Search

Copyright of Journal of Microscopy is the property of Wiley-Blackwell 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.)

To effectively detect and identify the structural faults of the rotating machinery, a new kind of symptom parameters of the vibration signal measured in multiple directions is proposed (hereinafter referred to as the structural feature symptom parameter SFSP), the method to extract anomaly signals based on the multi-band filter, the method to enhance the sensitivity of the symptom parameters using the least square mapping, and the automatic fault diagnosis system established for the structural faults of the rotating machinery in combination with multivariate analysis (principal component analysis) method. In addition, the validity of the diagnostic method was also verified with reference to the experimental data measured on the rotating machine in different states of structural faults. [ABSTRACT FROM AUTHOR]

Automatic modulation classification (AMC) is a promising technology to identify the modulation mode of the received signal in drone communication systems. Recently, benefiting from the outstanding classification performance of deep learning (DL), various deep neural networks (DNNs) have been introduced into AMC methods. Most current AMC methods are based on a local framework (LocalAMC) where there is only one device, or a centralized framework (CentAMC) where multiple local devices (LDs) upload their data to only one central server (CS). LocalAMC may not achieve ideal results due to insufficient data and finite computational power. CentAMC carries a significant risk of privacy leakage and the final data for training model in CS are quite massive. In this paper, we propose a practical and light AMC method based on decentralized learning with residual network (ResNet) in drone communication systems. Simulation results show that the ResNet-based decentralized AMC (DecentAMC) method achieves similar classification performance to CentAMC while improving training efficiency and protecting data privacy.

Automatic modulation classification (AMC) is a signal processing technology used to identify the modulation type of unknown signals without prior information such as modulation parameters for drone communications. In recent years, deep learning (DL) has been widely used in AMC methods due to its powerful feature extraction ability. The significant performance of DL-based AMC methods is highly dependent on large amount of data. However, with the increasingly complex signal environment and the emergence of new signals, several recognition tasks have difficulty obtaining sufficient high-quality signals. To address this problem, we propose an AMC method based on a deep residual neural network with masked modeling (DRMM). Specifically, masked modeling is adopted to improve the performance of a deep neural network with limited signal samples. Both complex-valued and real-valued residual neural networks (ResNet) play an important role in extracting signal features for identification. Several typical experiments are conducted to evaluate our proposed DRMM-based AMC method on the RadioML 2016.10A dataset and a simulated dataset, and comparison experiments with existing AMC methods are also conducted. The simulation results illustrate that our proposed DRMM-based AMC method achieves better performance in the case of limited signal samples with low signal-to-noise ratio (SNR) than other existing methods.

【Objective】 Aqueduct is a watercourse to carry away water from its source to distributing points, and its operation and safety are affected by many factors. The purpose of this paper is to analyze the response of water dynamics in the aqueduct to seismic load in an attempt to help aqueduct design under different scenarios. 【Method】 The analysis was based on numerical simulations. The seismic load was characterized by loading mode and loading frequency. The three-dimensional free-surface water flow in the aqueduct under different flow rate as affected by different combinations of loading characteristic factors was simulated using the FLOW-3D software, from which we calculated the force imposed by the flowing water to the aqueduct. 【Result】 The maximum vertical force induced by the seismic load was at the edge wall of the aqueduct proximal to the free water surface, which was more than ten-fold that at the bottom and in the midst of the aqueduct. When the seismic load was imposed in a single direction, the force induced by it to the aqueduct was 2.5% higher in the horizontal direction than in the vertical direction. With the frequency of seismic load increasing, the intensity of water body movement increased first and then declined when the frequency exceeded a threshold value. 【Conclusion】 Designing aqueducts potentially subject to seismic load should consider the strength of the side wall near the free water surface. While an increase in water flow rate reduces the force imposed by the water body to the side-walls under seismic load, it could risk overflow.

