Search

Polymeric elastomers are extensively employed to fabricate implantable medical devices. However, implantation of the elastomers can induce a strong immune rejection known as the foreign body response (FBR), diminishing their efficacy. Herein, we present a group of immunocompatible elastomers, termed easy-to-synthesize vinyl-based anti-FBR dense elastomers (EVADE). EVADE materials effectively suppress the inflammation and capsule formation in subcutaneous models of rodents and non-human primates for at least one year and two months, respectively. Implantation of EVADE materials significantly reduces the expression of inflammation-related proteins S100A8/A9 in adjacent tissues compared to polydimethylsiloxane. We also show that inhibition or knockout of S100A8/A9 leads to substantial attenuation of fibrosis in mice, suggesting a target for fibrosis inhibition. Continuous subcutaneous insulin infusion (CSII) catheters constructed from EVADE elastomers demonstrate significantly improved longevity and performance compared to commercial catheters. The EVADE materials reported here may enhance and extend function in various medical devices by resisting the local immune responses. Fibrous capsule formation limits the lifetime of implanted devices. Here, the authors report on an elastomer designed to reduce the immune response which shows enhanced device utility in both mice and non-human primate models, including in an insulin infusion catheter model. [ABSTRACT FROM AUTHOR]

The hot-compression bonding of Ti386 and TC4 titanium alloys takes place within a temperature range of 800 to 900 °C and under strains of 10–30% in this study. The analysis of Ti386/TC4 joints involves the interfacial microstructure and compression shear strength are conducted. Quantitative analysis reveals that the interfacial voids, as a consequence of surface roughness, notably decrease as deformation strains and temperatures increase, which leads to the highest shear strength 673.26 MPa at 900 °C/30%. The Ti386/TC4 interface exhibits distinct characteristics and presents two zones: αp(TC4)/β(Ti386) and αs + β(TC4)/β(Ti386). The αp(TC4) phase protrudes into the β(Ti386) side at the αp(TC4)/β(Ti386) interface due to the higher hardness of the stable primary αp(TC4). During the process of deformation bonding, dynamic recrystallization occurs within the interfacial region. On the TC4 side, dynamic recrystallization is primarily discontinuous dynamic recrystallization, initially observed in the lamellar αs + β areas before extending to the equiaxed αp areas. Conversely, on the Ti386 side, continuous dynamic recrystallization prevails, characterized by the transformation of low-angle grain boundaries to high-angle grain boundaries. [ABSTRACT FROM AUTHOR]

Domain adaptation (DA) aligns domain invariant features between different domains as much as possible, but domain-specific features are largely ignored. In this paper, our aim is to utilize specific information on the target domain to improve domain adaptation. To address this issue, a natural idea is to use another classifier to model-specific information independently, because it is difficult to combine the different features of the two domains in a single model. Specifically, we propose a two-branch network (TBN), which utilizes ground truth labels to train the source domain classifier and utilizes pseudo-labels to train the target domain classifier. Different classifiers were able to learn different domain-specific features. Considering the noise impact of pseudo-labels, we propose one-time clustering module to further boost the accuracy of pseudo-labels. In particular, TBN can be easily integrated into various DA methods to further improve their performance. The superiority of the proposed method is validated on several standard datasets. [ABSTRACT FROM AUTHOR]

Remote sensing imagery is indispensable in diverse domains, including geographic information systems, climate monitoring, agricultural planning, and disaster management. Nonetheless, cloud cover can drastically degrade the utility and quality of these images. Current deep learning-based cloud removal methods rely on convolutional neural networks to extract features at the same scale, which can overlook detailed and global information, resulting in suboptimal cloud removal performance. To overcome these challenges, we develop a method for cloud removal that leverages multi-scale spatial information perception. Our technique employs convolution kernels of various sizes, enabling the integration of both global semantic information and local detail information. An attention mechanism enhances this process by targeting key areas within the images, and dynamically adjusting channel weights to improve feature reconstruction. We compared our method with current popular cloud removal methods across three datasets, and the results show that our proposed method improves metrics such as PSNR, SSIM, and cosine similarity, verifying the effectiveness of our method in cloud removal. [ABSTRACT FROM AUTHOR]

Few-shot learning has achieved satisfactory progress over the years, but these methods implicitly hypothesize that the data in the base (source) classes and novel (target) classes are sampled from the same data distribution (domain), which is often invalid in reality. The purpose of cross-domain few-shot learning (CD-FSL) is to successfully identify novel target classes with a small quantity of labeled instances on the target domain under the circumstance of domain shift between the source domain and the target domain. However, in CD-FSL, the knowledge learned by the network on the source domain often suffers from the situation of inadaptation when it is transferred to the target domain, since the instances on the source and target domains do not obey the same data distribution. To surmount this problem, we propose a Target Oriented Dynamic Adaption (TODA) model, which uses a tiny amount of target data to orient the network to dynamically adjust and adapt during training. Specifically, this work proposes a domain-specific adapter to ameliorate the network inadaptability issues in transfer to the target domain. The domain-specific adapter can make the extracted features more specific to the tasks in the target domain and reduce the impact of tasks in the source domain by combining them with the mainstream backbone network. In addition, we propose an adaptive optimization method in the network optimization process, which assigns different weights according to the importance of different optimization tasks. Extensive experiments on several benchmark datasets demonstrate the effectiveness of our TODA method. [ABSTRACT FROM AUTHOR]

Background: Soft tissue recurrence of giant cell tumor of bone (GCTB) is rare. This study aims to provide its prevalence, recurrent locations, risk factors, effective detection methods and a modified classification for this recurrence. Methods: Patients with soft tissue recurrence after primary surgery for GCTB were screened from January 2003 to December 2022. General data, recurrence frequency, types according to an original classification (type-I: peripheral ossification; type-II: central ossification; type-III: without ossification), a modified classification with more detailed subtypes (type I-1: ≤ 1/2 peripheral ossification; type I-2: ≥ 1/2 peripheral ossification; type II-1: ≤ 1/2 central ossification; type II-2: ≥ 1/2 central ossification; type III: without ossification), locations, detection methods such as ultrasonography, X-ray, CT or MRI, Musculoskeletal Tumor Society (MSTS) scores were recorded. Multivariate regression analysis was conducted to identify risk factors for recurrence frequency. Results: A total of 558 recurrent cases were identified from 2009 patients with GCTB. Among them, 32 were soft tissue recurrence. The total recurrence rate was 27.78% (558/2009). Soft tissue recurrence rate was 5.73% among 558 recurrent cases, and 1.59% among 2009 GCTB patients, respectively. After excluding one patient lost to follow-up, 10 males and 21 females with the mean age of 28.52 ± 9.93 (16–57) years were included. The definitive diagnosis of all recurrences was confirmed by postoperative pathology. The interval from primary surgery to the first recurrence was 23.23 ± 26.12 (2–27) months. Eight recurrences occurred from primary GCTB located at distal radius, followed by distal femur (6 cases). Recurrence occurred twice in 12 patients and 3 times in 7 patients. Twenty-seven recurrences were firstly detected by ultrasonography, followed by CT or X-ray (10 cases in each). Types at the first recurrence were 5 cases in type-I, 8 in type-II and 18 in type-III. According to the modified classification, 3 patients in type I-1, 2 in type I-2, 1 in type II-1, 7 in type II-2, and 18 in type III. The mean MSTS score was 26.62 ± 4.21 (14—30). Neither Campanacci grade nor recurrence type, modified classification and other characters, were identified as risk factors. Conclusions: Soft tissue recurrence of GCTB may recur for more than once and distal radius was the most common location of primary GCTB that would suffer a soft tissue recurrence. Ultrasonography was a useful method to detect the recurrence. Since no risk factors were discovered, a careful follow-up with ultrasonography was recommended. [ABSTRACT FROM AUTHOR]

