Search

This paper investigates the capability of plain Vision Transformers (ViTs) for semantic segmentation using the encoder–decoder framework and introduce SegViTv2. In this study, we introduce a novel Attention-to-Mask (ATM) module to design a lightweight decoder effective for plain ViT. The proposed ATM converts the global attention map into semantic masks for high-quality segmentation results. Our decoder outperforms popular decoder UPerNet using various ViT backbones while consuming only about 5 % of the computational cost. For the encoder, we address the concern of the relatively high computational cost in the ViT-based encoders and propose a Shrunk++ structure that incorporates edge-aware query-based down-sampling (EQD) and query-based up-sampling (QU) modules. The Shrunk++ structure reduces the computational cost of the encoder by up to 50 % while maintaining competitive performance. Furthermore, we propose to adapt SegViT for continual semantic segmentation, demonstrating nearly zero forgetting of previously learned knowledge. Experiments show that our proposed SegViTv2 surpasses recent segmentation methods on three popular benchmarks including ADE20k, COCO-Stuff-10k and PASCAL-Context datasets. The code is available through the following link: https://github.com/zbwxp/SegVit. [ABSTRACT FROM AUTHOR]

Each year, underwater remotely operated vehicles (ROVs) collect thousands of hours of video of unexplored ocean habitats revealing a plethora of information regarding biodiversity on Earth. However, fully utilizing this information remains a challenge as proper annotations and analysis require trained scientists' time, which is both limited and costly. To this end, we present a Dataset for Underwater Substrate and Invertebrate Analysis (DUSIA), a benchmark suite and growing large-scale dataset to train, validate, and test methods for temporally localizing four underwater substrates as well as temporally and spatially localizing 59 underwater invertebrate species. DUSIA currently includes over ten hours of footage across 25 videos captured in 1080p at 30 fps by an ROV following pre-planned transects across the ocean floor near the Channel Islands of California. Each video includes annotations indicating the start and end times of substrates across the video in addition to counts of species of interest. Some frames are annotated with precise bounding box locations for invertebrate species of interest, as seen in Fig. 1. To our knowledge, DUSIA is the first dataset of its kind for deep sea exploration, with video from a moving camera, that includes substrate annotations and invertebrate species that are present at significant depths where sunlight does not penetrate. Additionally, we present the novel context-driven object detector (CDD) where we use explicit substrate classification to influence an object detection network to simultaneously predict a substrate and species class influenced by that substrate. We also present a method for improving training on partially annotated bounding box frames. Finally, we offer a baseline method for automating the counting of invertebrate species of interest. [ABSTRACT FROM AUTHOR]

Evidence from previous epidemiological studies on the effect of physical activity on the risk of Alzheimer's disease (AD) is conflicting. We performed a two-sample Mendelian randomization analysis to verify whether physical activity is causally associated with AD. This study used two-sample Mendelian randomization (MR) analysis to estimate the association between physical activity (including overall activity, sedentary behavior, walking, and moderate-intensity activity) and AD. Genetic instruments for physical activity were obtained from published genome-wide association studies (GWAS) including 91,105 individuals from UK Biobank. Summary-level GWAS data were extracted from the International Genomics of Alzheimer's Project IGAP (21,982 patients with AD and 41,944 controls). Inverse Variance Weighted (IVW) was used to estimate the effect of physical activity on AD. Sensitivity analyses including weighted median, MR-Egger, MR-PRESSO, and leave-one-out analysis were used to estimate pleiotropy and heterogeneity. Mendelian randomization evidences suggested a protective relationship between walking and AD (odds ratio (OR) = 0.30, 95% confidence interval (CI), 0.13–0.68, P = 0.0039). Genetically predicted overall activity, sedentary behavior, and moderate-intensity activity were not associated with AD. In summary, this study provided evidence that genetically predicted walking might associate with a reduced risk of AD. Further research into the causal association between physical activity and AD could help to explore the real relationship and provide more measures to reduce AD risk. [ABSTRACT FROM AUTHOR]

