YOLOv7-TS: A Traffic Sign Detection Model Based on Sub-Pixel Convolution and Feature Fusion.

In recent years, significant progress has been witnessed in the field of deep learning-based object detection. As a subtask in the field of object detection, traffic sign detection has great potential for development. However, the existing object detection methods for traffic sign detection in real-world scenes are plagued by issues such as the omission of small objects and low detection accuracies. To address these issues, a traffic sign detection model named YOLOv7-Traffic Sign (YOLOv7-TS) is proposed based on sub-pixel convolution and feature fusion. Firstly, the up-sampling capability of the sub-pixel convolution integrating channel dimension is harnessed and a Feature Map Extraction Module (FMEM) is devised to mitigate the channel information loss. Furthermore, a Multi-feature Interactive Fusion Network (MIFNet) is constructed to facilitate enhanced information interaction among all feature layers, improving the feature fusion effectiveness and strengthening the perception ability of small objects. Moreover, a Deep Feature Enhancement Module (DFEM) is established to accelerate the pooling process while enriching the highest-layer feature. YOLOv7-TS is evaluated on two traffic sign datasets, namely CCTSDB2021 and TT100K. Compared with YOLOv7, YOLOv7-TS, with a smaller number of parameters, achieves a significant enhancement of 3.63% and 2.68% in the mean Average Precision (mAP) for each respective dataset, proving the effectiveness of the proposed model.

An interpretable shapelets-based method for myocardial infarction detection using dynamic learning and deep learning.

Objective.Myocardial infarction (MI) is a prevalent cardiovascular disease that contributes to global mortality rates. Timely diagnosis and treatment of MI are crucial in reducing its fatality rate. Currently, electrocardiography (ECG) serves as the primary tool for clinical diagnosis. However, detecting MI accurately through ECG remains challenging due to the complex and subtle pathological ECG changes it causes. To enhance the accuracy of ECG in detecting MI, a more thorough exploration of ECG signals is necessary to extract significant features.Approach.In this paper, we propose an interpretable shapelet-based approach for MI detection using dynamic learning and deep learning. Firstly, the intrinsic dynamics of ECG signals are learned through dynamic learning. Then, a deep neural network is utilized to extract and select shapelets from ECG dynamics, which can capture locally specific ECG changes, and serve as discriminative features for identifying MI patients. Finally, the ensemble model for MI detection is built by integrating shapelets of multi-dimensional ECG dynamic signals.Main results.The performance of the proposed method is evaluated on the public PTB dataset with accuracy, sensitivity, and specificity of 94.11%, 94.97%, and 90.98%.Significance.The shapelets obtained in this study exhibit significant morphological differences between MI and healthy subjects.

Deterministic learning-based neural identification and knowledge fusion.

Recent deterministic learning methods have achieved locally-accurate identification of unknown system dynamics. However, the locally-accurate identification means that the neural networks can only capture the local dynamics knowledge along the system trajectory. In order to capture a broader knowledge region, this article investigates the knowledge fusion problem of deterministic learning, that is, the integration of different knowledge regions along different individual trajectories. Specifically, two kinds of knowledge fusion schemes are systematically introduced: an online fusion scheme and an offline fusion scheme. The online scheme can be viewed as an extension of distributed cooperative learning control to cooperative neural identification for sampled-data systems. By designing an auxiliary information transmission strategy to enable the neural network to receive information learned from other tasks while learning its own task, it is proven that the weights of all localized RBF networks exponentially converge to their common true/ideal values. The offline scheme can be regarded as a knowledge distillation strategy, in which the fused network is obtained by offline training through the knowledge learned from all individual system trajectories via deterministic learning. A novel weight fusion algorithm with low computational complexity is proposed based on the least squares solution under subspace constraints. Simulation studies show that the proposed fusion schemes can successfully integrate the knowledge regions of different individual trajectories while maintaining the learning performance, thereby greatly expanding the knowledge region learned from deterministic learning.

Dynamic learning from adaptive neural control for full-state constrained strict-feedback nonlinear systems.