Graph convolutional network (GCN) has led to state-of-the-art performance for structured data. The superior performance would be partly due to the convolutional operations that operate over local neighborhoods. However, the distant long-range dependencies in data are still challenging to capture since it requires deep stacks of convolutional operations. Moreover, missing links in structured data might further hurt the performance. This paper introduces non-locality augmented graph convolution blocks into GCN to capture long-range or even disconnected dependencies. Specifically, we propose a dictionary-based non-locality encoding approach in which the non-local information is encoded by both graph convolution and dictionary-based implicit convolution. Unlike previous non-local approaches, our non-local block does not rely on the exhaustive computation of the relationship of data pairs. Thus, it is suitable for GCN, which typically models a large number of data samples. What's more, the proposed non-local blocks could be embedded into arbitrarily GCN architectures. We demonstrate the efficacy of our non-local block on four benchmark datasets. [ABSTRACT FROM AUTHOR]

In recent decades, the development of multimedia and computer vision has sparked significant interest among researchers in the field of automatic image annotation. However, much of the research has primarily focused on using a single graph for annotating images in semi-supervised learning. Conversely, numerous approaches have explored the integration of multi-view or image segmentation techniques to create multiple graph structures. Yet, relying solely on a single graph proves to be challenging, as it struggles to capture the complete manifold of structural information. Furthermore, the computational complexity of building multiple graph structures based on multi-view or image segmentation is substantial and time-consuming. To address these issues, we propose a novel method called "Central Attention with Multi-graphs for Image Annotation." Our approach emphasizes the critical role of the central image region in the annotation process. Remarkably, we demonstrate that impressive performance can be achieved by leveraging just two graph structures, composed of central and overall features, in semi-supervised learning. To validate the effectiveness of our proposed method, we conducted a series of experiments on benchmark datasets, including Corel5K, ESPGame, and IAPRTC12. These experiments provide empirical evidence of our method’s capabilities. [ABSTRACT FROM AUTHOR]

Ciprofloxacin (CIP, C17H18FN3O3), one of the main pollution sources detected in water, has raised intense issues. In order to resolve this issue, it is essential to achieve efficient degradation of CIP under visible light. Herein, a set of two-dimensional composites MoS2/Bi2MoO6 were constructed, and proved via XRD, FT-IR, SEM, TEM, and XPS, respectively. In addition, the results recorded that when the optimized composites 0.5%-MoS2/Bi2MoO6 were applied, the degradation rate of CIP was up to 91% at 120 min under the irradiation of visible light (> 395 nm) which was superior to that of the pure MoS2 or Bi2MoO6. UV–Vis DRS, photocurrent, and electrical impedance analyses indicate that the outperformance of 0.5%-MoS2/Bi2MoO6 is due to the optimized band gap and specific surface area. More importantly, a photocatalytic mechanism was proposed based on the analysis of GC–MS (according to the references (Dang et al. in Chem Eng J 422(15):130103, 2021; Zhu et al. in Environ Sci Pollut Res Int 30(11):28874–28888, 2022)) and active species. The result was manifest: CIP was degraded progressively under the attack of photocatalytically generated h+,.O2−. This work opens up a category for the construction of efficient Bi2MoO6-based composites and proposes a hypothetical mechanism and pathway for CIP degradation. [ABSTRACT FROM AUTHOR]

In today's society, breast cancer is becoming more and more serious, and the promotion of related medical means is imminent. They require modern technology support led by artificial intelligence, such as computer-aided diagnosis system for breast cancer. The existing computer-aided diagnostic systems of breast cancer have high cost, high requirements for the configuration of hospital hardware facilities, and are basically limited to the classification of benign and malignant. The paper proposes L_MiniVGGNet (improved VGGNet) and T_MobileNetV2 (improved MobileNetV2) algorithms, and a simple integrated breast cancer pathologic image pre-classification system. The system uses L_MiniVGGNet and T_MobileNetV2 as the core algorithms to solve the problem of binary and quaternary classification of pathological images, respectively. It adapts to both tensorflow and pytorch frameworks. In this paper, BreakHis and BACH open data sets are used, and data enhancement and transfer learning algorithms are combined to train high-precision and low-memory models and save them. PyQt5 library is used to build human–computer interaction GUI system interface, and images are imported according to the path required by system users and diagnosis results are predicted. [ABSTRACT FROM AUTHOR]

The thyroid gland is a critical regulator of numerous physiological functions, and the presence of thyroid nodules often signals potential disorders. Accurate nodule segmentation from ultrasound images is imperative for effective diagnosis and treatment planning. Existing techniques often struggle due to intra-nodule variability. To address this, we introduce TNSeg, an innovative framework specifically designed for thyroid nodule segmentation. TNSeg incorporates two key components: a segmentation block and a discriminative block, and leverages adversarial training. In particular, the discriminator uses a fully convolutional decoder with skip connections to efficiently differentiate between real and synthetic samples. Further, we introduce a novel multi-scale joint loss function for adversarial training that employs a balanced sampling strategy, effectively resolving the difficulties associated with foreground-background differentiation and computational redundancy. Extensive evaluation proves TNSeg's superiority in achieving a Dice coefficient of 92.06%, Hd95 of 13.35, Jaccard index of 90.02%, and Precision of 94.01%, thereby demonstrating significant improvements in four commonly used segmentation quality metrics. [ABSTRACT FROM AUTHOR]

