Search

Automatic and efficient 3D object modeling has become critical in industrial applications. The advancement of deep convolutional neural networks (CNNs) has prompted researchers to use CNNs for learning 3D geometry information directly from images. However, the feature maps directly extracted by CNNs are more suitable for image processing tasks because they contain more deep texture information of the entire 2D image. Compared with this, 3D reconstruction tasks using CNNs demand geometric information about a specific object. The existing architecture mainly tries to infer the geometric structure through texture information, which leads to an uneven distribution of points in the generated point cloud object. To address this problem, we propose a cascade point set generation network (CPSGN) that deforms the predicted object while more effectively inferring the object’s 3D geometric information from the 2D image so that the distribution of the final object becomes more uniform and denser. The CPSGN consists of a point set generation part for producing a basic 3D object and a point deformation part for fine-tuning the basic 3D object. In addition, we designed a projection loss that optimizes the geometry of the model by measuring shape differences from multiple perspectives. Experimental results on different benchmark datasets indicate that the produced point-based model outperforms existing approaches.

Under hypoxia, mouse embryonic stem cells (mESCs) lose the ability to self-renew and begin to differentiate through down-regulation of LIFR-STAT3 pathway via hypoxia-inducible factor-1α (HIF-1α). However, it remains largely unknown what kinds of factors are involved in hypoxia-induced differentiation of mESCs. This study aims to identify the differentially expressed genes (DEGs) in early differentiation of mESCs under hypoxia. Here we utilized a Genefishing techniqueTM to discover the new DEGs during hypoxia-induced early differentiation in CCE mESCs. Next, we investigated the role of DEGs using morphological observation, alkaline phosphatase (ALP) assay, STAT3 activation analysis, and biomarkers analysis for stemness. We detected 19 DEGs under hypoxia and performed cloning with sequencing in six genes. We confirmed the expression patterns of five DEGs including H2afz and GOT1 by realtime PCR assay. Among them, H2afz was significantly decreased under hypoxia, depending on HIF-1α. H2afz-overexpressing CCE mESCs maintained their ALP activity and stem cell markers (Nanog and Rex1), even in hypoxic condition. On the other hand, the early differentiation markers such as FGF5 and STAT5a, which had been increased in hypoxic conditions, were reduced by H2afz overexpression. We discovered that H2afz could be a new target gene that functions in hypoxia-induced differentiation in mESCs and have revealed that it is involved in maintaining the pluripotency of mESCs in the early stages of differentiation. These findings will provide insights into mechanisms of hypoxia-mediated differentiation of mESCs during early development.

The relationship between the process condition of wire-drawing and the microstructural evolution during the subsequent annealing was investigated for the production of high-purity copper wires. Three wire-drawing process conditions were considered according to the reduction in the area per pass and the die angle, and their effects on the evolution of the grain structure and texture during annealing were investigated. Strain distributions, calculated using the finite element method, showed that the shear strain near the wire surface varies according to the process condition. The electron backscatter diffraction results indicate that the location on the cross-section of the wire, where secondary recrystallization initiates during annealing, is closely related to the shear strain during wire-drawing. Meanwhile, the effect of shear deformation on the grain structure and texture after primary recrystallization is insignificant. Drawing process conditions with a small reduction in the area per pass and die angle resulted in a relatively small shear strain and gave rise to the onset of secondary recrystallization at the central region of the wires. In contrast, the condition with a large reduction in the area per pass and die angle led to a high shear strain and the onset of secondary recrystallization at the peripheral region of the wire. The observed primary and secondary recrystallization behavior during annealing was explained from the viewpoint of dynamic material phenomena and inhomogeneity in the local recrystallization texture, regulated by the extent of shear deformation during wire-drawing.

The western Korean Peninsula preserves evidence for multiple Proterozoic to Phanerozoic orogenic processes along the convergent continental margins. After the Late Palaeozoic to Early Mesozoic coll...

Natural environments usually have a larger dynamic range than the dynamic range that can be acquired by an optical camera with a single shot. In this paper, we propose a multiexposure fusion method that effectively fuses in a direct manner differently exposed images of a high dynamic range scene into a high-quality image. First, we present a developed joint weight by considering the exposure level measurement of the local and global luminance components of the input images. Second, we introduce a designed multiscale edge-preserving smoothing (MEPS) model for direct representing the weight maps. Third, two scale-aware factors for the MEPS model are adaptively determined without manual interference to obtain an optimal representation effect for each scale of the weight maps. The proposed adaptive MEPS model does not require Gaussian filtering steps to first smoothen the weight maps. It significantly reduces spatial artifacts in the fused image. We compare the proposed method with eight existing methods on 30 sequences from two databases with different characteristics. The experimental results indicate that the proposed approach achieves better imaging performance than the existing state-of-the-art methods, on both quantitative and qualitative evaluation. Moreover, it maintains a high computational efficiency.