This study focuses on the learning and control issues of strict-feedback systems with full-state constraints. To achieve learning capability under constraints, transformation mapping is utilized to convert the original system with full-state constraints into a quasi-pure-feedback unconstrained system. Utilizing the system transformation technique, only a single neural network (NN) is required to identify the unknown dynamics within the transformed system. Combining the dynamic surface control design, a novel adaptive neural control scheme is developed to ensure that all closed-loop signals are uniformly bounded, and every system state remains within the predefined constraint range. In addition, the precise convergence of NN weights is further transformed into an exponential stability problem for a category of linear time-varying systems under persistent excitation conditions. Subsequently, the converged NN weights are efficiently stored and utilized to create a learning controller to achieve better control performance while abiding by the full-state constraints. The viability of this control strategy is demonstrated via simulations.

Structural maintenance of chromosomes (SMC) proteins are required for DNA elimination in Paramecium.

Chromosome (SMC) proteins are a large family of ATPases that play important roles in the organization and dynamics of chromatin. They are central regulators of chromosome dynamics and the core component of condensin. DNA elimination during zygotic somatic genome development is a characteristic feature of ciliated protozoa such as Paramecium This process occurs after meiosis, mitosis, karyogamy, and another mitosis, which result in the formation of a new germline and somatic nuclei. The series of nuclear divisions implies an important role of SMC proteins in Paramecium sexual development. The relationship between DNA elimination and SMC has not yet been described. Here, we applied RNA interference, genome sequencing, mRNA sequencing, immunofluorescence, and mass spectrometry to investigate the roles of SMC components in DNA elimination. Our results show that SMC4-2 is required for genome rearrangement, whereas SMC4-1 is not. Functional diversification of SMC4 in Paramecium led to a formation of two paralogues where SMC4-2 acquired a novel, development-specific function and differs from SMC4-1. Moreover, our study suggests a competitive relationship between these two proteins.

Comparative study of the dynamics of colloidal glass and gel.

We investigate and compare the difference in the dynamics of two arrested states: colloidal glass and colloidal gel. Real-space experiments reveal two distinct nonergodicity origins for their slow dynamics, namely, cage effects for the glass and attractive bondings for the gel. Such distinct origins lead to a faster decay of the correlation function and a smaller nonergodicity parameter of the glass than those of the gel. We also find that the gel exhibits stronger dynamical heterogeneity compared with the glass due to the greater correlated motions in the gel. Moreover, a logarithmic decay in the correlation function is observed as the two nonergodicity origins merge, consistent with the mode coupling theory.

New Results on Rapid Dynamical Pattern Recognition via Deterministic Learning From Sampling Sequences.

Rapid dynamical pattern recognition based on the deterministic learning method (DLM-based RDPR) aims to rapidly recognize the most similar dynamical pattern pair from perspectives of differences in inherent system dynamics. The basic mechanism is to use available recognition errors to reflect the differences in the dynamics of dynamical pattern pairs and then to make a decision based on a minimal recognition error (MRE) principle. This article focuses on providing a rigorous theoretical analysis of the MRE principle in DLM-based RDPR under the sampled-data framework. Specifically, we seek a unified methodology from the similarity definition to the measure implementation and then to derive general sufficient conditions and necessary conditions for the MRE principle. The main idea is to: 1) from the average signal energy aspect, define a time-dependent dynamics-based similarity in dynamical pattern pairs and reestablish the measure of recognition errors generated from the DLM-based RDPR; 2) introduce the energy-based Lyapunov method to establish the interrelation between the dynamical distance and the recognition error; and 3) derive sufficient conditions and necessary conditions from two directions of the interrelation. The proposed conditions distinguish themselves from virtually all of the existing DLM-based RDPR works with only sufficient conditions in the sense that it is shown in a rigorous analysis that under what conditions, the pattern pair recognized based on the MRE principle is indeed the most similar one. Therefore, the proposed work makes the DLM-based RDPR possess good interpretability and provides strong theoretical guidance in engineering applications.