To evaluate whether single acetabular column can be reserved and the effect of reconstruction with femoral head plus total hip replacement (THR) for primary malignant peri-acetabulum tumors. From 2007 to 2015, nineteen patients with primary malignant peri-acetabulum tumors were enrolled. All cases underwent single column resection with clear surgical margins. Ten of the 19 tumor's resections were assisted by computer navigation. Femoral heads were applied to reconstruct anterior or posterior column defects; THR was used for joint reconstruction. The surgical safety, oncologic outcome and prosthesis survivorship and function were evaluated by regular follow-up. The average follow-up period was 65.9 months. Surgical margins contained wide resection in 12 cases and marginal resection in 7 cases. One patient with Ewing's sarcoma died 14 months postoperative due to lung metastasis. One case with chondrosarcoma had recurrence. One prosthesis was removed due to infection. The average MusculoSkeletal Tumor Society (MSTS) function score was 83.7%. Due to the relative small number of cases, there was no significant difference in the recurrence rate and prosthesis failure rate between the navigation group and non-navigation group. Single column resection and reconstruction with femoral head autograft plus THR is an effective, safe method with less complication rate and better functional outcome for patients with peri-acetabular tumors. [ABSTRACT FROM AUTHOR]

The study of few-shot remote sensing image classification has received significant attention. Although meta-learning-based algorithms have been the primary focus of recent examination, feature fusion methods stress feature extraction and representation. Nonetheless, current feature fusion methods, like the multi-head mechanism, are restricted by their complicated network structure and challenging training process. This manuscript presents a simplified multi-head mechanism for obtaining multiple feature representations from a single sample. Furthermore, we perform specific fundamental transformations on remote-sensing images to obtain more suitable features for information representation. Specifically, we reduce multiple feature extractors of the multi-head mechanism to a single one and add an image transformation module before the feature extractor. After transforming the image, the features are extracted resulting in multiple features for each sample. The feature fusion stage is integrated with the classification prediction stage, and multiple linear classifiers are combined for multi-decision fusion to complete feature fusion and classification. By combining image transformation with feature decision fusion, we compare our results with other methods through validation tests and demonstrate that our algorithm simplifies the multi-head mechanism while maintaining or improving classification performance. [ABSTRACT FROM AUTHOR]

History matching, as an essential part of reservoir development, aims to infer high-dimensional geological parameters of a reservoir with a small amount of observation. Despite the rapid development of optimization algorithms, finding optimal solutions for history matching is still challenging because of the large number of parameters that depend on the grid blocks of the numerical simulation model. Motivated by the divide-and-conquer strategy, in this work a novel probabilistic cooperative coevolutionary framework based on local parameterization (LP-PCC) is constructed to improve the convergence of the history matching of large-scale problems. First, the high-dimensional model parameters are decomposed based on smooth local parameterization, in which the divided low-dimensional parameters can reconstruct smooth boundaries of the geological structure during optimization. After that, a contribution-based cooperative coevolutionary algorithm is adopted to optimize the low-dimensional parameters in a round-robin fashion and allocate the computational resources reasonably. To further improve the performance of cooperative coevolution, a new probabilistic method integrated with contribution information is presented to select the subcomponents to be optimized. This framework incorporates domain knowledge for decomposition and a probabilistic mechanism to select subcomponents with large probability, which enhances both convergence and exploration in cooperative coevolution. Two synthetic reservoir cases are designed to validate the effectiveness and efficiency of the proposed method. The numerical results indicate that, compared with traditional strategies, the method can obtain better history-matching results and be adapted to large-scale reservoir problems. [ABSTRACT FROM AUTHOR]

Domain adaptation (DA) improves the generalization ability of models across source and target domains with different distributions. Current methods aim to reduce domain distribution divergence to learn transferable features. However, in most real cases, the number of available labeled samples is limited, and annotating labels requires significant time and labor costs, making it difficult to achieve high accuracy. To address this problem, we propose a novel method called Deep Geometric and Statistic Alignment for Fewer labeled Domain Adaptation (GSA4FDA). This method achieves the target task with fewer labeled source samples by combining manifold learning and leveraging the local geometric structure of sufficient unlabeled source samples. For domain alignment, we employ a joint geometric-statistical alignment and embed it into a specific layer of a deep convolutional neural network (CNN) to obtain high-level semantic information, taking into account the complementary nature of the two aspects. Specifically, we use the Nyström method and Maximum Mean Discrepancy (MMD) to compensate for the geometrical and statistical shift between domains. The experimental results on several datasets demonstrate the superiority of our method, particularly when there are limited labeled source samples available. [ABSTRACT FROM AUTHOR]

Binary transition metal sulfides have garnered widespread concentration result from their superior electrical conductivity and outstanding capacitance. However, their poor cycling stability hinders their applications in energy storage devices. The objective of this study is to devise and prepare graphene and NiCo2S4 composite (NiCo2S4@graphene) using a simple one-step hydrothermal modality. Graphene is used as a conductive substrate, and NiCo2S4 nanoparticles are formed in situ and homogeneously anchored on graphene nanosheets through C-S-C covalent bonds. For example, the NiCo2S4@graphene composite has a high specific capacitance of 918.0 C g−1 at a current density of 1 A g−1 and enhanced cycling stability (90.1% after 6000 cycles). In addition, the asymmetric supercapacitor was fabricated with NiCo2S4@graphene as the positive electrode and graphene (GR) as the negative electrode, and the device provided a maximum energy density of 49.8 Wh kg−1 at a power density of 845.3 Wh kg−1. Besides, the capacitance retention rate was as high as 92.0% after 1000 cycles. The superior electrochemical properties of the NiCo2S4@graphene material verified its huge potential for realistic applications. [ABSTRACT FROM AUTHOR]

In recent years, deep learning-based methods for remote sensing image dehazing have achieved remarkable progress, especially with the incorporation of attention mechanisms to enhance feature learning. However, the channel attention method uses global average pooling to aggregate feature information, which makes it sensitive to specific points of influence. In particular, remote sensing images are captured in a wide range, and are more likely to have special points such as abrupt color changes and confusing landscapes within the images. This paper proposes a dynamic feature attention network (DFA-Net) that includes a dynamic channel module and a pixel attention module to address this challenge. The dynamic channel attention module adds a network branche to learn the weight of spatial information and dynamically aggregates features to reduce the influence from outliers such as color mutations and improve the defogging effect. The pixel attention module is also added to learn the importance of different pixels and enhance the feature extraction capability. Experiments on several datasets have shown that DFA-Net is suitable for remote sensing image dehazing in various scenes, and has achieved the best dehazing performance. [ABSTRACT FROM AUTHOR]