Background Cancer stem-like cells (CSCs) play a pivotal role in lung tumor formation and progression. Nerve injury-induced protein 1 (Ninjurin1, Ninj1) has been implicated in lung cancer; however, the pathological role of Ninj1 in the context of lung tumorigenesis remains largely unknown. Methods The role of Ninj1 in the survival of non-small cell lung cancer (NSCLC) CSCs within microenvironments exhibiting hazardous conditions was assessed by utilizing patient tissues and transgenic mouse models where Ninj1 repression and oncogenic KrasG12D/+ or carcinogen-induced genetic changes were induced in putative pulmonary stem cells (SCs). Additionally, NSCLC cell lines and primary cultures of patient-derived tumors, particularly Ninj1high and Ninj1low subpopulations and those with gain- or loss-of-Ninj1 expression, and also publicly available data were all used to assess the role of Ninj1 in lung tumorigenesis. Results Ninj1 expression is elevated in various human NSCLC cell lines and tumors, and elevated expression of this protein can serve as a biomarker for poor prognosis in patients with NSCLC. Elevated Ninj1 expression in pulmonary SCs with oncogenic changes promotes lung tumor growth in mice. Ninj1high subpopulations within NSCLC cell lines, patient-derived tumors, and NSCLC cells with gain-of-Ninj1 expression exhibited CSC-associated phenotypes and significantly enhanced survival capacities in vitro and in vivo in the presence of various cell death inducers. Mechanistically, Ninj1 forms an assembly with lipoprotein receptor-related protein 6 (LRP6) through its extracellular N-terminal domain and recruits Frizzled2 (FZD2) and various downstream signaling mediators, ultimately resulting in transcriptional upregulation of target genes of the LRP6/β-catenin signaling pathway. Conclusions Ninj1 may act as a driver of lung tumor formation and progression by protecting NSCLC CSCs from hostile microenvironments through ligand-independent activation of LRP6/β-catenin signaling.

Classical Chinese poetry is a kind of formatted text with phonological patterns. It has been a big challenge for the classical Chinese poetry generation. Because deep learning technology has made great progress in image and text processing, we present a classical Chinese poetry generation model based on Transformer-XL that can capture longer-term dependency than the original transformer. Our model creates the first character according to the title specified by a user and then concatenates the generated character as the input to predict the next character until reaches the predefined number of characters. Experimental results show that our model can generate poems that have good format, consistency and rhythm.

In recent years, palm vein recognition has obtained significant attention as its uniqueness, stable features, and high recognition rate. Although state-of-art deep learning methods can outperform several research domains, the lack of sufficiently large data for vein-based biometric recognition can suffer from generalization problems and degrades the model accuracy. Our approach trained Generative Adversarial Nets (GAN) with triplet loss for classification as an additional task. Lately, triplet networks are widely applied as it learns the latent space representation between neighbors and performs significantly higher accuracy even for insufficient data size. Moreover, in practical application, the quality of acquired vein images is low due to external factors and affects the recognition accuracy. To overcome this problem, we propose a CNN-based Encoder-Decoder network for vein segmentation to utilize the accuracy performance. Jerman enhancement filter is applied to enhance the vein ROI images for labeling the ground truth mask images for training the Encoder-Decoder network.

There are two stages in bottom-up human pose estimation method, joint detection and joint candidate grouping. Optimizing grouping algorithms can significantly improve the performance of pose estimation. In this paper, we introduce body center, a center point of person instance, and anchor point, a corresponding assistant position point, for the task of grouping. The anchor point is the center of joint and body center, which can help joint grouping to the corresponding person instance like an anchor. The body center and anchor point can be predicted simultaneously with the joint candidate by the same backbone. So, this new grouping method can fully exploit the features extracted by the step of joint detection. On the COCO keypoints dataset, the proposed method performs on par with the existing state-of-the-art bottom-up method in accuracy.

The Infrared and visible image fusion technique is to generate an integrated image that can simultaneously preserve more texture information and thermal target from the raw images. To achieve this goal, a new infrared and visible fusion based on a multi-scale decomposition and texture preservation model is proposed. First, the base layers and detail layers images are obtained through a novel multi-scale decomposition method. Then, an adaptive saliency weighting rule is designed to obtain the fused base image. To maintain important image information from the raw images as much as possible, a texture preservation model is present. Specifically, we first apply a “max-absolute” rule to obtain pre-fused images and then calculate a Frobenius norm operator between pre-fused images and the target detail fusion image. Finally, the merged image can be obtained through an add operator. Experimental results show that compared with other state-of-the-art fusion methods, our method can preserve the texture details and infrared targets from the original images in the fusion image in terms of subjective effects and objective indicators.

Copper (Cu) foils are used as anode current collectors in secondary batteries and to increase their capacity thin foils 8 μm thick, or less, have been commercialized. The change of mechanical properties according to the thickness of the foils was investigated from the viewpoint of the crystallographic microstructure. The elongation tends to decrease with decreasing thickness of the Cu foils but is limited by an increase in the ratio (D/t) of the grain size (D) to the thickness (t), which can lead to anisotropic plastic deformation. Two electroplating methods are proposed based on the concentration of the additive in the plating solution and by modulating the current density to develop multilayer laminate structures with alternating layers exhibiting different recrystallization behaviors to enhance the mechanical properties of Cu foils. Cycling between two electroplating solutions with different additive concentrations enables the control of the recrystallization behavior, although it likely presents challenges at the production scale. Alternatively, the periodic variation of the current density is used to alter the microstructure and recrystallization behaviors, enabling the formation of locally irregular grain extrusions between the layers. Mechanical measurements reveal a Hall-Petch relationship between the grain size and yield strength in the multilayer laminated foils.