Deterministic learning-based neural control for output-constrained strict-feedback nonlinear systems.

This paper studies learning from adaptive neural control of output-constrained strict-feedback uncertain nonlinear systems. To overcome the constraint restriction and achieve learning from the closed-loop control process, there are several significant steps. Firstly, a state transformation is introduced to convert the original constrained system output into an unconstrained one. Then an equivalent n-order affine nonlinear system is constructed based on the transformed unconstrained output state in norm form by the system transformation method. By combining dynamic surface control (DSC) technique, an adaptive neural control scheme is proposed for the transformed system. Then all closed-loop signals are uniformly ultimately bounded and the system output tracks the expected trajectory well with satisfying the constraint requirement. Secondly, the partial persistent excitation condition of the radial basis function neural network (RBF NN) could be verified to achieve. Therefore, the uncertain dynamics can be precisely approximated by RBF NN. Subsequently, the learning ability of RBF NN is achieved, and the knowledge acquired from the neural control process is stored in the form of constant neural networks (NNs). By reutilizing the knowledge, a novel learning controller is established to improve the control performance when facing the similar or same control task. The proposed learning control (LC) scheme can avoid repeating the online adaptation of neural weight estimates, which saves computing resources and improves transient performance. Meanwhile, the LC method significantly raises the tracking accuracy and the speed of error convergence while satisfying of the constraint condition simultaneously. Simulation studies demonstrate the efficiency of this proposed control scheme.

A dynamic learning-based ECG feature extraction method for myocardial infarction detection.

Objective.Myocardial infarction (MI) is one of the leading causes of human mortality in all cardiovascular diseases globally. Currently, the 12-lead electrocardiogram (ECG) is widely used as a first-line diagnostic tool for MI. However, visual inspection of pathological ECG variations induced by MI remains a great challenge for cardiologists, since pathological changes are usually complex and slight.Approach.To have an accuracy of the MI detection, the prominent features extracted from in-depth mining of ECG signals need to be explored. In this study, a dynamic learning algorithm is applied to discover prominent features for identifying MI patients via mining the hidden inherent dynamics in ECG signals. Firstly, the distinctive dynamic features extracted from the multi-scale decomposition of dynamic modeling of the ECG signals effectively and comprehensibly represent the pathological ECG changes. Secondly, a few most important dynamic features are filtered through a hybrid feature selection algorithm based on filter and wrapper to form a representative reduced feature set. Finally, different classifiers based on the reduced feature set are trained and tested on the public PTB dataset and an independent clinical data set.Main results.Our proposed method achieves a significant improvement in detecting MI patients under the inter-patient paradigm, with an accuracy of 94.75%, sensitivity of 94.18%, and specificity of 96.33% on the PTB dataset. Furthermore, classifiers trained on PTB are verified on the test data set collected from 200 patients, yielding a maximum accuracy of 84.96%, sensitivity of 85.04%, and specificity of 84.80%.Significance.The experimental results demonstrate that our method performs distinctive dynamic feature extraction and may be used as an effective auxiliary tool to diagnose MI patients.

Observer-based dynamical pattern recognition via deterministic learning.

In this paper, based on the sampled-data observer and the deterministic learning theory, a rapid dynamical pattern recognition approach is proposed for univariate time series composed of the output signals of the dynamical systems. Specifically, locally-accurate identification of inherent dynamics of univariate time series is first achieved by using the sampled-data observer and the radial basis function (RBF) networks. The dynamical estimators embedded with the learned knowledge are then designed by resorting to the sampled-data observer. It is proved that generated estimator residuals can reflect the difference between the system dynamics of the training and test univariate time series. Finally, a recognition decision-making scheme is proposed based on the residual norms of the dynamical estimators. Through rigorous analysis, recognition conditions are given to guarantee the accurate recognition of the dynamical pattern of the test univariate time series. The significance of this paper lies in that the difficult problems of dynamical modeling and rapid recognition for univariate time series are solved by incorporating the sampled-data observer design and the deterministic learning theory. The effectiveness of the proposed approach is confirmed by a numerical example and compressor stall warning experiments.