Existing low-light image enhancement techniques face challenges in achieving high visual quality and computational efficiency, as well as in effectively removing noise and adjusting illumination in extremely dark scenes. To address these problems, in this paper, we propose an illumination enhancement network based on Retinex theory for fast and accurate brightening of images in low-illumination scenes. Two learning-based networks are carefully constructed: decomposition network and enhancement network. The decomposition network is responsible for decomposing the low-light input image into the initial reflectance and illumination map. The enhanced network includes two sub-modules: the illumination enhancement module and the reflection denoising module, which are used for efficient brightness enhancement and accurate reflectance. Specially, we have established a cascaded iterative lighting learning process and utilized weight sharing to conduct accurate illumination estimation. Additionally, unsupervised training losses are defined to improve the generalization ability of the model. The proposed illumination enhancement framework enables noise suppression and detail preservation of the final decomposition results. To establish the efficacy and superiority of the model, on the widely applicable LOL dataset, our approach achieves a significant 9.16% increase in PSNR compared to the classical Retinex-Net, and a remarkable enhancement of 19.26% compared to the latest SCI method. [ABSTRACT FROM AUTHOR]

Graph neural networks' application in automatic image annotation is becoming more mature. However, there are still several problems. First, the feature data of the original image obtained by the feature extraction algorithm, such as color features and gradient features, all have the problem of slight intra-class variance and significant inter-class variance. Second, merely utilize the graph convolution neural networks to construct samples or labeled graphs, limiting multimodality's fusion and expansion. This paper uses a parallel graph convolution network based on feature fusion for automatic image annotation. By fusing the sample features, the inherent defects of the features extracted by a single model are reduced, and the annotation performance under the condition of semi-supervised learning is improved. Experiments on three benchmark image annotation datasets show that this method is superior to the existing methods. [ABSTRACT FROM AUTHOR]

Intuitionistic fuzzy Best Worst method (IFBWM) is an effective method to deal with multi-criteria decision making problems via intuitionistic fuzzy reference comparisons of experts, which has attracted increasing attention from different decision fields. However, there are drawbacks in the existing IFBWMs, which can result in unreasoning and incorrect ranking order of criteria or alternatives. To overcome the existing drawbacks of the IFBWMs, we propose a new IFBWM considering the multiplicative consistency of intuitionistic fuzzy reference comparisons. First, combining the multiplicative consistent intuitionistic fuzzy preference relation with the fully multiplicative consistency of the intuitionistic fuzzy reference comparisons, we build the mathematical programming model for obtaining the normalized intuitionistic fuzzy weight vector. Then, we define the concept of the consistency ratio for evaluating the multiplicative consistency of intuitionistic fuzzy reference comparisons. Afterwards, an algorithm for repairing the inconsistency of intuitionistic fuzzy reference comparisons is developed, which only adjust the preference degree of the best criterion over the worst criterion. Subsequently we propose a new IFBWM with the multiplicative consistency. Furthermore, we apply the proposed method to Advanced Mathematics textbooks selection. The computational results show that the consistency ratio is 0.0091, implying the intuitionistic fuzzy reference comparisons are acceptable, and the ranking order of the alternatives C 1 ≻ C 2 ≻ C 4 ≻ C 3 is consistent with the initial judgment of experts and the actual usage of Advanced Mathematics textbooks in universities of China. The computational results of the practical example through the other methods reveal that there are some troubles in the consistency ratio, the weights of alternatives or the ranking order of the alternative. Therefore, the results suggest that the proposed method may have significant effects on the theoretical development and practical application of IFBWM. [ABSTRACT FROM AUTHOR]

Lignin is the largest natural aromatic biopolymer, but usually treated as industrial biomass waste. The development of lignin/polymer biocomposites can promote the high value utilization of lignin and the greening of polymers. However, the weak interfacial interaction between industrial lignin and polymer induces poor compatibility and serious agglomeration in polymer owing to the strong intermolecular force of lignin. As such, it is extremely difficult to prepare high performance lignin/polymer biocomposites. Recently, we proposed the strategy of in situ construction of interfacial dynamic bonds in lignin/polymer composites. By taking advantage of the abundant oxygen-containing polar groups of lignin, we inserted dynamic bonding connection such as hydrogen bonds and coordination bonds into the interphase between lignin and the polymer matrix to improve the interfacial interactions. Meanwhile, the natural amphiphilic structure characteristics of lignin were utilized to construct the hierarchical nanophase separation structure in lignin/polymer composites. The persistent problems of poor dispersity and interfacial compatibility of lignin in the polymer matrix were effectively solved. The lignin-modified polymer composites achieved simultaneously enhanced strength and toughness. This concise review systematically summarized the recent research progress of our group toward building high-performance lignin/polymer biocomposites through the design of interfacial dynamic bonds (hydrogen bonds, coordination bonds, and dynamic covalent bonds) between lignin and different polymer systems (polar plastics, rubber, polyurethane, hydrogels, and other polymers). Finally, the future development direction, main challenges, and potential solutions of lignin application in polymers were presented. [ABSTRACT FROM AUTHOR]

Intestinal ischemia/reperfusion (I/R) injury is a severe clinical condition without optimal diagnostic markers nor clear molecular etiological insights. Plasma exosomal circular RNAs (circRNAs) are valuable biomarkers and therapeutic targets for various diseases, but their role in intestinal I/R injury remains unknown. Here we screen the expression profile of circRNAs in intestinal tissue exosomes collected from intestinal I/R mice and identify circEZH2_005 as a significantly downregulated exosomal circRNA. In parallel, circEZH2_005 is also reduced in the plasma of clinical cardiac surgery patients who developed postoperative intestinal I/R injury. Exosomal circEZH2_005 displays a significant diagnostic value for intestinal injury induced by I/R. Mechanistically, circEZH2_005 is highly expressed in intestinal crypt cells. CircEZH2_005 upregulation promotes the proliferation of Lgr5+ stem cells by direct interaction with hnRNPA1, and enhanced Gprc5a stability, thereby alleviating I/R-induced intestinal mucosal damage. Hence, exosomal circEZH2_005 may serve as a biomarker for intestinal I/R injury and targeting the circEZH2_005/hnRNPA1/Gprc5a axis may be a potential therapeutic strategy for intestinal I/R injury. Currently, intestinal ischemia/reperfusion injury lacks ideal diagnostic markers and effective therapeutic targets. Here, the authors show exosomal circEZH2_005 is an intestinal injury biomarker and protects the intestine from ischemia/reperfusion injury by promoting crypt cell proliferation via affecting hnRNPA1- mediated Gprc5a stability. [ABSTRACT FROM AUTHOR]