Hypoxia‐inducible factor‐1 (HIF‐1) plays an important role in cellular responses to hypoxia, including the transcriptional activation of several genes involved in tumor angiogenesis. Melatonin, also known as N‐acetyl‐5‐methopxytryptamine, is produced naturally by the pineal gland and has anti‐angiogenic effects in cancer through its ability to modulate HIF‐1α activity. However, the use of melatonin as a therapeutic is limited by its low oral bioavailability and short half‐life. Here, we synthesized melatonin‐like molecules with enhanced HIF‐1α targeting activity and less toxicity and investigated their effects on tumor growth and angiogenesis, as well as the underlying molecular mechanisms. Among melatonin derivatives, N‐butyryl‐5‐methoxytryptamine (NB‐5‐MT) showed the most potent HIF‐1α targeting activity. This molecule was able to (a) reduce the expression of HIF‐1α at the protein level, (b) reduce the transcription of HIF‐1α target genes, (c) reduce reactive oxygen species (ROS) generation, (d) decrease angiogenesis in vitro and in vivo, and (e) suppress tumor size and metastasis. In addition, NB‐5‐MT showed improved anti‐angiogenic activity compared with melatonin due to its enhanced cellular uptake. NB‐5‐MT is thus a promising lead for the future development of anticancer compounds with HIF‐1α targeting activity. Given that HIF‐1α is overexpressed in the majority of human cancers, the melatonin derivative NB‐5‐MT could represent a novel potent therapeutic agent for cancer.

This study was designed to determine whether α-humulene, a major constituent in many plants used in fragrances, has a protective role against gastric injury in vivo and in vitro. A rat model of hydrochloric acid (HCl)/ethanol-induced gastritis and human mast cells (HMC-1) were used to investigate the mucosal protective effect of α-humulene. α-Humulene significantly inhibited gastric lesions in HCl/ethanol-induced acute gastritis and decreased gastric acid secretion pyloric ligation-induced gastric ulcers in vivo. In addition, α-humulene reduced the amount of reactive oxygen species and malondialdehyde through upregulation of prostaglandin E2 (PGE2) and superoxide dismutase (SOD). In HMC-1 cells, α-humulene decreased intracellular calcium and increased intracellular cyclic adenosine monophosphate (cAMP) levels, resulting in low histamine levels. α-Humulene also reduced the expression levels of cytokine genes such as interleukin (IL)-1β, IL-6, tumor necrosis factor (TNF) by downregulating nuclear factor-κB (NF-κB) nuclear translocation. Finally, α-humulene upregulated the expression levels of mucin 5AC (Muc5ac), Muc6, trefoil factor 1 (Tff1), trefoil factor 2 (Tff2), and polymeric immunoglobulin receptor (pigr). α-Humulene may attenuate HCl/ethanol-induced gastritis by inhibiting histamine release and NF-κB activation and stimulating antioxidants and mucosal protective factors, particularly Muc5ac and Muc6. Therefore, these data suggest that α-humulene is a potential drug candidate for the treatment of stress-induced or alcoholic gastritis.

In bottom-up multi-person pose estimation, grouping joint candidates into the appropriately structured corresponding instance of a person is challenging. In this paper, a new bottom-up method, the Partitioned CenterPose (PCP) Network, is proposed to better cluster the detected joints. To achieve this goal, we propose a novel approach called Partition Pose Representation (PPR) which integrates the instance of a person and its body joints based on joint offset. PPR leverages information about the center of the human body and the offsets between that center point and the positions of the body’s joints to encode human poses accurately. To enhance the relationships between body joints, we divide the human body into five parts, and then, we generate a sub-PPR for each part. Based on this PPR, the PCP Network can detect people and their body joints simultaneously, then group all body joints according to joint offset. Moreover, an improved l1 loss is designed to more accurately measure joint offset. Using the COCO keypoints and CrowdPose datasets for testing, it was found that the performance of the proposed method is on par with that of existing state-of-the-art bottom-up methods in terms of accuracy and speed.

Several nonmarine basins were formed in SW Korea under an extensional or transtensional tectonic regime during the Cretaceous. The Neungju Basin is one of these basins, filled by alluvial to lacustrine deposits interbedded with thick, amalgamated tuff beds. Facies analysis in the northern part of the basin shows that the basin fill is composed of four facies associations, representing (1) alluvial fan, (2) alluvial plain, (3) sandflat, and (4) marginal playa lake environments. The basin fill can be subdivided into four stratigraphic units by interbedded thick amalgamated tuff beds. The lowermost unit (unit I) is composed of alluvial plain deposits. The lack of any vertical depositional trend indicates a balance between basin subsidence and sediment supply. The middle unit (units II and III) show regional variations. The extensive lake formed in the central to northern parts of the basin, which are later partly filled by proximal sediments from surrounding areas. In contrast, only alluvial plain to sandflat environments occurred in the southern part. The uppermost unit (unit IV) shows a transition from alluvial fan to sandflat, comprising a fining-upward trend that reflects a gradual decrease in sediment supply and slope-gradient during denudation of uplifted source areas. The variations in depositional patterns were possibly caused by the relationship between volcanism-driven sediment supply and basin subsidence. Basin subsidence may have been sufficient to compensate the volcanism-driven sediment supply prior to the deposition of units II and III, resulting in the conformable deposition of lake sediments over thick amalgamated tuff beds. In contrast, volcanism-driven sediment supply may have reduced the generation of accommodation prior to the deposition of unit IV, resulting in the deposition of conglomerates over thick tuff beds. The traditional depositional model composed of syn-eruption aggradation and inter-eruption degradation may not sufficiently explain the development of sedimentary succession, in which the balance between volcanism-driven sediment supply versus basin subsidence coeval to volcanism is variable.