【Background】 In order to alleviate the shortage of water resources, China has established many water transfer projects. Due to its long water transfer distance, large water delivery volume, and numerous water passing buildings along the line, the control process is very complicated. Unsteady flow will inevitably appear in the channel operation scheduling process, and the Saint-Venant equations are an important way to describe and solve the unsteady flow. 【Objective】 With the construction of large-scale water transfer projects and the complexity of the operation scheduling and control process, the traditional method of solving the sparse matrix of the original Saint-Venant equations has been unable to meet the requirements of calculation volume and calculation speed. In order to find efficient and stable algorithms for solving large and sparse linear equations to improve the speed of solving Saint-Venant equations. 【Method】 In this paper, four algorithms for solving Saint-Venant equations based on the four-point eccentric scheme are summarized and improved, and the calculation efficiency of different algorithms is compared through simulation experiments. 【Result】 From the simulation results, it can be seen that when the number of calculation sections is small (less than 500), the calculation time of all methods is basically the same; when the number of calculation sections is large (more than 500), the speed of the four algorithms is improved compared with the traditional algorithm.When the number of cross sections is 1520, the calculation speed of the four algorithms is 4 times that of the traditional algorithm. When the number of cross sections is 3040, the calculation speed is more than 10 times. 【Conclusion】 The improved GE and PM algorithms have faster calculation speed and higher calculation efficiency when the number of cross sections is larger. The results of this paper have reference value for improving the calculation speed of large channel non-constant flow simulation. The method can be applied to canal system automatic control technology such as MPC control, LQR control point, etc., to improve the running speed of simulation program.

The Industrial Internet of Things (IIoT) refers to the application of the IoT in the industrial field. The development of fifth-generation (5G) communication technology has accelerated the world’s entry into the era of the industrial revolution and has also promoted the overall optimization of the IIoT. In the IIoT environment, challenges such as complex operating conditions and diverse data transmission have become increasingly prominent. Therefore, studying how to collect and process a large amount of real-time data from various devices in a timely, efficient, and reasonable manner is a significant problem. To address these issues, we propose a three-level networking model based on distributed sensor self-networking and cloud server platforms for networking. This model can collect monitoring data for a variety of industrial scenarios that require data collection. It enables the processing and storage of key information in a timely manner, reduces data transmission and storage costs, and improves data transmission reliability and efficiency. Additionally, we have designed a feature fusion network to further enhance the amount of feature information and improve the accuracy of industrial data recognition. The system also includes data preprocessing and data visualization capabilities. Finally, we discuss how to further preprocess and visualize the collected dataset and provide a specific algorithm analysis process using a large manipulator dataset as an example.

【Background】 Open channels have been widely used to transfer water in many projects. Automatically controlling them is a prerequisite in realizing intelligent management, optimizing water distribution, and achieving flexible water supply in emergency. Traditional automatic control systems normally use water level, flow rate, or storage pool as objective to control the channel operation based on some logic analysis. Existing control operation modes include keeping the upstream and downstream water levels constant, equal-volume method and control-volume method, with the control-volume method being most flexible and able to quickly respond to any changes. The disadvantage of the control-volume method is that it only uses the volume of a single channel section as its control objective. Therefore, its storage capacity is limited, and it cannot complement adjacent channels. 【Objective】 The objective of this paper is to resolve the inferiorities of the control-volume method, including limited storage capacity, inability to adjust the changes in the storage capacity between adjacent channels when using the storage capacity of a single section as the control objective. 【Method】 The proposed method was based on the balanced operation mode of multi-channel pool storage capacity with equal downstream water depth. We established a robust and easy-to-implement model to calculate the gate target flow based on the change in storage capacity, which not only keeps the total storage capacity of the channel unchanged, but also makes the downstream target water depth in each channel pool consistent. As a validation and verification, we applied the model to Jiping main canal on the east route of the south-north water transfer project. 【Result】 In the multi-canal pool storage balance mode, multiplying the storage difference by a weight coefficient can effectively reduce the gate flow overshoot and shorten the time it takes the system to stabilize. Under normal operating conditions, the regulation pressure on the head reservoir in the canal calculated by the proposed method was lower than that calculated by the conventional downstream constant water level control mode. Overall, the proposed method improved all dimensionless performance indexes, with the system stabilization time reduced by up to 10 hours. Considering that the difference in pool lengths in a canal could be more than 10 times, it is found that when there are large flow changes, adding a control gate to reduce the length of a single pool can further reduce the overall stabilization time and improve all dimensionless performance indexes. 【Conclusion】 Multi-channel pool storage control works better than the single channel pool storage control system, and the method we proposed lays a foundation for its further study and improvement.