Purpose: Docosahexaenoic acid (DHA) is an important omega-3 unsaturated fatty acid and has been widely applied in medicine, food additives, and feed ingredients. The fermentative production of DHA using microorganisms, including Schizochytrium sp., attracted much attention due to its high production efficiency and environment friendly properties. An efficient laboratory evolution approach was used to improve the strain's performance in this study. Methods: A multi-pronged laboratory evolution approach was applied to evolve high-yield DHA-producing Schizochytrium strain. We further employed comparative transcriptional analysis to identify transcriptional changes between the screened strain HS01 and its parent strain GS00. Results: After multiple generations of ALE, a strain HS01 with higher DHA content and lower saturated fatty acids content was obtained. Low nitrogen conditions were important for enhancing DHA biosynthesis in HS01. The comparative transcriptional analysis results indicated that during the fermentation process of HS01, the expression of key enzymes in the glycolysis, the pentose phosphate pathway and the tricarboxylic acid cycle were up-regulated, while the expression of polyketide synthase genes and fatty acid synthesis genes were similar to those in GS00. Conclusion: The results suggest that the improved DHA production capacity of HS01 is not due to enhancement of the DHA biosynthesis pathway, but rather related to modulation of central metabolism pathways. [ABSTRACT FROM AUTHOR]

Background: Pancreatic cancer (PC) is amongst the most lethal gastrointestinal tumors, which is the seventh leading reason of cancer-related mortality worldwide. Previous studies have indicated that circular RNAs (circRNAs), which is a new type of endogenous noncoding RNA (ncRNA), can mediate tumor progression in diverse tumor types including PC. Whereas precise roles regarding circRNAs and their underlying regulatory mechanisms in PC remain unknown. Methods: In the current study, we employed next generation sequencing (NGS) to characterize abnormally expressed circRNAs among PC tissues. Next, we assessed expression levels of one identified circRNA, circ-STK39, in PC cell lines and tissues. Then, using bioinformatics analysis, luciferase reporter, Transwell migration, EdU and CCK-8 assays, we examined the regulatory mechanisms and targets of circ-STK39. Finally, our group explored the circ-STK39 role in PC tumor growth and metastasis in vivo. Results: Our team discovered that circ-STK39 expression increased in PC tissues and cells, suggesting that circ-STK39 may have a role in PC progression. Downregulation of circ-STK39 inhibited PC proliferation and migration. Bioinformatics and luciferase reporter outcomes demonstrated that TRAM2 and miR-140-3p were circ-STK39 downstream targets. TRAM2 overexpression reversed the miR-140-3p overexpression effects upon migration, proliferation and the epithelial-mesenchymal transition (EMT). Conclusion: In this regard, we showed that circ-STK39 downregulation led to decreased migration, proliferation and the EMT of PC via the miR-140-3p/TRAM2 axis. [ABSTRACT FROM AUTHOR]

Over-track buildings in metro depots become more and more prevalent in metropolis with the increase in population and the decrease of construction land. However, the train-induced vibration greatly reduces the comfort of occupants living in the over-track buildings. Owing to complex vibration sources and numerous transfer paths, it is a challenging task to analyze and predict the vibration characteristics accurately in the building. In this paper, a field measurement campaign of vibration was conducted in the Guanhu metro depot in Guangzhou, China. A novel formulation based on the operational transfer path analysis (OTPA) and singular value decomposition (SVD) is proposed to analyze the measured data and predict the train-induced vibration in the building. In this study, the vibration contributions from each transfer path to the target points in the building were obtained and the predominant transfer paths were identified further. Besides, the vibration at target points in the building was predicted using the vibration at path points and transmissibility functions from transfer paths. This study provides insights into the prediction and evaluation of vibration transmissions from vibration source to upper floors of over-track building. [ABSTRACT FROM AUTHOR]

Centerless grinding is the main technology for mass processing of rotary parts, which can improve the roundness of rotary parts. With the demand for higher levels of roundness accuracy (within 0.3 μm), the development and operating costs of centerless grinding machines have increased dramatically. Taking cylindrical roller of bearing as research object, a new lapping process of double-disc arc-contact lapping (DDACL) is presented in this work. On the processing mechanism, it is expounded how DDACL overcomes the three fundamental technical difficulties of centerless grinding process in geometric lobing, workpiece instability and work-regenerative chatter vibration, and the unique rounding mechanism of DDACL process is established. At the same time, the machine has simple structure, low precision requirement, and low cost. After lapping with 45 steel groove on the developed test bench, the roundness accuracy of the cylindrical roller is effectively controlled within 0.2 μm, and the accuracy of some rollers can even reach 130 nm, which has significant precision advantages. According to the established DDACL rounding mechanism, the influence of 45 steel and PTFE groove material and electromagnetic driving force on the rounding speed of parts in the test is well explained. The process method proposed in this work paves the way for improving the roundness accuracy of other rotary parts, such as those seen in tapered roller and spherical roller. [ABSTRACT FROM AUTHOR]

Dictionary learning has drawn increasing attention for its impressive performance in obtaining the high-fidelity representations of data and extracting semantics. However, when there exists distribution divergence between source and target data, the representations of target data based on the learned dictionary from source data fail to reveal the intrinsic nature of target tasks, which consequently degrades the target performance severely. To tackle this problem, we propose a Shared Dictionary Learning (SDL) method in this paper. SDL learns a shared dictionary by implementing both geometric and statistical adaptations. SDL utilizes the Nyström method to exploit the geometric relationships between domains. Specifically, SDL uses the Nyström method to construct a variable source graph based on the target graph eigensystem and employs the Nyström approximation error to measure the distance between the variable source graph and the ground truth source graph to formalize the geometric divergence. Thus, a domain-invariant graph can be constructed by minimizing the approximation error and can be used to bridge two domains geometrically. Simultaneously, SDL captures the latent statistical commonality underlying two domains via minimizing the Maximum Mean Discrepancy (MMD) distance between domains. Finally, SDL achieves a shared dictionary and a set of corresponding new representations to handle cross-distribution data classification. Extensive experimental results on several popular datasets demonstrate the superiority of SDL. [ABSTRACT FROM AUTHOR]