Crushing Characteristics Comparison between Aluminum/CFRP and Aluminum/CFRP/Aluminum Hybrid Tubes.

This study experimentally investigated the axial crushing characteristics of the hybrid tubes with the configuration of aluminum/carbon fiber-reinforced polymer (CFRP) (1/1) and aluminum/CFRP/aluminum (2/1). The effects of geometry size and fiber lay-up sequence on the axial crushing energy-absorption performances and failure modes of the two types of hybrid tubes were compared. The results showed that the energy absorption of the specimens with [0°/90°] lay-up sequence was better than that of the ones with [45°/-45°] lay-up sequence for both types of hybrid tubes. The proper length of the tubes should be selected to avoid too small a length-to-diameter ratio so that a stable and controllable progressive crushing failure mode can be achieved. When the crushing failure process was relatively stable, the specific energy absorption and crushing force efficiency of the 2/1 hybrid tubes were not affected by the geometric size. The energy absorption of the hybrid tubes was higher than the sum of the energy absorption of all the corresponding individual tubes, showing a positive hybrid effect.

Transcriptomic landscape of hepatic lymph nodes, peripheral blood lymphocytes and spleen of swamp buffaloes infected with the tropical liver fluke Fasciola gigantica.

The tropical liver fluke Fasciola gigantica is a parasitic helminth that has been frequently reported to infect mammals, typically involving water buffaloes. In this study, we characterized the tissue transcriptional landscape of buffaloes following infection by F. gigantica. RNAs were isolated from hepatic lymph nodes (hLNs), peripheral blood lymphocytes (pBLs), and spleen at 3-, 42- and 70-days post-infection (dpi), and all samples were subjected to RNA sequencing analyses. At 3 dpi, 2603, 460, and 162 differentially expressed transcripts (DETs) were detected in hLNs, pBLs, and spleen, respectively. At 42 dpi, 322, 937, and 196 DETs were detected in hLNs, pBLs, and spleen, respectively. At 70 dpi, 376, 334, and 165 DETs were detected in hLNs, pBLs, and spleen, respectively. Functional enrichment analysis identified upregulated immune-related pathways in the infected tissues involved in innate and adaptive immune responses, especially in hLNs at 42 and 70 dpi, and pBLs at 3 and 42 dpi. The upregulated transcripts in spleen were not enriched in any immune-related pathway. Co-expression network analysis further identified transcriptional changes associated with immune response to F. gigantica infection. Receiver operating characteristic (ROC) curve analysis showed that 107 genes in hLNs, 32 genes in pBLs, and 36 genes in spleen correlated with F. gigantica load. These findings provide new insight into molecular mechanisms and signaling pathways associated with F. gigantica infection in buffaloes.

Proteomic Profiling of the Liver, Hepatic Lymph Nodes, and Spleen of Buffaloes Infected with Fasciola gigantica.

In the present study, we used an isobaric tag for relative and absolute quantitation (iTRAQ) proteomics technology to characterize the differentially expressed proteins (DEPs) in the liver, hepatic lymph nodes (hLNs), and spleen of buffaloes infected with Fasciola gigantica (F. gigantica). We also used the parallel reaction monitoring (PRM) method to verify the expression levels of the DEPs in the three infected tissues. At three days post-infection (dpi), 225, 1821, and 364 DEPs were detected in the liver, hLNs, and spleen, respectively. At 42 dpi, 384, 252, and 214 DEPs were detected in the liver, hLNs, and spleen, respectively. At 70 dpi, 125, 829, and 247 DEPs were detected in the liver, hLNs, and spleen, respectively. Downregulation of metabolism was prominent in infected livers at all time points, and upregulation of immune responses was marked in the hLNs during early infection (three dpi); however, no changes in the immune response were detected at the late stages of infection (42 and 70 dpi). Compared to the hLNs, there was no significant upregulation in the levels of immune responses in the infected spleen. All the identified DEPs were used to predict the subcellular localization of the proteins, which were related to extracellular space and membrane and were involved in host immune responses. Further PRM analysis confirmed the expression of 18 proteins. These data provide the first simultaneous proteomic profiles of multiple organs of buffaloes experimentally infected with F. gigantica.

iTRAQ-based Quantitative Proteomics Analysis Identifies Host Pathways Modulated during Toxoplasma gondii Infection in Swine.