Canny edge detection principle and morphological operations have been used for depth edge detection. Several image smoothing filters were proposed for the enhancement of the detection task. However, an image smoothing filter can blur the edges of an image. In this paper, we propose an enhancement of depth edge detection using an edge-preserving filter, bilateral filter. The filter smooths an image and reduces the edge blurring effects across the edge such as halos and phantom. We maintain a canny edge detection principle and incorporate it with morphological properties. The results show that, this method can detect edges of a depth image better than the method without the edge-preserving filter such as gaussian and median blur.

Sketch-to-photo recognition is an important challenge in face recognition because it requires matching face images in different domains. Even the deep learning, which has recently been deployed in face recognition, is not efficient for face sketch recognition due to the limited sketch datasets. In this paper, we propose a novel face sketch recognition approach based on transfer learning. We design a three-channel convolutional neural network architecture in which the triplet loss is adopted in order to learn discriminative features and reduce intra-class variations. Moreover, we propose a hard triplet sample selection strategy to augment the number of training samples and avoid slow convergence. With the proposed method, facial features from digital photos and from sketches taken from the same person are closer; the opposite occurs if the digital photo and sketch are from different identities. Experimental results on multiple public datasets indicate that the proposed face sketch recognition method outperforms the existing approaches.

Through silicon via (TSV) is one of the most important technologies in 3-dimensional wafer/chip stacking. However, there are some issues related to defect-free TSV filling and Cu pumping. In this study, a defect-free TSV filling was achieved using iodide ions. The working mechanism of iodide ions in TSV filling was systemically investigated by electrochemical measurements; It was found that the formation of CuI on the electrode surface is a key process for inhibiting Cu ion reduction. This inhibition effect of iodide ions enables defect-free TSV filling. Furthermore, the study of microstructure of the filled Cu revealed that the electrochemical reduction of CuI during the TSV filling forms small Cu grains in TSV. The small Cu grains cause Cu pumping in subsequent processes, which damages semiconductor devices. We achieved a defect-free TSV filling with enlarged Cu grains using a two-step filling method that effectively promotes the direct reduction of Cu ions rather than electrochemical reduction of CuI.

Objective Positive symptoms, such as delusion and hallucination, commonly include negative emotional content in schizophrenia. We investigated the neural basis implicated during the processing of strong negative emotional words in patients with schizophrenia. Methods In our study, 35 patients with schizophrenia and 19 healthy controls were recruited, and the participants were asked to passively view the words that contained swearing and neutral content during functional magnetic resonance imaging. Results Patients with schizophrenia, compared to healthy controls, showed hypoactivation to the swear and neutral words stimuli in the left inferior frontal gyrus, left middle frontal gyrus, and left angular/supramarginal gyrus. More specifically, patients with remitted schizophrenia were found to have greater activation to the stimuli in the left middle/inferior frontal gyrus than patients with active schizophrenia. Furthermore, in the analysis of regions of interests, the left inferior and middle frontal gyrus activity was related to the severity of positive symptoms, including delusion and suspiciousness. Conclusion Our results suggest that patients with schizophrenia have difficulty in semantic processing and inhibitory control of swear words, and these abnormalities may be connected with the severity of positive symptoms.

Face recognition is an important biometric due to its non-intrusive collection of data that can be applied to surveillance systems. However, the human pose is unconstrained under surveillance, and there is only one frontal face available in the gallery in most scenarios, and these two factors challenge face recognition performance. A goodly amount of research has been published to solve this problem. Specially, deep learning-based methods learn generic models between poses to synthesize different poses of a given face, however, generic synthesis models can lose the face identity while warping the face, which deteriorates the discriminative capability of learned features. In this paper, we proposed the deep Siamese networks to learn identity-aware and pose-invariant features, adding contrastive loss to the face synthesis model to preserve the face identity while synthesizing the face. In addition, we trained various face synthesis models with different target poses as supervisory signals, the learned pose-invariant features were incorporated by another Siamese network, resulting in deeper pose-invariant and identity-aware features. The proposed network is free of landmark estimation and face pose, and it is in real time. We tested the proposed algorithm in the FERET and CAS-PEAL datasets, and experimental results demonstrated that our network achieved superior performance to that of recently published algorithms for cross-pose face recognition, especially the 2D deep learning-based algorithms.

Face sketch recognition aims at matching face sketch images to face photo images. The main challenge lies in modality discrepancy between face photo and sketch images. In this work, we propose a new facial sketch-to-photo recognition approach by adopting VGG-Face deep learning network, with which face images can be represented by compact and highly discriminative feature vectors. Different from existing VGG-Face based methods which directly match face sketches to photos, we firstly transform the gallery photos to sketches for decreasing the modality difference. Experimental results on multiple face photo-sketch datasets indicate the superiority of our method.

