In this paper, we present a novel deep learning based method for video anomaly detection and localization. The key idea of our approach is that the latent space representations of normal samples are ...trained to accord with a specific prior distribution by the proposed deep neural network - Multivariate Gaussian Fully Convolution Adversarial Autoencoder (MGFC-AAE), while the latent representations of anomalies do not. In order to extract deep features from input samples as latent representations, a convolutional neural network (CNN) is employed for the encoder of the deep network. Based on the probability that the test sample is associated with the prior distribution, an energy-based method is applied to obtain its anomaly score. A two-stream framework is utilized to integrate the appearance and motion cues to achieve more comprehensive detection results, taking the gradient and optical flow patches as inputs for each stream. Besides, a multi-scale patch structure is put forward to handle the perspective of some video scenes. Experiments are conducted on three public datasets, results verify that our framework can accurately detect and locate abnormal objects in various video scenes, achieving competitive performance when compared with other state-of-the-art works.
Designable and ultrathin covalent organic framework nanosheets (CONs) with good photoelectric activity are promising candidates for the construction of photoelectrochemical (PEC) biosensors for the ...detection of low-abundance biological substrates. However, achieving highly sensitive PEC properties by using emerging covalent organic framework nanosheets (CONs) remains a great challenge due to the polymeric nature and poor photoelectric activity of CONs. Herein, we report for the first time the preparation of novel composites and their PEC sensing properties by electrostatic self-assembly of ultrathin CONs (called TTPA-CONs) with Ti3C2Tx. The prepared TTPA-CONs/Ti3C2Tx composites can be used as photocathodes for PEC detection of prostate-specific antigen (PSA) with high sensitivity, low detection limit, and good stability. This work not only expands the application of CONs but also opens new avenues for the development of efficient PEC sensing platforms.
The post spacing of slit dams is a key parameter that controls the trapping efficiency of these open-type countermeasures. In this study we conduct flume experiments of quasi-monodisperse spherical ...particles passing through slits to investigate the relationship between the trapping efficiency and the pile-up geometry, as well as the impact of the latter on the interaction between granular flows and slit dams. The ratio of the slit width relative to the particle size b/d is varied while the flume inclination is held constant. Tests reveal that at a critical ratio b/d∼2.3 trapping is most unstable. The trapping efficiency influences the geometry of the granular pile-up that deposits behind the dam after it has been jammed. When >52% of the granular mass is trapped (occurring at b/d≤2.3), the height of the final deposit changes with the trapping efficiency, whereas below this threshold value (when b/d≥3.1)the flow-wise length becomes more sensitive to the trapping. Impact force measurements further shed light on the mechanisms relating the trapping efficiency and the geometry and their effect on the total force fluctuations.
•Flume experiments of granular materials impacting slit dams reveal a critical relative post spacing of b/d∼2.3 at which the trapping is considered most unstable.•The deposit geometry depends on the trapping efficiency. When the efficiency is >52 %, deposit height is sensitive to it but below this threshold only the flow-wise length changes.•The total impact force reflects the pile-up process and geometry. The trapping efficiency influences the magnitude of the impact force and its evolution as particles exit from the slits.
Enhancing network feature representation capabilities and reducing the loss of image details have become the focus of semantic segmentation task. This work proposes the bilateral attention network ...for semantic segmentation. The authors embed two attention modules in the encoder and decoder structures . Specifically, high‐level features of the encoder structure integrate all channel maps through dense channel relationships learned by the channel correlation coefficient attention module. The positively correlated channels promote each other, and the negatively correlated channels suppress each other. In the decoder structure, low‐level features selectively emphasize the edge detail information in the feature map through the position attention module. The feature expression of semantic segmentation is improved by feature fusion of the two attention modules to obtain more accurate segmentation results . Finally, to verify the effectiveness of the model, the authors conduct experiments on the PASCAL VOC 2012 and Cityscapes scene analysis benchmark data sets and achieve a mean intersection‐over‐union of 74.92% and 66.63%, respectively.