In a time division duplex (TDD) based massive multiple-input multiple-output (MIMO) system, a base station (BS) is required to obtain accurate estimation of channel state information (CSI) for a user terminal (UT). Because of the time-varying nature of the channel, the length of pilot signals is limited and the number of orthogonal pilot signals is finite. Hence, the same pilot signals are required to be reused in neighboring cells and thus its channel estimation performance is deteriorated by pilot contamination from the neighboring cells. The minimum mean square error (MMSE) channel estimation can be used to reduce the influence of pilot contamination. However, it needs to know the covariance matrix of channels for all the UTs, which is unknown to the BS in practice. In this paper, we propose two methods of deep learning aided channel estimation to reduce the influence of pilot contamination. One method uses a neural network consisting of fully connected layers, while the other method uses a convolutional neural network (CNN). The neural network, particularly the CNN, plays a role in extracting features of the spatial information from the contaminated signals. The former method is better in terms of the training speed, however, the latter one can estimate the channel more accurately. We evaluate the proposed methods under two scenarios, i.e., perfect timing synchronization and imperfect one. Simulation results confirm that the proposed methods are better than the LS and covariance estimation methods via normalized mean square error (NMSE). In addition, we also investigate the impact of channel aging, and show that including some expected data into training datasets can avoid the great degradation of estimation quality.

Objective The methionine synthase reductase ( MTRR ) gene encodes the MTRR enzyme involved in the metabolic pathway of homocysteine. Several studies investigated the effect of the MTRR rs1532268 gene polymorphism on the risk of gastric cancer (GC), but the results have been inconsistent. Methods We performed a comprehensive and systematic search of PubMed, Google Scholar, MEDLINE, Science Direct, Scopus, CNKI, and Web of Science. Five studies were included in this meta-analysis to determine whether MTRR rs1532268 polymorphism contributes to the risk of GC. Results Pooled data indicated that the MTRR rs1532268 polymorphism significantly increased GC risk under the allele comparison model (odds ratio [OR] = 1.14, 95% confidence interval [CI] = 1.01–1.29) and dominant model (OR = 1.14, 95% CI = 1.00–1.30). In the analysis stratified by ethnicity, no relationship was found in Whites or Asians. Conclusion Our meta-analysis suggests a positive correlation between MTRR rs1532268 polymorphism and GC development.

With the explosion of global data,centralized cloud computing cannot provide low-latency,high-efficiency video surveillance services.A distributed edge computing model was proposed,which directly processed video data at the edge node to reduce the transmission pressure of the network,eased the computational burden of the central cloud server,and reduced the processing delay of the video surveillance system.Combined with the federated learning algorithm,a lightweight neural network was used,which trained in different scenarios and deployed on edge devices with limited computing power.Experimental results show that,compared with the general neural network model,the detection accuracy of the proposed method is improved by 18%,and the model training time is reduced.

Due to the fact that in dynamic wide-area optical backbone network the accuracies of the existing prediction methods were insufficient,a novel prediction method on quality of transmission (QoT) of optical channel was proposed based on ensemble learning theory.Firstly,under the framework of stacked ensemble learning,a base-learner including five multilayer perceptron (MLP) model was built,which could achieve homomorphic ensemble learning of sample data through parallel combination.Subsequently,the new training set fused from the predicted results of the preceding base-learner was used to training the meta-learner composed of a single MLP.The simulation results show that compared with the used deep neural network,the proposed method can obtain a more excellent nonlinear approximation in the scenarios of the single-channel and multi-channels,and the prediction accuracies have the improvements of 1.93% and 3.82% respectively.

Soil acidification and boron (B) starvation are two dominant abiotic stress factors impacting citrus production in the red soil region of southern China. To evaluate the combined effects of low pH and B deficiency on plant growth, gas exchange parameters, and the concentrations of B and other mineral nutrients, ‘HB’ pummelo seedlings were treated under B deficiency (0 μM H3BO3) or adequate B (23 μM H3BO3) conditions at various low pH levels (4.0, 5.0, and 6.0). The seedlings were grown with modified half-strength Hoagland’s solution under greenhouse conditions for 12 weeks. Plant biomass, leaf area, seedling height, and root traits were remarkably inhibited by low pH and B deficiency stresses, and these parameters were extremely reduced with the decrease in pH levels. After 12 weeks of treatment, typical stress symptoms associated with B deficiency in citrus leaf were observed, with more severe symptoms observed at pH 4.0 and 5.0 than at pH 6.0. Leaf gas exchange parameter measurements showed that leaf photosynthesis was significantly inhibited under both low pH and B-deficient conditions. Notably, the lower the pH level, the greater the inhibition under both normal and deficient B conditions. Further investigations of the mineral nutrient concentrations showed that under both low pH and B deficiency, the concentrations of B and other mineral nutrients were influenced remarkably, particularly at pH 4.0 and 5.0. The physiological and nutritional results of the ‘HB’ pummelo seedlings indicated that low pH can exacerbate the effects of B deficiency to a certain extent.