Highlights: A new way to achieve high yield and large area oxide thin film transfer is developed. Three different film compositions are demonstrated: SrRuO3, CeO2, and CeO2/STO nanocomposite films. Cracks, wrinkles, and damages are prevented by the new transfer method. They are commonly introduced by conventional transfer processes. Vertically aligned nanocomposite (VAN) structures can further improve the transfer yield. Possible mechanisms related to increased fracture toughness are proposed. We have opened up a route to large-scale oxide thin-film-based electronic device applications.In this work, we have developed a new method for manipulating and transferring up to 5 mm × 10 mm epitaxial oxide thin films. The method involves fixing a PET frame onto a PMMA attachment film, enabling transfer of epitaxial films lifted-off by wet chemical etching of a Sr3Al2O6 sacrificial layer. The crystallinity, surface morphology, continuity, and purity of the films are all preserved in the transfer process. We demonstrate the applicability of our method for three different film compositions and structures of thickness ~ 100 nm. Furthermore, we show that by using epitaxial nanocomposite films, lift-off yield is improved by ~ 50% compared to plain epitaxial films and we ascribe this effect to the higher fracture toughness of the composites. This work shows important steps towards large-scale perovskite thin-film-based electronic device applications. [ABSTRACT FROM AUTHOR]

APTEEN is a typical routing protocol for wireless sensor networks, but when clustering, cluster heads are randomly selected, which makes it easy to select nodes with low residual energy as cluster heads, thus forming network holes. At the same time, in multi hop transmission, the cluster head near the sink node is overloaded due to forwarding a large amount of data. Aiming at these problems, this paper introduces PSO algorithm and proposes ADCH-EPE-APTEEN routing protocol. In order to make the particle swarm optimization algorithm more suitable for routing protocol, this paper first proposes DCA-PSO based on particle swarm optimization algorithm. DCA-PSO uses the classification adaptive change inertia weight for different states of particles in the optimization process. At the same time, dynamic learning factor is used to improve the influence of particle experience and other particle experience on convergence speed and to improve the optimization accuracy and speed of the algorithm. Secondly, when APTEEN routing protocol is networked, the cluster head is selected by DCA-PSO, in which the residual energy of nodes, the location of nodes, and the energy distribution around nodes are fully taken into consideration. Thirdly, considering the distance between nodes and the sink node, the residual energy and the energy distribution around nodes, a front area is set up in the cluster located in the upper half and the assistant cluster head is selected in the front area through the DCA-PSO algorithm. The simulation results show that the DCA-PSO algorithm can significantly improve the search speed and precision compared with the PSO algorithm. Compared with APTEEN routing protocol, the ADCH-EPE-APTEEN routing protocol can reduce the network energy consumption rate and extend network lifetime by 173%. [ABSTRACT FROM AUTHOR]

In typical CuNWs-based conductors, CuNWs are deposited on or semi-embedded in the surface of substrates, which cannot avoid the oxidation. Here, we report a strategy to fabricate highly conductive and stretchable conductor with enhanced reliability and robustness through fully embedding CuNWs into the surface layer of poly(dimethylsiloxane) (PDMS) and followed with a high-intensity pulsed light technique. The light energy absorbed by the film not only removes the oxides on the surface of nanowires but also enhances the inter-nanowire connection to achieve high conductivity. The fully embedded CuNWs/PDMS structure showed excellent oxidation resistance in high-humidity and high-temperature environments (85 °C, 85% RH), and maintained a low sheet resistance when subjected to 30% of strain for 1000 cycles. Stretchable dipole antenna was fabricated using the fully embedded CuNWs/PDMS conductors, which retained its sensitivity to specific radio frequencies even after 500 cycles of the stretching/releasing process. Furthermore, a stretchable conductive heater was demonstrated, verifying the applicability of our fully embedded conductor for wearable electronics. We believe this work might open up new opportunities for the wide range of practical applications of CuNW-based conductors. [ABSTRACT FROM AUTHOR]