Metallic zinc (Zn) presents a compelling alternative to conventional electrochemical energy storage systems due to its environmentally friendly nature, abundant availability, high water ...compatibility, low toxicity, low electrochemical potential (−0.762 V vs. SHE), and cost-effectiveness. While considerable efforts have been devoted to enhancing the physical and chemical properties of zinc-ion battery materials to improve battery efficiency and longevity, research on multi-physics coupled modeling for a deeper understanding of battery performance remains relatively scarce. In this study, we established a comprehensive two-dimensional model for single-flow zinc–nickel redox batteries to investigate electrode reactions, current-potential behaviors, and concentration distributions, leveraging theories such as Nernst–Planck and Butler–Volmer. Additionally, we explored the distribution of the velocity field using the Brinkman theory in porous media and the Navier–Stokes equations in free-flow channels. The validated model, informed by experimental data, not only provides insights into the performance of the battery, but also offers valuable recommendations for advancing single-flow zinc–nickel battery technology. Our findings offer promising avenues for enhancing the design and performance of not only zinc–nickel flow batteries, but also applicable for other flow battery designs.
The corner reflector is an effective means of interference for radar seekers due to its high jamming intensity, wide frequency band, and combat effectiveness ratio. Properly arranging multiple corner ...reflectors in an array can form dilution jamming that resembles ships, substantially enhancing the interference effect. This results in a significant decline in the precision attack efficiency of radar seekers. Hence, it is critical to accurately identify corner reflector array. The common recognition methods involve extracting features on the high-resolution range profile (HRRP) and polarization domain. However, the former is constrained by the number of corner reflectors, while the latter is affected by the accuracy of polarization measurement, both of which have limited performance on the identification of corner reflector array. In terms of the evident variations in physical structures, there must be differences in their scattering characteristics. To highlight the differences, this paper proposes a new method based on the concept of mismatched filtering, which involves changing the frequency modulation slope of the chirp signal in the filter. Then, the variance of width and intervals within a specific scope are extracted as features to characterize these differences, and an identification process is designed in combination with the support vector machine. The simulation experiments demonstrate that the proposed method exhibits stable discriminative performance and can effectively combat dilution jamming. Its accuracy rate exceeds 0.86 when the signal-to-noise ratio is greater than 0 dB. Compared to the HRRP methods, the recognition accuracy of the proposed algorithm improves 15% in relation to variations in the quantity of corner reflectors.
Two-dimensional (2D) covalent organic framework nanosheets (CONs) are attracting increasing research attention because of their unique properties derived from their ultrathin thickness, high ...surface-to-volume atomic ratio, and extremely large surface area. 2D CONs can provide high transport pathways for charge carriers (e.g., electrons, holes and ions) through either the conjugated skeletons or the open channels. Therefore, they have shown great potential in energy related applications. In this review, we firstly introduce the recent developments and characteristics of 2D CONs by focusing on the two typical synthetic methods, i.e., top-down and bottom-up methods. Then, the energy-related applications in energy storage and conversion of 2D CONs are summarized. Finally, we give our personal views on the challenges and perspectives for the future research of 2D CONs and their composites.
Covalent organic frameworks nanosheets (CONs) have received extensive attention due to their unique properties and wide applications. This review summarizes the synthetic methods and applications of CONs, particularly in the field of energy storage (e.g., batteries) and energy conversion (e.g., photocatalysis and electrocatalysis), providing guidance and perspectives for the future research of 2D CONs and their composites
Display omitted .
Abstract
Background
Hepatocellular carcinoma (HCC) was the sixth common malignancies characteristic with highly aggressive in the world. It was well established that tumor mutation burden (TMB) act ...as indicator of immunotherapeutic responsiveness in various tumors. However, the role of TMB in tumor immune microenvironment (TIME) is still obscure.
Method
The mutation data was analyzed by employing “maftools” package. Weighted gene co-expression network analysis (WGCNA) was implemented to determine candidate module and significant genes correlated with TMB value. Differential analysis was performed between different level of TMB subgroups employing R package “limma”. Gene ontology (GO) enrichment analysis was implemented with “clusterProfiler”, “enrichplot” and “ggplot2” packages. Then risk score signature was developed by systematical bioinformatics analyses. K-M survival curves and receiver operating characteristic (ROC) plot were further analyzed for prognostic validity. To depict comprehensive context of TIME, XCELL, TIMER, QUANTISEQ, MCPcounter, EPIC, CIBERSORT, and CIBERSORT-ABS algorithm were employed. Additionally, the potential role of risk score on immune checkpoint blockade (ICB) immunotherapy was further explored. The quantitative real-time polymerase chain reaction was performed to detect expression of HTRA3.