Toxoplasma gondii is a leading cause of foodborne illness and consumption of undercooked pig meat is a major risk factor for acquiring toxoplasmosis, which causes a substantial burden on society. Here, we used isobaric tags for relative and absolute quantification (iTRAQ) labelling coupled with liquid chromatography-tandem mass spectrometry (LC-MS/MS) to identify cellular proteins and pathways altered during T. gondii infection in pigs. We also used parallel reaction monitoring-based LC-MS/MS to verify the levels of protein expression of infected spleens and mesenteric lymph nodes (MLNs). At 6 days post-infection (dpi), 156, 391, 170, 292, and 200 differentially expressed proteins (DEPs) were detected in the brain, liver, lung, MLNs and spleen, respectively. At 18 dpi, 339, 351, 483, 388, and 303 DEPs were detected in the brain, liver, lung, MLNs and spleen, respectively. Although proteins involved in immune responses were upregulated in all infected tissues, protein expression signature in infected livers was dominated by downregulation of the metabolic processes. By weighted gene co-expression network analysis, we could further show that all proteins were clustered into 25 co-expression modules and that the pink module significantly correlated with the infection status. We also identified 163 potential anti-T. gondii proteins (PATPs) and provided evidence that two PATPs (HSP70.2 and PDIA3) can reduce T. gondii burden in porcine macrophages in vitro. This comprehensive proteomics analysis reveals new facets in the pathogenesis of T. gondii infection and identifies key proteins that may contribute to the pig's defense against this infection.

Complex and dynamic transcriptional changes allow the helminth Fasciola gigantica to adjust to its intermediate snail and definitive mammalian hosts.

BACKGROUND: The tropical liver fluke, Fasciola gigantica causes fasciolosis, an important disease of humans and livestock. We characterized dynamic transcriptional changes associated with the development of the parasite in its two hosts, the snail intermediate host and the mammalian definitive host. RESULTS: Differential gene transcription analysis revealed 7445 unigenes transcribed by all F. gigantica lifecycle stages, while the majority (n = 50,977) exhibited stage-specific expression. Miracidia that hatch from eggs are highly transcriptionally active, expressing a myriad of genes involved in pheromone activity and metallopeptidase activity, consistent with snail host finding and invasion. Clonal expansion of rediae within the snail correlates with increased expression of genes associated with transcription, translation and repair. All intra-snail stages (miracidia, rediae and cercariae) require abundant cathepsin L peptidases for migration and feeding and, as indicated by their annotation, express genes putatively involved in the manipulation of snail innate immune responses. Cercariae emerge from the snail, settle on vegetation and become encysted metacercariae that are infectious to mammals; these remain metabolically active, transcribing genes involved in regulation of metabolism, synthesis of nucleotides, pH and endopeptidase activity to assure their longevity and survival on pasture. Dramatic growth and development following infection of the mammalian host are associated with high gene transcription of cell motility pathways, and transport and catabolism pathways. The intra-mammalian stages temporally regulate key families of genes including the cathepsin L and B proteases and their trans-activating peptidases, the legumains, during intense feeding and migration through the intestine, liver and bile ducts. While 70% of the F. gigantica transcripts share homology with genes expressed by the temperate liver fluke Fasciola hepatica, gene expression profiles of the most abundantly expressed transcripts within the comparable lifecycle stages implies significant species-specific gene regulation. CONCLUSIONS: Transcriptional profiling of the F. gigantica lifecycle identified key metabolic, growth and developmental processes the parasite undergoes as it encounters vastly different environments within two very different hosts. Comparative analysis with F. hepatica provides insight into the similarities and differences of these parasites that diverged > 20 million years ago, crucial for the future development of novel control strategies against both species.