Kushen (Radix Sophorae flavescentis) is used to treat ulcerative colitis, tumors, and pruritus. Recently, phaseolin, formononetin, matrine, luteolin, and quercetin, through a network pharmacology approach, were tentatively identified as five bioactive constituents responsible for the anti-inflammatory effects of S. flavescentis. However, the role of phaseolin (one of the primary components of S. flavescentis) in the direct regulation of inflammation and inflammatory processes is not well known. In this study, the beneficial role of phaseolin against inflammation was explored in lipopolysaccharide (LPS)-induced inflammation models of RAW 264.7 macrophages and zebrafish larvae. Phaseolin inhibited LPS-mediated production of nitric oxide (NO) and the expression of inducible nitric oxide synthase (iNOS), without affecting cell viability. In addition, phaseolin suppressed pro-inflammatory mediators such as cyclooxygenase 2 (COX-2), interleukin-1β (IL-1β), tumor necrosis factor α (TNF-α), monocyte chemoattractant protein-1 (MCP-1), and interleukin-6 (IL-6) in a dose-dependent manner. Furthermore, phaseolin reduced matrix metalloproteinase (MMP) activity as well as macrophage adhesion in vitro and the recruitment of leukocytes in vivo by downregulating Ninjurin 1 (Ninj1), an adhesion molecule. Finally, phaseolin inhibited the nuclear translocation of nuclear factor-kappa B (NF-κB). In view of the above, our results suggest that phaseolin could be a potential therapeutic candidate for the management of inflammation.

Person re-identification aims to match pedestrian images across multiple surveillance camera views. It is still a challenging task due to the partial occlusion of pedestrian images, variations in the illumination of surveillance cameras, and similar appearances of pedestrians and so on. In order to improve the representation ability of pedestrian features extracted from the convolutional neural networks, in this paper, we proposed an improved split-attention architecture for person re-identification. Specifically, we first divide the feature map into two sub-groups and then split the features in each subgroup into three more fine-grained sub-feature maps. Moreover, in order to minimize the inter-class similarity and maximize the intra-class similarity, we use circle loss and identification loss to optimize our network together. Circle loss makes the similarity scores learn at different paces, which benefits deep feature learning. The circle loss not only makes the model have higher optimization flexibility but also makes the convergence target of the model more definite. Unlike many methods that use complex convolutional neural networks to represent pedestrian feature maps in a layer-wise manner, our proposed method improves the representation ability of pedestrian features at a more fine-grained level. We evaluated the performance of our proposed network on two large-scale person re-identification benchmark datasets Market-1501 and DukeMTMC-reID. Experimental results show that the proposed split-attention network outperforms the state-of-the-art methods on both datasets with only using pedestrian global features.

Semantic segmentation, which refers to pixel-wise classification of an image, is a fundamental topic in computer vision owing to its growing importance in robot vision and autonomous driving industries. It provides rich information about objects in the scene such as object boundary, category, and location. Recent methods for semantic segmentation often employ an encoder-decoder structure using deep convolutional neural networks. The encoder part extracts feature of the image using several filters and pooling operations, whereas the decoder part gradually recovers the low-resolution feature maps of the encoder into a full input resolution feature map for pixel-wise prediction. However, the encoder-decoder variants for semantic segmentation suffer from severe spatial information loss, caused by pooling operations or convolutions with stride, and does not consider the context in the scene. In this paper, we propose a dense upsampling convolution method based on guided filtering to effectively preserve the spatial information of the image in the network. We further propose a novel local context convolution method that not only covers larger-scale objects in the scene but covers them densely for precise object boundary delineation. Theoretical analyses and experimental results on several benchmark datasets verify the effectiveness of our method. Qualitatively, our approach delineates object boundaries at a level of accuracy that is beyond the current excellent methods. Quantitatively, we report a new record of 82.86% and 81.62% of pixel accuracy on ADE20K and Pascal-Context benchmark datasets, respectively. In comparison with the state-of-the-art methods, the proposed method offers promising improvements., 11 pages, published in sensors journal

A large portion of the research for the task of image semantic segmentation focuses on optimizing deep learning networks to improve the state-of-the-art results. In this work, however, we take a different approach in order to enhance the performance of state-of-the-art methods for image semantic segmentation. Benefiting from the internal learning process of deep learning models we propose a Cascade Data Augmentation (CDA) technique that can effectively increases the size of data based on the characteristics of the dataset in hand. CDA utilizes basic image transformation techniques such as geometric and photometric transformations to augment the size of data. In particular, CDA blends these transformations by assigning two adjustable parameters that are 1) the probability of applying the transformation operation and 2) the magnitude of the operation. We evaluate the effectiveness of our method on the validation set of Pascal Context benchmark dataset for image semantic segmentation. Experimental results prove that incorporating CDA with baseline deep learning models can increase their performance. In particular we attain a average gain of ≈ 4.72 pixel accuracy and ≈ 1.9 mean intersection over union Pascal-Context dataset. Moreover, CDA is a generic method that can be integrated with any deep learning model and attains improved performance while adding trivial extra training time of the network.