To treat high-salt urea wastewater by microbial hydrolysis, it is necessary to overcome the dual problems of incomplete removal of nitrogen (N) from mixed strains and inhibition of microbial activity by high salt (NaCl) concentrations. In this paper, the mechanism of NaCl tolerance of Halomonas sp. H36 was investigated. Using molecular biology and enzymatic methods, it was proven that the strain's N-removal enzymes (urease; ammonia monooxygenase, AMO; nitrite reductase, NIR; nitrate reductase, NAR) played a key role in the removal of N, and the N-removal pathway was clarified. For the strain used to treat simulated ship domestic sewage, the urea nitrogen (CO(NH2)2-N)-removal rate was 88.52%, the ammonia nitrogen (NH4+-N)-removal rate was 91.16%, the total nitrogen (TN)-removal rate was 90.25%, and nitrite nitrogen (NO2−-N) and nitrate nitrogen (NO3−-N) did not accumulate. It was proven for the first time that Halomonas sp. H36 has the function of simultaneous urea hydrolysis-nitrification–denitrification with urea as the initial substrate and can simultaneously remove urea nitrogen and inorganic nitrogen from high-salt urea wastewater. [ABSTRACT FROM AUTHOR]

In this paper, we consider multi-sensor with partly overlapping field of view (FoV) in the labeled random finite set (L-RFS) framework. This is different from most existing multi-sensor tracking algorithms, where the sensors are assumed to have the same FoV. We describe the partly overlapping FoV by modeling probability field of detection for individual sensors in whole observation area and can be seen as the same range of FoV. We consider all these using generalized labeled multi-Bernoulli filter in labeled RFS framework. Besides, we also propose a measurement-driven target birth model. Finally, the effectiveness of the proposed algorithm is verified by experiments. [ABSTRACT FROM AUTHOR]

Copyright of Canadian Journal of Anaesthesia / Journal Canadien d'Anesthésie is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Current resolvable group targets tracking algorithms can only track a small number of group targets assuming the standard deviation vectors (SDVs) are known. Based on generalized labeled multi-Bernoulli filter, this paper first proposes a serial tracking algorithm, in order to save the computational cost of filter tracking large-batch and multi-structure group targets. Then, the least squares and the k-means algorithms are used to jointly estimate the SDVs and the group target states. The experimental simulation results verify the effectiveness of the proposed methods. [ABSTRACT FROM AUTHOR]

CH4 and CO2 are the main greenhouse effect contributors that cause global warming and undesirable climate change. Simultaneously converting CH4 and CO2 into value-added syngas via dry reforming (DRM) is an attractive route but highly active catalyst is demanded. A novel Ni-Pt/MgO-Al2O3 aerogel was developed via a sol-gel combined with supercritical deposition method and the effect of Pt doping in DRM was studied. The catalytic activities of Ni-Pt/MgO-Al2O3 aerogel catalysts followed the order: Ni10Pt0.1 > Ni10Pt0.05 > Ni10Pt0.2 > Ni10. 87% CH4 and 82% CO2 can be stably converted to syngas at 800 °C on the optimal Ni10Pt0.1. Characterizations including XRD, BET, HRTEM, H2-TPR, TG/DSC, and FESEM analysis revealed that increasing the Pt/Ni ratio from 0 to 0.01 helped to transfer Ni2Al2O4 spinel to bulk NiOx species, which is conducive to higher activity owing to the improvement of reducibility of the catalyst. However, a further increase in the Pt/Ni ratio to 0.02 resulted in the transformation of the bulk NiOx species to surface NiOx, which deteriorated the Ni-Pt catalyst performance. Kinetic experiments on the optimal Ni10Pt0.1 showed a CH4 activation energy of 31.8 kJ/mol, which is comparable to Pt-Ni catalysts with higher Pt contents. This work developed a promising catalyst for DRM and will help utilize abundant CH4 and CO2 resources in a clean and efficient way. Highlights: Ni–Pt bimetallic catalysts were prepared via sol–gel combined with supercritical deposition method. Ni species and their reduction behavior can be tuned by Pt/Ni ratio. Ni10Pt0.1 exhibited the best catalytic performance. [ABSTRACT FROM AUTHOR]

Few-shot learning (FSL), aiming to address the problem of data scarcity, is a hot topic of current researches. The most commonly used FSL framework is composed of two components: (1) Pre-train. Employing the base data to train a feature extraction model (FEM). (2) Meta-test. Utilizing the trained FEM to extract novel data's feature embedding and then recognize them with the to-be-designed classifier. Due to the constraint of limited labeled samples, some researchers attempt to exploit the unlabeled samples to strengthen the classifier by introducing self-training strategy. However, a single classifier (based on scarce labeled samples) usually incorrectly identifies unlabeled sample classes as its insufficient discrimination, which is dubbed as Single-Classifier-Misclassify-Data (SCMD) problem. To address this fundamental problem, we design a Co-learning (CL) method for FSL. Typically, we find that different classifiers have different adaptability to the same feature distribution. Hence we try to exploit two basic classifiers to separately infer pseudo-labels for unlabeled samples, and crossly expand them to the labeled data. The two complementary classifiers make the predicted accuracy more reliable. We evaluate our CL on five benchmark datasets (e.g., mini-ImageNet, tiered-ImageNet, CIFAR-FS, FC100, CUB), and can exceed other state-of-the-arts 0.63–4.6%. The outstanding performance has demonstrated the efficiency of our method. [ABSTRACT FROM AUTHOR]

Large scale labeled samples are expensive and difficult to obtain, hence few-shot learning (FSL), only needing a small number of labeled samples, is a dedicated technology. Recently, the graph-based FSL approaches have attracted lots of attention. It is helpful to model pair-wise relations among samples according to the similarity of features. However, the data in the reality usually have high-order relations, which can not be modeled by the traditional graph-based methods. To address this challenge, we introduce hypergraph structure and propose the Dual-Modal Hypergraph Few-Shot Learning (DMH-FSL) method to model the relations from different perspectives to model the high-order relations between samples. Specifically, we construct a dual-modal (e.g., feature-modal and label-modal) hypergraph, the feature-modal hypergraph construct incidence matrix with samples' features and the label-modal hypergraph construct incidence matrix with samples' labels. In addition, we employ two hypergraph convolution methods to perform flexible aggregation of samples from different modals. The proposed DMH-FSL method is easy to extend to other graph-based methods. We demonstrate the efficiency of our DMH-FSL method on three benchmark datasets. Our algorithm has at least an increase of 2.62% in Stanford40(from 72.20 to 74.82%), 0.85% in mini-ImageNet(from 50.33 to 51.18%) and 1.61% in USE-PPMI(from 78.77 to 80.38%) in few-shot learning experiments. What's more, the cross-domain experimental results evaluate our method's adaptability in real-world applications to some extent. [ABSTRACT FROM AUTHOR]