Global Transcriptome Profiling of Multiple Porcine Organs Reveals Toxoplasma gondii-Induced Transcriptional Landscapes.

We characterized the porcine tissue transcriptional landscapes that follow Toxoplasma gondii infection. RNAs were isolated from liver, spleen, cerebral cortex, lung, and mesenteric lymph nodes (MLNs) of T. gondii-infected and uninfected (control) pigs at days 6 and 18 postinfection, and were analyzed using next-generation sequencing (RNA-seq). T. gondii altered the expression of 178, 476, 199, 201, and 362 transcripts at 6 dpi and 217, 223, 347, 119, and 161 at 18 dpi in the infected brain, liver, lung, MLNs and spleen, respectively. The differentially expressed transcripts (DETs) were grouped into five expression patterns and 10 sub-clusters. Gene Ontology enrichment and pathway analysis revealed that immune-related genes dominated the overall transcriptomic signature and that metabolic processes, such as steroid biosynthesis, and metabolism of lipid and carboxylic acid, were downregulated in infected tissues. Co-expression network analysis identified transcriptional modules associated with host immune response to infection. These findings not only show how T. gondii infection alters porcine transcriptome in a tissue-specific manner, but also offer a gateway for testing new hypotheses regarding human response to T. gondii infection.

Metabolomic signature of mouse cerebral cortex following Toxoplasma gondii infection.

BACKGROUND: The protozoan parasite Toxoplasma gondii infects and alters the neurotransmission in cerebral cortex and other brain regions, leading to neurobehavioral and neuropathologic changes in humans and animals. However, the molecules that contribute to these changes remain largely unknown. METHODS: We have investigated the impact of T. gondii infection on the overall metabolism of mouse cerebral cortex. Mass-spectrometry-based metabolomics and multivariate statistical analysis were employed to discover metabolomic signatures that discriminate between cerebral cortex of T. gondii-infected and uninfected control mice. RESULTS: Our results identified 73, 67 and 276 differentially abundant metabolites, which were involved in 25, 37 and 64 pathways at 7, 14 and 21 days post-infection (dpi), respectively. Metabolites in the unsaturated fatty acid biosynthesis pathway were upregulated as the infection progressed, indicating that T. gondii induces the biosynthesis of unsaturated fatty acids to promote its own growth and survival. Some of the downregulated metabolites were related to pathways, such as steroid hormone biosynthesis and arachidonic acid metabolism. Nine metabolites were identified as T. gondii responsive metabolites, namely galactosylsphingosine, arachidonic acid, LysoSM(d18:1), L-palmitoylcarnitine, calcitetrol, 27-Deoxy-5b-cyprinol, L-homophenylalanine, oleic acid and ceramide (d18:1/16:0). CONCLUSIONS: Our data provide novel insight into the dysregulation of the metabolism of the mouse cerebral cortex during T. gondii infection and have important implications for studies of T. gondii pathogenesis.

Global serum proteomic changes in water buffaloes infected with Fasciola gigantica.

BACKGROUND: The liver fluke Fasciola gigantica modulates several signaling pathways in infected buffaloes to facilitate its survival and establishment of persistent infection. In response to the parasite invasion, buffaloes activate innate and adaptive immune responses to counter the parasite infection. To detect new proteins that might be involved in the interaction between F. gigantica and the buffaloes, and that also might serve as biomarkers for fasciolosis, we used proteomic techniques to study the serum proteome of buffaloes during F. gigantica infection. Here, we used an isobaric tags for relative and absolute quantitation (iTRAQ)-based quantitative proteomic approach to identify serum proteins that are differentially expressed in infected buffaloes compared to uninfected control buffaloes. Additionally, we applied a parallel reaction monitoring (PRM) assay to validate specific proteins identified by the iTRAQ method. RESULTS: A total of 313, 459 and 399 proteins were identified at 3, 42 and 70 days post-infection, respectively; of these 92, 93 and 138 were differentially abundant proteins. Some of the identified differentially abundant proteins, including complement factor H related 5, complement component C6, complement component C7, amine oxidase, plasma serine protease inhibitor and lysozyme, are known to be involved in complement system activation, blood coagulation, platelet activation, lymphocyte's adhesion and lysozyme hydrolysis. Analysis of data for all three time points after infection identified six significantly upregulated proteins in infected serum that separated infected and uninfected buffaloes into distinct clusters. Further PRM analysis confirmed the expression of five proteins, namely MHC class I antigen, Beta-2-microglobulin, NID2 protein, Fetuin-B and Fibrinogen gamma-B chain. CONCLUSIONS: These findings provide novel insights into the serum proteomics signature of buffaloes during F. gigantica infection.