The human pose estimation is a problem of localizing human joints in a single image, and that is still a challenge in the field of computer vision. The hourglass network has been used in many researches to achieve good performance in human pose estimation problems. For human pose estimation problem, not only high-level features but also low-level features are important for understanding the whole human body. However, the vanilla hourglass network has the problem of passing only high-level features to the next stack. Therefore, we propose a network structure that can solve the problems of the vanilla hourglass by using an additional skip connection. The proposed skip connection improves network performance by passing relative low-level features to the next stack. In addition, the skip connection is a simple element-wise Sum operation, so there is no increase in the number of parameters. In this work, we use the well-known human pose estimation data set, MPII, to evaluate the proposed method. We conducted experiments to evaluate the objective performance of the proposed method, and it was confirmed through this evaluation that the proposed method improves the performance of human pose estimation of the vanilla hourglass network.

Slow-cycling cancer cells (SCCs) with a quiescence-like phenotype are believed to perpetrate cancer relapse and progression. However, the mechanisms that mediate SCC-derived tumor recurrence are poorly understood. Here, we investigated the mechanisms underlying cancer recurrence after chemotherapy, focusing on the interplay between SCCs and the tumor microenvironment. We established a preclinical model of SCCs by exposing non-small-cell lung cancer (NSCLC) cells to either the proliferation-dependent dye carboxyfluorescein diacetate succinimidyl ester (CFSE) or chemotherapeutic drugs. An RNA sequencing analysis revealed that the established SCCs exhibited the upregulation of a group of genes, especially epidermal growth factor (EGF). Increases in the number of vascular endothelial growth factor receptor (VEGFR)-positive vascular endothelial cells and epidermal growth factor receptor (EGFR) activation were found in NSCLC cell line- and patient-derived xenograft tumors that progressed upon chemotherapy. EGFR tyrosine kinase inhibitors effectively suppressed the migration and tube formation of vascular endothelial cells. Furthermore, activating transcription factor 6 (ATF6) induced the upregulation of EGF, and its antagonism effectively suppressed these SCC-mediated events and inhibited tumor recurrence after chemotherapy. These results suggest that the ATF6-EGF signaling axis in SCCs functions to trigger the angiogenesis switch in residual tumors after chemotherapy and is thus a driving force for the switch from SCCs to actively cycling cancer cells, leading to tumor recurrence.

Objective: We propose and develop a novel biclustering (N-BiC) approach for performing N-way biclustering of neuroimaging data. Our approach is applicable to an arbitrary number of features from both imaging and behavioral data (e.g., symptoms). We applied it to structural MRI data from patients with schizophrenia. Methods: It uses a source-based morphometry approach [i.e., independent component analysis of gray matter segmentation maps] to decompose the data into a set of spatial maps, each of which includes regions that covary among individuals. Then, the loading parameters for components of interest are entered to an exhaustive search, which incorporates a modified depth-first search technique to carry out the biclustering, with the goal of obtaining submatrices where the selected rows (individuals) show homogeneity in their expressions of selected columns (components) and vice versa. Results: Findings demonstrate that multiple biclusters have an evident association with distinct brain networks for the different types of symptoms in schizophrenia. The study identifies two components: inferior temporal gyrus (16) and brainstem (7), which are related to positive (distortion/excess of normal function) and negative (diminution/loss of normal function) symptoms in schizophrenia, respectively. Conclusion: N-BiC is a data-driven method of biclustering MRI data that can exhaustively explore relationships/substructures from a dataset without any prior information with a higher degree of robustness than earlier biclustering applications. Significance: The use of such approaches is important to investigate the underlying biological substrates of mental illness by grouping patients into homogeneous subjects, as the schizophrenia diagnosis is known to be relatively nonspecific and heterogeneous.

Cristacarpin is a novel prenylated pterocarpan that reportedly exhibits broad anti-cancer activity by enhancing endoplasmic reticulum stress. However, whether and how cristacarpin affects in-flammatory processes remain largely unknown. In the present study, the anti-inflammatory effect of cristacarpin on lipopolysaccharide (LPS)-induced inflammation was investigated using zebrafish embryos, RAW 264.7 macrophages, and mouse uveitis models. In the non-toxic concentration range (from 20 to 100 μM), cristacarpin suppressed pro-inflammatory mediators such as interleukin (IL)-6 and tumor necrosis factor (TNF)-α, while stimulating anti-inflammatory mediators such as IL-4 and IL-10 in LPS-stimulated RAW 264.7 cells and uveitis mouse models. Cristacarpin decreased cell adhesion of macrophages through downregulation of the expression of Ninjurin1 and matrix metalloproteinases. Furthermore, cristacarpin reduced macrophage migration in zebrafish embryos in vivo. Cristacarpin also increased cytosolic levels of inhibitor of nuclear factor-κB and suppressed the nuclear translocation of nuclear factor κ-light-chain-enhancer of activated B cells. Collectively, our results suggest that cristacarpin is a potential therapeutic candidate for developing ocular anti-inflammatory drugs.