A serious challenge faced by manufacturers of large-aperture aspheric optical components of glass ceramics is the long processing time. Ultrasonic vibration–assisted grinding (UVG) allows one to effectively shorten the subsequent polishing process by several times, which is essential for grinding of aspheric components. However, the surface creation mechanism of UVG-treated glass ceramics is rarely studied. Herein, rotary ultrasonic vibration–assisted vertical grinding (RUVG) and parallel grinding (RUPG) are applied to polish the aspheric glass ceramics. Particular attention is paid to the surface formation mechanism of UVG-processed ceramics. The single-grain kinematic functions are created and the contact characteristics between the grinding wheel and aspheric surface are analyzed for the two UVG methods in terms of contact area, velocity, and trajectory. In addition, aspheric grinding texture is simulated and comparative experiments are conducted correspondingly. According to the results, the rotary ultrasonic vibration mainly influences the microscopic grinding marks. Besides, the aspheric surface form accuracy of Pt and RMS value in RUVG is 2.16 and 3.71 times lower than those in RUPG, respectively, whereas the surface roughness–related parameters (mean deviation Sa and maximum height of profile Sz) in RUVG are 6.36% and 4.56% higher than those in RUPG. This indicates that RUVG is more suitable for high precision and efficiency grinding of the aspheric surface than RUPG due to the fact that the polishing depth is primarily determined by surface form accuracy rather than surface roughness. Thus, the current research enables an in-depth understanding of surface generation mechanism in rotary ultrasonic vibration–assisted grinding, pointing out its benefits in the high-efficiency aspheric surface manufacturing. [ABSTRACT FROM AUTHOR]

Zizania latifolia is easily lignified after harvesting, leading to the degradation of food quality and commercial value. Thus, this study evaluated the effect of radio frequency (RF) treatments on lignification inhibition of Zizania latifolia. The results showed that the lignin content of Zizania latifolia treated with RF decreased significantly compared with the control group. At the 7th day of storage, the phenylalanine ammonia lyase activity of the 90 W RF treatment group decreased by 52.9% compared with the initial value. The activities of peroxidase and polyphenol oxidase in the stems of Zizania latifolia were significantly (p < 0.05) decreased after RF treatments. Besides, a decrease in conversion rate of O2− and H2O2 to downstream products was observed, indicating that the related invertases were inhibited by RF treatment. All of these showed that RF treatments contribute to inhibit or delay the lignification of Zizania latifolia, providing a better taste and quality for products. [ABSTRACT FROM AUTHOR]

Recombinant human thrombopoietin (rHuTPO) is a drug that is used clinically to promote megakaryocyte and platelet generation. Here, we report the mitigative effect of rHuTPO (administered after exposure) against severe whole body irradiation in mice. Injection of rHuTPO for 14 consecutive days following exposure significantly improved the survival rate of lethally irradiated mice. RHuTPO treatment notably increased bone marrow cell density and LSK cell numbers in the mice after sub-lethal irradiation primarily by promoting residual HSC proliferation. In lethally irradiated mice with hematopoietic cell transplantation, rHuTPO treatment increased the survival rate and enhanced hematopoietic cell engraftment compared with the placebo treatment. Our observations indicate that recombinant human TPO might have a therapeutic role in promoting hematopoietic reconstitution and HSC engraftment. [ABSTRACT FROM AUTHOR]