A novel raised source and recessed drain dopingless tunneling FET (RSRD-DL-TFET) is proposed in this paper. By using TCAD simulation, the DC electrical characteristics, energy-band diagrams, carrier concentrations and band-to-band tunneling generation rate of RSRD-DL-TFET are analyzed, as compared with planar dopingless tunneling FET (PDL-TFET). Besides, analog/RF and transient performances for both devices are also investigated, including transconductance, parasitic gate capacitance, cut-off frequency, gain bandwidth product and fall propagation delay. Benefiting from the design of raised source and recessed drain, the proposed device not only creates the line tunneling in the source region, but also reduces the ambipolar tunneling distance at drain/channel interface. As a result, the proposed RSRD-DL-TFET could provide higher on-current, steeper subthreshold swing, lower ambipolar current, higher cut-off frequency and better transient performance than PDL-TFET. [ABSTRACT FROM AUTHOR]

Segmented operations, such as segmented sum, segmented scan and segmented sort, are important building blocks for parallel irregular algorithms. We in this work propose a new parallel primitive called segmented merge. Its function is in parallel merging q sub-segments to p segments, both of possibly nonuniform lengths which easily cause the load balancing and the vectorization problems on massively parallel processors, such as GPUs. Our algorithm resolves these problems by first recording the boundaries of segments and sub-segments, then assigning roughly the same number of elements for GPU threads, and finally iteratively merging the sub-segments in each segment in the form of binary tree until there is only one sub-segment in each segment. We implement the segmented merge primitive on GPUs and demonstrate its efficiency on parallel sparse matrix transposition (SpTRANS) and sparse matrix–matrix multiplication (SpGEMM) operations. We conduct a comparative experiment with NVIDIA vendor library on two GPUs. The experimental results show that our algorithm achieve on average 3.94× (up to 13.09×) and 2.89× (up to 109.15×) speedup on SpTRANS and SpGEMM, respectively. [ABSTRACT FROM AUTHOR]

Utilizing the abundant biogas resources in an efficient and sustainable way helps to mitigate the energy and environmental problems. CH4 oxy-CO2 reforming reaction is a promising technic and an efficient catalyst for the reaction is in great demand. In this work, Ru was incorporated into monometallic Co/MgO-Al2O3 aerogel catalyst via supercritical drying process combined with sol–gel method and a series of Co-Ru bimetallic aerogel catalyst with different Ru contents was synthesized. Catalytic performance screening for the catalysts indicated that combination of Ru and Co significantly raised the activity of monometallic catalysts, e.g., the activity of Co15-Ru0.2 (Co content: 15 wt.%, Ru content: 0.2 wt.%) was 64.7% and 63.9% higher than those of the monometallic Co15 and Ru0.5 catalysts, respectively. The results also indicated that Co/Ru ratio had a great influence on the catalytic activity: Co15-Ru0.2 > Co15-Ru0.1 ≈ Co15-Ru0.5 > Co15-Ru0.05 > Ru0.5 > Co15. XRD, BET, H2-TPR, XPS, TEM, TG/DSC, and FESEM analyses were performed, disclosing that the reducibility and active metal particle size of the catalysts can be manipulated by the Co/Ru ratio due to the transportation of bulk Co2+ to the surface. The Co15-Ru0.2 catalyst had a mean active metal particle size below 10 nm and medium reducibility, resulting in a high activity and stable performance. Highlights: A Co-Ru/MgO-Al2O3 aerogel was prepared by sol-gel combined with supercritical deposition method. The addition of Ru destabilized the Co spinel and promoted the reduction of Co species. The synergy between Co-Ru promoted the catalytic performance which were better than the monometallic catalysts. Co15-Ru0.2 showed the highest CH4 reforming activity and resistance to inactive carbon. [ABSTRACT FROM AUTHOR]

Artificial muscle materials promise incredible applications in actuators, robotics and medical apparatus, yet the ability to mimic the full characteristics of skeletal muscles into synthetic materials remains a huge challenge. Herein, inspired by the dynamic sacrificial bonds in biomaterials and the self-strengthening of skeletal muscles by physical exercise, high performance artificial muscle material is prepared by rearrangement of sacrificial coordination bonds in the polyolefin elastomer via a repetitive mechanical training process. Biomass lignin is incorporated as a green reinforcer for the construction of interfacial coordination bonds. The prepared artificial muscle material exhibits high actuation strain (>40%), high actuation stress (1.5 MPa) which can lift more than 10,000 times its own weight with 30% strain, characteristics of excellent self-strengthening by mechanical training, strain-adaptive stiffening, and heat/electric programmable actuation performance. In this work, we show a facile strategy for the fabrication of intelligent materials using easily available raw materials. Artificial muscles have a wide range of applications yet truly mimetic designs remain a challenge. Here, the authors use dynamic sacrificial bonds which are rearranged via a mechanical training process to optimise the characteristics of self-strengthening, strain-adaptive stiffening and actuation. [ABSTRACT FROM AUTHOR]

Recent years have witnessed the proliferation of online fashion blogs and communities, where a large amount of fashion model images with chic clothes in various scenarios are publicly available. To facilitate users to find the corresponding clothes, we focus on studying how to generate pure wellshaped clothing items with the best view from the complex model images. Towards this end, inspired by painting, where the initial sketches and following coloring are both essential, we propose a two-stage shape-constrained clothing generative framework, dubbed as PaintNet. PaintNet comprises two coherent components: shape predictor and texture renderer. The shape predictor is devised to predict the intermediate shape map for the to-be-generated clothing item based on the latent representation learning, while the texture renderer is introduced to generate the final clothing image with the guidance of the predicted shape map. Extensive qualitative and quantitative experiments conducted on the public Lookbook dataset verify the effectiveness of PaintNet in clothing generation from fashion model images. Moreover, we also explore the potential of PaintNet in the task of cross-domain clothing retrieval, and the experiment results show that PaintNet can achieve, on average, 5.34% performance improvement over the traditional non-generative retrieval methods. [ABSTRACT FROM AUTHOR]

The prominent protein producing workhorse Trichoderma reesei secretes a typical yellow pigment that is synthesized by a gene cluster including two polyketide synthase encoding genes sor1 and sor2. Two transcription factors (YPR1 and YPR2) that are encoded in the same cluster have been shown to regulate the expression of the sor genes. However, the physiological relevance of the yellow pigment synthesis in T. reesei is not completely clear. In this study, a yellow pigment hyper-producer OEypr1 and three yellow pigment non-producers, OEypr1-sor1, Δypr1, and OEypr2, were constructed. Their phenotypic features in mycelial growth, conidiation, cell wall integrity, stress tolerance, and cellulase production were determined. Whereas hyperproduction of the yellow pigment caused significant defects in all the physiological aspects tested, the non-producers showed similar colony growth, but improved conidiation, maintenance of cell wall integrity, and stress tolerance compared to the control strain. Moreover, in contrast to the severely compromised extracellular cellobiohydrolase production in the yellow pigment hyperproducer, loss of the yellow pigment hardly affected induced cellulase gene expression. Our results demonstrate that interfering with the yellow pigment synthesis constitutes an engineering strategy to endow T. reesei with preferred features for industrial application. [ABSTRACT FROM AUTHOR]