Results
TMB value was positively correlated with older age, male gender and early T status. A total of 75 intersection genes between TMB-related genes and differentially expressed genes (DEGs) were screened and enriched in extracellular matrix-relevant pathways. Risk score based on three hub genes significantly affected overall survival (OS) time, infiltration of immune cells, and ICB-related hub targets. The prognostic performance of risks score was validated in the external testing group. Risk-clinical nomogram was constructed for clinical application. HTRA3 was demonstrated to be a prognostic factor in HCC in further exploration. Finally, mutation of TP53 was correlated with risk score and do not interfere with risk score-based prognostic prediction.
Conclusion
Collectively, a comprehensive analysis of TMB might provide novel insights into mutation-driven mechanism of tumorigenesis further contribute to tailored immunotherapy and prognosis prediction of HCC.
Time-efficient anomaly detection and localization in video surveillance still remains challenging due to the complexity of "anomaly". In this paper, we propose a cuboid-patch-based method ...characterized by a cascade of classifiers called a spatial-temporal cascade autoencoder (ST-CaAE), which makes full use of both spatial and temporal cues from video data. The ST-CaAE has two main stages, defined by two proposed neural networks: a spatial-temporal adversarial autoencoder (ST-AAE) and a spatial-temporal convolutional autoencoder (ST-CAE). First, the ST-AAE is used to preliminarily identify anomalous video cuboids and exclude normal cuboids. The key idea underlying ST-AAE is to obtain a Gaussian model to fit the distribution of the regular data. Then in the second stage, the ST-CAE classifies the specific abnormal patches in each anomalous cuboid with reconstruction error based strategy that takes advantage of the CAE and skip connection. A two-stream framework is utilized to fuse the appearance and motion cues to achieve more complete detection results, taking the gradient and optical flow cuboids as inputs for each stream. The proposed ST-CaAE is evaluated using three public datasets. The experimental results verify that our framework outperforms other state-of-the-art works.
Hepatocellular carcinoma (HCC) ranks the sixth prevalent tumors with high mortality globally. Alternative splicing (AS) drives protein diversity, the imbalance of which might act an important factor ...in tumorigenesis. This study aimed to construct of AS-based prognostic signature and elucidate the role in tumor immune microenvironment (TIME) and immunotherapy in HCC.
Univariate Cox regression analysis was performed to determine the prognosis-related AS events and gene set enrichment analysis (GSEA) was employed for functional annotation, followed by the development of prognostic signatures using univariate Cox, LASSO and multivariate Cox regression. K-M survival analysis, proportional hazards model, and ROC curves were conducted to validate prognostic value. ESTIMATE R package, ssGSEA algorithm and CIBERSORT method and TIMER database exploration were performed to uncover the context of TIME in HCC. Quantitative real-time polymerase chain reaction was implemented to detect ZDHHC16 mRNA expression. Cytoscape software 3.8.0 were employed to visualize AS-splicing factors (SFs) regulatory networks.
A total of 3294 AS events associated with survival of HCC patients were screened. Based on splicing subtypes, eight AS prognostic signature with robust prognostic predictive accuracy were constructed. Furthermore, quantitative prognostic nomogram was developed and exhibited robust validity in prognostic prediction. Besides, the consolidated signature was significantly correlated with TIME diversity and ICB-related genes. ZDHHC16 presented promising prospect as prognostic factor in HCC. Finally, the splicing regulatory network uncovered the potential functions of splicing factors (SFs).
Herein, exploration of AS patterns may provide novel and robust indicators (i.e., risk signature, prognostic nomogram, etc.,) for prognostic prediction of HCC. The AS-SF networks could open up new approach for investigation of potential regulatory mechanisms. And pivotal players of AS events in context of TIME and immunotherapy efficiency were revealed, contributing to clinical decision-making and personalized prognosis monitoring of HCC.