Th2-related cytokines are associated with Fasciola gigantica infection and evasion in the natural host, swamp buffalo.

The infection of ruminants by Fasciola spp. always induces a non-protective Th2-type immune response. However, little is known about changes in the local and systemic immune environment during F. gigantica migration in buffalo. In this study, native swamp buffaloes were each infected with 500 viable F. gigantica metacercariae. Mesenteric lymph node (MLN), hepatic lymph node (HLN), spleen, and serum samples were collected from control and infected buffaloes at 3, 10, 28, 42, 70, and 98 days post-infection (DPI). The mRNA expression levels of the Th1- and Th2-related cytokines IL-2, IL-4, IL-5, IL-6, IL-10, IL-12p40, IFN-γ, TNF-α, and CD4 were measured during different infection stages in the MLNs, spleens, and HLNs using quantitative real-time PCR (qRT-PCR). Levels of the specific anti-ESP isotype antibodies IgG, IgG1, and IgG2 were used to reflect changes in humoral immunity. The results of this study indicated that swamp buffaloes were susceptible to F. gigantica infection, and that susceptibility to this infection was closely related to the cytokine environment associated with the Th2-type immune response. The MLNs showed a mixed Th1- and Th2-type immune response during the acute infection stages, after which the production of these cytokines returned to normal. Cytokine expression in the HLNs also expressed a mixed Th1- and Th2-type immune response during the early infection stages. When the infection became chronic, the typical Th2 immune response was induced in the HLNs. At the acute infection stages, the spleen exhibited a Th2 immune response. Nevertheless, cytokines associated with the Th1 and Th2 immune responses were upregulated at 98 DPI. In addition, the total IgG and IgG1 of the parasite-specific antibodies increased. This suggested that the Th2-related cytokines and IgG1 induced by F. gigantica infection might mediate successful F. gigantica infection in the natural host, swamp buffalo.

Vehicle Detection in Urban Traffic Surveillance Images Based on Convolutional Neural Networks with Feature Concatenation.

Vehicle detection with category inference on video sequence data is an important but challenging task for urban traffic surveillance. The difficulty of this task lies in the fact that it requires accurate localization of relatively small vehicles in complex scenes and expects real-time detection. In this paper, we present a vehicle detection framework that improves the performance of the conventional Single Shot MultiBox Detector (SSD), which effectively detects different types of vehicles in real-time. Our approach, which proposes the use of different feature extractors for localization and classification tasks in a single network, and to enhance these two feature extractors through deconvolution (D) and pooling (P) between layers in the feature pyramid, is denoted as DP-SSD. In addition, we extend the scope of the default box by adjusting its scale so that smaller default boxes can be exploited to guide DP-SSD training. Experimental results on the UA-DETRAC and KITTI datasets demonstrate that DP-SSD can achieve efficient vehicle detection for real-world traffic surveillance data in real-time. For the UA-DETRAC test set trained with UA-DETRAC trainval set, DP-SSD with the input size of 300 × 300 achieves 75.43% mAP (mean average precision) at the speed of 50.47 FPS (frames per second), and the framework with a 512 × 512 sized input reaches 77.94% mAP at 25.12 FPS using an NVIDIA GeForce GTX 1080Ti GPU. The DP-SSD shows comparable accuracy, which is better than those of the compared state-of-the-art models, except for YOLOv3.