Hallucinations characterize schizophrenia, with approximately 59% of patients reporting auditory hallucinations and 27% reporting visual hallucinations. Prior neuroimaging studies suggest that hallucinations are linked to disrupted communication across distributed (sensory, salience-monitoring and subcortical) networks. Yet, our understanding of the neurophysiological mechanisms that underlie auditory and visual hallucinations in schizophrenia remains limited. This study integrates two resting-state functional magnetic resonance imaging (fMRI) analysis methods – amplitudes of low-frequency fluctuations (ALFF) and functional network connectivity (FNC) – to explore the hypotheses that (1) abnormal FNC between salience and sensory (visual/auditory) networks underlies hallucinations in schizophrenia, and (2) disrupted hippocampal oscillations (as measured by hippocampal ALFF) beget changes in FNC linked to hallucinations. Our first hypothesis was supported by the finding that schizophrenia patients reporting hallucinations have higher FNC between the salience network and an associative auditory network relative to healthy controls. Hippocampal ALFF was negatively associated with FNC between primary auditory cortex and the salience network in healthy subjects, but was positively associated with FNC between these networks in patients reporting hallucinations. These findings provide indirect support favoring our second hypothesis. We suggest future studies integrate fMRI with electroencephalogram (EEG) and/or magnetoencephalogram (MEG) methods to directly probe the temporal relation between altered hippocampal oscillations and changes in cross-network functional communication.

Objective Auditory hallucinations (AHs) are a core symptom of schizophrenia. We investigated the neural signature of AHs by comparing hallucinating patients with schizophrenia with non-hallucinating patients with schizophrenia. Methods We recruited hallucinating patients with schizophrenia meeting the criteria for persistent, prominent, and predominant AHs (n=10) and non-hallucinating patients with schizophrenia (n=12). Various clinical assessments were performed incluing Psychotic Symptom Rating Scale for Auditory Hallucinations. Using fludeoxyglucose (18F) positron emission tomography, regional differences in neural activity between the groups were analyzed. Results The regions of interest analysis showed significantly lower standardized uptake value ratio (SUVR) in the superior, middle, and inferior frontal gyri, and higher SUVR in the putamen in patients with AHs versus patients without AHs. These findings were confirmed in the voxel-wise analysis. Conclusion Our findings indicate that hypoactivity in the frontal and cingulate gyri, coupled with hyperactivity in the temporal gyrus and putamen, may contribute to the pathophysiology of AHs.

Fluorine is a unique atom that imparts distinct properties to bioactive molecules upon incorporation. Herein, we prepare and study fluorinated derivatives of the nanomolar affine peripherally restricted dual CB1R/CB2R agonist; CRA13 and its analogs. Binding affinity evaluation relative to CRA13 proved the stronger binding affinity of compound 7c to CB1R and CB2R by 6.95 and 5.64 folds. Physicochemical properties evaluation proved compound 7c improved lipophilicity profile suggesting some enhanced BBB penetration relative to CRA13. Radiosynthesis of 18F-labeled compound 7c was conducted conveniently affording pure hot ligand. In vivo PET study investigation demonstrated efficient distribution of 18F-labeled compound 7c in peripheral tissues visualizing peripheral CB1R/CB2R generating time-activity-curves showing good standard uptake values. Despite enhanced BBB penetration and increased cannabinoid receptors binding affinity, low brain uptake of 7c was observed. In silico docking study explained the measured binding affinities of compounds 7a-d to CB1R. While most of previous efforts aimed to develop central cannabinoid PET imaging agents, 18F-labeled compound 7c might be a promising agent serving as a universal CB1R/CB2R PET imaging agents for diagnosis and therapy of various diseases correlated with peripheral cannabinoid system. It might also serve as a lead compound for development of PET imaging of peripheral and central cannabinoid systems.

Person re-identification is widely regarded as an image retrieval problem. Given a pedestrian-of-interest (query) image in one camera, the person re-identification system aims to identify the pictures of the same person from an image pool (gallery). Due to the differences in camera pixels, pose, illumination, occlusion, and intra-class variations across different cameras, the task of person re-identification remained challenging to the community of computer vision scientists. In this paper, we propose a multi-task network, based on a uniform partition network, which computes the identification loss and verification loss of two input images simultaneously. Given a pair of images as input, the system predicts the identities of the two input images and outputs a similarity score at the same time, to indicate whether they belong to the same identity or not. To get more fine-grained part-level features, we adopted the part-based convolutional baseline network for feature extraction of each input image and output a convolutional descriptor consisting of six local features. Our model achieved 81.19% mAP and 93.34% rank-1 accuracy on Market-1501 datasets. It also achieved 72.12% mAP and 85.59% rank-1 accuracy on DukeMTMC-reID. Comparing them with those of state-of-the-art, our model outperformed the state-of-the-art by a margin of 3.79% mAP, 1.03% rank-1, and 6.02% mAP, 3.79% rank-1 on Market-1501 and DukeMTMC-reID, respectively.

Due to its tentacle poison and huge body, giant jellyfish (Nemopilema nomurai) poses challenging issues to the environment and ecosystems. Here we developed, upcycling a giant jellyfish extract as a reducing agent, a green synthetic method of gold nanoparticles (JF-AuNPs) which possess biological activities. The colloidal solutions of JF-AuNPs were blue, violet, purple and pink depending on the extract concentration. UV-visible spectra exhibited two surface plasmon resonance bands at 5 4 0 ∼ 550 nm and 810 nm. Spherical shapes with an average size of 35.2 ± 8.7 nm and triangular nanoplates with an average height of 70.5 ± 30.3 nm were observed. A face-centered cubic structure was confirmed by high-resolution X-ray diffraction. JF-AuNPs exhibited significant cytotoxic effect against HeLa cancer cells but not against normal cells such as NIH-3T3 and Raw 264.7 cells. In HeLa cells, JF-AuNPs decreased the phosphorylation of AKT and ERK, which are crucial for cell proliferation. Also, JF-AuNPs decreased NO secretion and iNOS expression levels, resulting in anti-inflammatory effects in LPS-inflamed macrophages. Collectively, we established a green synthesis of anti-tumorigenic and anti-inflammatory JF-AuNPs using the extract of jellyfish sea wastes. Thus, beneficial effects of JF-AgNPs must be weighed in further studies in vivo and it can be potent nanomedicine for future applications.

