ResNet like structure can also be trained with the proposed method, which shows the fact that the proposed method can be applied to various structures of networks.

In the article Yuan and Shu (2006) 17, Yuan and Shu have developed discontinuous Galerkin (DG) methods based on non-polynomial approximation spaces for solving time dependent problems. The authors established L 2 error estimates of the proposed methods and also presented a criterion for the choice of basis functions of the non-polynomial spaces to have the same approximation rates as those of polynomial finite element spaces of the same dimension. However, the verification that some of approximation spaces do satisfy the criterion has been performed only to limited types of approximation spaces. In this regards, the aim of this short note is to fill the gap. We prove that the criterion of Yuan and Shu can be satisfied by a wide class of basis functions, including the well-known basis functions such as trigonometric and exponential polynomials.

The aim of this study is to develop a novel sixth-order weighted essentially nonoscillatory (WENO) finite difference scheme. To design new WENO weights, we present two important measurements: a discontinuity detector (at the cell boundary) and a smoothness indicator. The interpolation method is implemented by using exponential polynomials with tension parameters such that they can be tuned to the characteristics of the given data, yielding better approximation near steep gradients without spurious oscillations, compared to the WENO schemes based on algebraic polynomials at lower computational cost. A detailed analysis is performed to verify that the proposed scheme provides the required convergence order of accuracy. Some numerical experiments are presented and compared with other sixth-order WENO schemes to demonstrate the new algorithm's ability.

In this study, we present a new class of quasi-interpolatory non-stationary Hermite subdivision schemes reproducing exponential polynomials. This class extends and unifies the well-known Hermite schemes, including the interpolatory schemes. Each scheme in this family has tension parameters which provide design flexibility, while obtaining at least the same or better smoothness compared to an interpolatory scheme of the same order. We investigate the convergence and smoothness of the new schemes by exploiting the factorization tools of non-stationary subdivision operators. Moreover, a rigorous analysis for the approximation order of the non-stationary Hermite scheme is presented. Finally, some numerical results are presented to demonstrate the performance of the proposed schemes. We find that the quasi-interpolatory scheme can circumvent the undesirable artifacts appearing in interpolatory schemes with irregularly distributed control points.

This paper presents a nonlinear image interpolation algorithm. The suggested method is based on the moving least squares (MLS) projection technique, but introduces a fundamental modification. The algebraic polynomial-based MLS methods provide very satisfactory results. However, the associated approximation space is shift-and-scale invariant so that it cannot be adjusted according to the characteristic of a given data. As a result, when upsampling images, it has a limitation in producing sharp edges such that edges are often blurred in the magnified images. To recover sharper edges, we need to reduce smoothing parameter or adapt a new parameter sharpening the edges. Motivated by this observations, we propose a novel MLS method governed by a set of exponential polynomials with tension parameters such that they can be tuned to the characteristic of given data. Moreover, for a better match to the local structures around the edges, the suggested algorithm uses weights which consider the edge orientation. Numerical results are presented and compared, visually and by using some quantitative fidelity measures (PSNR, EPSNR, SSIM and FSIM), to the bicubic spline interpolation and other recently developed nonlinear methods. The results demonstrate the new algorithm's ability to magnify an image while preserving edge features.

This paper is concerned with fifth-order weighted essentially non-oscillatory (WENO) scheme with a new smoothness indicator. As the so-called WENO-JS scheme (Jiang and Shu in J Comput Phys 126:202---228, 1996) provides the third-order accuracy at critical points where the first and third order derivatives do not becomes zero simultaneously, several fifth-order WENO scheme have been developed through modifying the known smoothness indicators of WENO-JS. Recently a smoothness indicator based on $$L^1$$L1-norm has been proposed by Ha et al. (J Comput Phys 232:68---86, 2013) (denoted by WENO-NS). The aim of this paper is twofold. Firstly, we further improve the smoothness indicator of WENO-NS and secondly, using this measurement, we suggest new nonlinear weights by simplifying WENO-NS weights. The proposed WENO scheme provides the fifth-order accuracy, even at critical points. Some numerical experiments are provided to demonstrate that the present scheme performs better than other WENO schemes of the same order.