There is increasing focus on applying deep learning methods to electrocardiograms (ECGs), with recent studies showing that neural networks (NNs) can predict future heart failure or atrial fibrillation from the ECG alone. However, large numbers of ECGs are needed to train NNs, and many ECGs are currently only in paper format, which are not suitable for NN training. We developed a fully-automated online ECG digitisation tool to convert scanned paper ECGs into digital signals. Using automated horizontal and vertical anchor point detection, the algorithm automatically segments the ECG image into separate images for the 12 leads and a dynamical morphological algorithm is then applied to extract the signal of interest. We then validated the performance of the algorithm on 515 digital ECGs, of which 45 were printed, scanned and redigitised. The automated digitisation tool achieved 99.0% correlation between the digitised signals and the ground truth ECG (n = 515 standard 3-by-4 ECGs) after excluding ECGs with overlap of lead signals. Without exclusion, the performance of average correlation was from 90 to 97% across the leads on all 3-by-4 ECGs. There was a 97% correlation for 12-by-1 and 3-by-1 ECG formats after excluding ECGs with overlap of lead signals. Without exclusion, the average correlation of some leads in 12-by-1 ECGs was 60–70% and the average correlation of 3-by-1 ECGs achieved 80–90%. ECGs that were printed, scanned, and redigitised, our tool achieved 96% correlation with the original signals. We have developed and validated a fully-automated, user-friendly, online ECG digitisation tool. Unlike other available tools, this does not require any manual segmentation of ECG signals. Our tool can facilitate the rapid and automated digitisation of large repositories of paper ECGs to allow them to be used for deep learning projects. [ABSTRACT FROM AUTHOR]

A cross-sectional study based on the community was conducted to explore the association between sleep status and LUTS among middle and old-aged men. Male residents in Zhengzhou aged 40 or older were recruited. Participants received the Pittsburgh Sleep Quality Index questionnaire and the International Prostate Symptom Score questionnaire to evaluate sleep status and the severity of lower urinary tract symptoms (LUTS), respectively. Logistic regression analyses and linear regression analyses were performed to evaluate the relationship between sleep quality and sleep duration and LUTS. A total of 5785 participants were enrolled. Multivariable analyses showed a positive relationship between sleep quality and LUTS (β 0.716, 95% CI 0.647–0.784), and poor sleepers were significantly associated with moderate or severe LUTS (OR 2.486, 95% CI 2.095–2.950). U-shaped dose–response relationship revealed that sleeping less than 5.8 h/day and more than 7.9 h/day was related to moderate or severe LUTS and more than 7.9 h/day of sleep duration was associated with poor voiding and storage symptoms (P for nonlinearity < 0.001). Similar relationship was observed between sleep status and nocturia. It showed a significantly positive relationship between sleep quality and LUTS. U-shaped dose–response relationships between sleep duration and LUTS were observed. [ABSTRACT FROM AUTHOR]

Artificial neural network (ANN) is the main tool to dig data and was inspired by the human brain and nervous system. Several studies clarified its application in medicine. However, none has applied ANN to predict the efficacy of folic acid treatment to Hyperhomocysteinemia (HHcy). The efficacy has been proved to associate with both genetic and environmental factors while previous studies just focused on the latter one. The explained variance genetic risk score (EV-GRS) had better power and could represent the effect of genetic architectures. Our aim was to add EV-GRS into environmental factors to establish ANN to predict the efficacy of folic acid therapy to HHcy. We performed the prospective cohort research enrolling 638 HHcy patients. The multilayer perception algorithm was applied to construct ANN. To evaluate the effect of ANN, we also established logistic regression (LR) model to compare with ANN. According to our results, EV-GRS was statistically associated with the efficacy no matter analyzed as a continuous variable (OR = 3.301, 95%CI 1.954–5.576, P < 0.001) or category variable (OR = 3.870, 95%CI 2.092–7.159, P < 0.001). In our ANN model, the accuracy was 84.78%, the Youden's index was 0.7073 and the AUC was 0.938. These indexes above indicated higher power. When compared with LR, the AUC, accuracy, and Youden's index of the ANN model (84.78%, 0.938, 0.7073) were all slightly higher than the LR model (83.33% 0.910, 0.6687). Therefore, clinical application of the ANN model may be able to better predict the folic acid efficacy to HHcy than the traditional LR model. When testing two models in the validation set, we got the same conclusion. This study appears to be the first one to establish the ANN model which added EV-GRS into environmental factors to predict the efficacy of folic acid to HHcy. This model would be able to offer clinicians a new method to make decisions and individual therapeutic plans. [ABSTRACT FROM AUTHOR]