Spermidine is a naturally occurring polyamine compound that has recently emerged with anti-aging properties and suppresses inflammation and oxidation. However, its mechanisms of action on anti-inflammatory and antioxidant effects have not been fully elucidated. In this study, the potential of spermidine for reducing pro-inflammatory and oxidative effects in lipopolysaccharide (LPS)-stimulated macrophages and zebrafish was explored. Our data indicate that spermidine significantly inhibited the production of pro-inflammatory mediators such as nitric oxide (NO) and prostaglandin E2 (PGE2), and cytokines including tumor necrosis factor-α and interleukin-1β in RAW 264.7 macrophages without any significant cytotoxicity. The protective effects of spermidine accompanied by a marked suppression in their regulatory gene expression at the transcription levels. Spermidine also attenuated the nuclear translocation of NF-κB p65 subunit and reduced LPS-induced intracellular accumulation of reactive oxygen species (ROS) in RAW 264.7 macrophages. Moreover, spermidine prevented the LPS-induced NO production and ROS accumulation in zebrafish larvae and was found to be associated with a diminished recruitment of neutrophils and macrophages. Although more work is needed to fully understand the critical role of spermidine on the inhibition of inflammation-associated migration of immune cells, our findings clearly demonstrate that spermidine may be a potential therapeutic intervention for the treatment of inflammatory and oxidative disorders.

This work examines the filling of Through Silicon Vias (TSV) by Ni deposition from a NiSO4 + NiCl2 + H3BO3 electrolyte containing a branched polyethyleneimine suppressor. Feature filling occurs due to the interaction of transport limited suppressor adsorption and its consumption by potential dependent metal deposition. The interaction between surface topography and suppressor transport yields a sharp transition from passive to active deposition within the TSV. The transition is associated with significant incorporation of the suppressor, or its components, within the Ni deposit that results in grain refinement evident by electron backscatter diffraction (EBSD). Potential waveforms that progressively shift the location of the passive-active transition upward to optimize feature filling were examined. The evolution of feature filling and deposit microstructure are compared to predictions of a three-dimensional model that reflect critical behavior associated with suppressor-derived, S-shaped negative differential resistance (S-NDR). The model uses adsorption and consumption kinetics obtained from voltammetric measurements of the critical potential associated with suppression breakdown. Good agreement between experiment and simulation is demonstrated.

Person re-identification (re-ID) has been gaining in popularity in the research community owing to its numerous applications and growing importance in the surveillance industry. Recent methods often employ partial features for person re-ID and offer fine-grained information beneficial for person retrieval. In this paper, we focus on learning improved partial discriminative features using a deep convolutional neural architecture, which includes a pyramid spatial pooling module for efficient person feature representation. Furthermore, we propose a multi-task convolutional network that learns both personal attributes and identities in an end-to-end framework. Our approach incorporates partial features and global features for identity and attribute prediction, respectively. Experiments on several large-scale person re-ID benchmark data sets demonstrate the accuracy of our approach. For example, we report rank-1 accuracies of 85.37% (+3.47 %) and 92.81% (+0.51 %) on the DukeMTMC re-ID and Market-1501 data sets, respectively. The proposed method shows encouraging improvements compared with the state-of-the-art methods.

Mammalian claudin-5 (cldn5), a zebrafish cldn5a homolog, is essential to blood-brain barrier (BBB) integrity. Previously, the existence of an endothelial tight junction-based BBB with cldn5a expression in the cerebral microvessels was reported in zebrafish. However, the role of cldn5a in the cerebral microvessels of developing zebrafish has not been elucidated. Here, we further investigated the functional integrity of cldn5a in developing zebrafish by injecting cldn5a morpholinos. At 7 days post-fertilization, cldn5a immunoreactivity was detected on the brain surface, ventricular ependyma, and cerebral mircovessels but disappeared following cldna5a knockdown. Cldn5a morphants showed size-selective leakage of tracers through the BBB and downregulated expression of glucose transporter 1 (glut1) in the cerebral microvessels. In addition, leakiness in the blood-cerebrospinal fluid barrier was observed, implying the overall abnormal development of blood-neural barriers. The results of our study suggest that cldn5a is required for building and maintaining the blood-neural barrier during zebrafish development.

The microstructure of copper filled through silicon vias deposited in a CuSO(4) + H(2)SO(4) electrolyte containing micromolar concentrations of deposition rate suppressing poloxamine and chloride additives is explored using electron backscatter diffraction. Regions with distinct microstructures are observed in the vias, including conformal deposition and seam formation localized adjacent to the bottom that can transition to bottom-up filling higher in the features. The presence and extent of each microstructure depends on applied potential as well as additive concentration. Deposition in the presence of higher chloride concentration yields a strong (110) growth texture in regions where bottom-up filling exhibits a horizontal growth front profile while (110) textured or untextured growth is observed for different conditions where upward growth proceeds with a v-notch profile.