Summary: An inflammatory myofibroblastic tumor (IMT) is a rare invasive soft tissue mass with intramuscular penetration that is primarily treated via a surgical procedure. However, with unclear boundaries and a high rate of relapse, there is no standard treatment for recurrence or unresectable tumors. It is noteworthy that approximately half of IMTs harbor genetic rearrangements of the anaplastic lymphoma kinase (ALK). ALK inhibitors have been used successfully in the treatment of IMTs with a variety of ALK fusions. Here, we present a case of a 15-year-old patient with IMT around the hip. Next-generation sequencing (NGS) revealed an LRRFIP1-ALK fusion, which has not yet been reported in the literature. Crizotinib, an ALK inhibitor, was effective in the treatment of this patient, indicating that ALK inhibitors may be effective for IMT with LRRFIP1-ALK fusions. This report expands the list of gene fusions in IMTs and highlights a new target for treatment. [ABSTRACT FROM AUTHOR]

The purpose of this study was to evaluate the functional results, complications and related factors of prosthesis reconstruction after malignant tumor resection of primary proximal humeral, and also evaluate whether soft tissue reconstruction with mesh patch and anchors can improve clinical results. From 2002 to 2016, forty-one patients were enrolled in this study. The pathological diagnosis contained 27 cases of osteosarcoma, 7 cases of chondrosarcoma and other primary malignant bone tumors. Both mesh patch and anchors were used in the reconstruction of joint capsule and the surrounding soft tissues in 27 cases. The mean postoperative follow-up was 60.6 months. The average active abduction angle and passive abduction angle was 33.5 (5–71) degrees and 72.4 (52–104) degrees. The prosthetic humeral head displacement was over 2 cm in 5 cases (12.2%). The average MSTS score was 23.1. The overall 5-year survival rate of prosthesis was 88.2%. The length of osteotomy, whether preserving deltoid muscle, whether applying mesh patch and anchors had significant effects on the abduction angle of shoulder joint; the length of osteotomy, whether applying mesh patch and anchors had significant effects on the degree of upward displacement of prosthesis. The application of both mesh patch and anchors in prosthesis reconstruction achieved more stable result and better function of shoulder joint. To ensure the stability of shoulder joint and the firm wrapping of surrounding soft tissue are key factors affecting the postoperative function. [ABSTRACT FROM AUTHOR]

Highlights: A facile strategy is proposed to construct novel CoO and NiO electrode material with interior (Cu+) and exterior (metallic Cu0) decoration for supercapacitor. The co-doped CoO electrode presents four times higher specific capacitance and more cycle stability than that of pristine CoO. The conductivity and electron transport rate of CoO with Cu0/Cu+ co-doping have been effectively improved and confirmed by detailed theoretical calculations.Although CoO is a promising electrode material for supercapacitors due to its high theoretical capacitance, the practical applications still suffering from inferior electrochemical activity owing to its low electrical conductivity, poor structural stability and inefficient nanostructure. Herein, we report a novel Cu0/Cu+ co-doped CoO composite with adjustable metallic Cu0 and ion Cu+ via a facile strategy. Through interior (Cu+) and exterior (Cu0) decoration of CoO, the electrochemical performance of CoO electrode has been significantly improved due to both the beneficial flower-like nanostructure and the synergetic effect of Cu0/Cu+ co-doping, which results in a significantly enhanced specific capacitance (695 F g−1 at 1 A g−1) and high cyclic stability (93.4% retention over 10,000 cycles) than pristine CoO. Furthermore, this co-doping strategy is also applicable to other transition metal oxide (NiO) with enhanced electrochemical performance. In addition, an asymmetric hybrid supercapacitor was assembled using the Cu0/Cu+ co-doped CoO electrode and active carbon, which delivers a remarkable maximal energy density (35 Wh kg−1), exceptional power density (16 kW kg−1) and ultralong cycle life (91.5% retention over 10,000 cycles). Theoretical calculations further verify that the co-doping of Cu0/Cu+ can tune the electronic structure of CoO and improve the conductivity and electron transport. This study demonstrates a facile and favorable strategy to enhance the electrochemical performance of transition metal oxide electrode materials. [ABSTRACT FROM AUTHOR]

We propose that clusters interconnected with network topologies having minimal mean path length will increase their processing speeds. We approach our heuristic by constructing clusters of up to 32 nodes having torus, ring, Chvatal, Wagner, Bidiakis and optimal topology for minimal mean path length and by simulating the performance of 256 nodes clusters with the same network topologies. The optimal (or near-optimal) low-latency network topologies are found by minimizing the mean path length of regular graphs. The selected topologies are benchmarked using ping-pong messaging, the MPI collective communications and the standard parallel applications including effective bandwidth, FFTE, Graph 500 and NAS parallel benchmarks. We established strong correlations between the clusters' performances and the network topologies, especially the mean path lengths, for a wide range of applications. In communication-intensive benchmarks, optimal graphs enabled network topologies with multifold performance enhancement in comparison with mainstream graphs. It is striking that mere adjustment of the network topology suffices to reclaim performance from the same computing hardware. [ABSTRACT FROM AUTHOR]

Alzheimer's disease (AD) is a multifactorial disease that affects more than 5 million Americans. Multiple pathways might be involved in the AD pathogenesis. The implication of lipid genetic susceptibility on brain gene expression is yet to be investigated. The current study included 192 brain samples from AD patients who were enrolled in the ROSMAP study. The samples were genotyped and imputed to the HRC Reference Panel. Lipid polygenetic risk score was constructed from the weighted sum of genetic variants associated with low-density lipoprotein cholesterol (LDL-C). The gene expression was profiled by RNA sequencing, and the association of gene expression with lipid polygenetic risk scores was tested by linear regression models adjusted for age, sex and APOE e4 alleles. Three genes were found to associate with lipid polygenetic risk scores, including HMCN2 (P = 3.6 × 10–7), PDLIM5 (P = 1.2 × 10–6), and FHL5 (P = 2.0 × 10–6). Network analysis revealed multiple related pathways, including dopaminergic synapse (P = 4.5 × 10–5), circadian entrainment (P = 1.1 × 10–4), and cholinergic synapse (P = 2.3 × 10–4). Our study underscores the importance of lipid regulation and metabolism to AD heterogeneity. [ABSTRACT FROM AUTHOR]