Ag sinter joint is gaining greater attention for use with wide-bandgap die attach modules in high-temperature applications. In this study, we have investigated both the necking growth of Ag grain and the mechanical properties of sintered Ag with different Ag particle shapes in a low-temperature, pressure-less sintering process in air and N2. The necking growth ratio of Ag grains plays an important role in the mechanical properties of sintered Ag. Large necking growth corresponds to stronger tensile fracture strength of sintered Ag paste and stronger die shear strength. Self-generated Ag nanoparticles (AgNPs) with of a size less than 10 nm were observed under an air atmosphere, an atmosphere that produces larger necking growth than does that of N2. In addition, self-generated AgNPs were observed in sintered Ag flake particles specimens. Ag flake particles also lead to larger necking growth than that of sintered Ag spherical particle specimens. Self-generated AgNPs accumulated together due to their high surface energy, thus improving the necking growth of Ag grain and leading to a stronger fracture strength. In addition, it was revealed that both the oxygen element and the grain residual strain of initial Ag particles play important roles in the AgNPs self-generation during the sintering process. This study enhances our understanding of the sintering mechanism of Ag particles and the relationship between grain necking growth and its mechanical properties. [ABSTRACT FROM AUTHOR]

With the number of processor cores increasing in chip multiprocessors, the network on chip becomes a reliable structure with its perfect parallel communication performance. The traditional static router suffers a bad performance because of low buffer utilization for application-oriented network on chip. In this paper, a dynamic router which can take into account the unbalanced traffic for application-oriented network on chip is proposed. The router combines the inter-port and intra-port buffer allocation mechanism, which makes efficient use of buffer resources and avoids the head of line blocking. Furthermore, the proposed router can solve the problem of unbalanced load effectively between different ports on the same router. Simulation results show that the on-chip routers balance traffic between ports and increase the buffer utilization by 21.8%, thus optimizing delay and throughput performance for application-oriented network on chip. [ABSTRACT FROM AUTHOR]

An amendment to this paper has been published and can be accessed via a link at the top of the paper. [ABSTRACT FROM AUTHOR]

Gap junction-mediated cell-cell interactions are highly conserved and play essential roles in cell survival, proliferation, differentiation and patterning. We report that Connexin 32 (Cx32)-mediated gap junctional intercellular communication (GJIC) is necessary for human embryonic stem cell-derived hepatocytes (hESC-Heps) during step-wise hepatic lineage restriction and maturation. Vitamin K2, previously shown to promote Cx32 expression in mature hepatocytes, up-regulated Cx32 expression and GJIC activation during hepatic differentiation and maturation, resulting in significant increases of hepatic markers expression and hepatocyte functions. In contrast, negative Cx32 regulator 2-aminoethoxydiphenyl borate blocked hESC-to-hepatocyte maturation and muted hepatocyte functions through disruption of GJIC activities. Dynamic gap junction organization and internalization are phosphorylation-dependent and the p38 mitogen-activated protein kinases pathway (MAPK) can negatively regulate Cxs through phosphorylation-dependent degradation of Cxs. We found that p38 MAPK inhibitor SB203580 improved maturation of hESC-Heps correlating with up-regulation of Cx32; by contrast, the p38 MAPK activator, anisomycin, blocked hESC-Heps maturation correlating with down-regulation of Cx32. These results suggested that Cx32 is essential for cell-cell interactions that facilitate driving hESCs through hepatic-lineage maturation. Regulators of both Cx32 and other members of its pathways maybe used as a promising approach on regulating hepatic lineage restriction of pluripotent stem cells and optimizing their functional maturation. [ABSTRACT FROM AUTHOR]