In this paper, we propose a transfer learning-based dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) method for classifying fibroadenoma and invasive ductal carcinoma (IDC) in breast tumors. A total of 207 breast tumors from patients were collected and identified by pathologic diagnosis within 15 days after enhanced DCE-MRI examination; 119 patients (average age 50.52±10.33 years) had fibroadenomas, and 88 patients (average age 42.20±10.10 years) had IDCs. Two lesion-level models were built based on the InceptionV3 and VGG19 models, which were pretrained with the ImageNet dataset. The effects of different depths of transfer learning were examined. The network's performance was evaluated through five-fold cross validation. In the lesion-level models, the model based on Inception V3 obtained better results (area under the receiver operating characteristic curve (AUC) = 0.89) when the weights were frozen before layer-276. The other model based on VGG19 obtained better results (AUC = 0.87) when the weights were frozen before layer-13. Compared with the image-level models, both lesion-level models displayed better discrimination (accuracy increased by 13% and 14%) based on the VGG19 and Inception V3 models, respectively. Our research confirms that transfer learning can be utilized to classify fibroadenomas and IDCs in DCE-MRI images. Different depths of transfer learning result in different performances, and our proposed lesion-level model notably improves the classification accuracy.

Automatic modulation recognition (AMR) plays an important role in various communications systems. It has the ability of adaptive modulation and can adapt to various complex environments. Automatic modulation recognition is also widely used in orthogonal frequency division multiplexing (OFDM) systems. However, because the recognition accuracy of traditional methods to extract the features of OFDM signals is very limited. In order to solve these problems, many deep learning based AMR methods have been proposed to improve the recognition performance. However, most of these AMR methods neglect the harmful effect by carrier phase offset (PO) which often appears in realistic communications systems. Hence it is required to consider the PO effect for designing the OFDM system. Unlike conventional methods, we propose a convolutional neural network (CNN) based AMR method for considering PO in the OFDM system. The proposed method is used to eliminate the PO to achieve the high classification accuracy. Experiment results are provided to confirm the proposed method when comparing to conventional methods.

The deep convolutional neural network (DeCNN) is considered one of promising techniques for classifying the high-spatial-resolution remote sensing (HSRRS) scenes, due to its powerful feature extraction capabilities. It is well-known that huge high-quality labeled datasets are required for achieving the better classification performances and preventing overfitting, during the training DeCNN model process. However, the lack of high-quality datasets limits the applications of DeCNN. In order to solve this problem, in this article, we propose a HSRRS image scene classification method using transfer learning and the DeCNN (TL-DeCNN) model in a few shot HSRRS scene samples. Specifically, three typical DeCNNs of VGG19, ResNet50, and InceptionV3, trained on the ImageNet2015, the weights of their convolutional layer for that of the TL-DeCNN are transferred, respectively. Then, TL-DeCNN just needs to fine-tune its classification module on the few shot HSRRS scene samples in a few epochs. Experimental results indicate that our proposed TL-DeCNN method provides absolute dominance results without overfitting, when compared with the VGG19, ResNet50, and InceptionV3, directly trained on the few shot samples.

Alcohol use disorders (AUD) is characterized by persistent or intermittent alcohol cravings and compulsive drinking. The functional changes in the central nervous system (CNS) after alcohol consumption are alcohol-associated cognitive impairment and mood disorders, which are major health issues reported in AUDs. Studies have shown that transferring the intestinal microbiota from AUDs patients to germ-free animals causes learning and memory dysfunction, depression and anxiety-like behavior, indicating the vital role of intestinal microbiota in development of neuropsychiatric disorders in AUD. Intestinal flora composition of AUD patients are significantly different from normal people, suggesting that intestinal flora imbalance orchestrate the development of neuropsychiatric disorders in AUD. Studies suggests that gut microbiome links bidirectional signaling network of the enteric nervous system (ENS) to central nervous system (CNS), forming gut-microbe-brain axis (brain-gut axis). In this review, we discussed pathogenesis and possible treatment of AUD-induced cognitive deficits, anxiety, and depression disorders. Further, we described the mechanism of intestinal flora imbalance and dysfunction of hippocampus-amygdala-frontal cortex (gut-limbic circuit system dysfunction). Therefore, we postulate therapeutic interventions of gut-brain axis as novel strategies for treatment of AUD-induced neuropsychiatric disorders.

Drone-aided ubiquitous applications play important roles in our daily lives. Accurate recognition of drones is required in aviation management due to their potential risks and disasters. Radiofrequency (RF) fingerprinting-based recognition technology based on deep learning (DL) is considered an effective approach to extracting hidden abstract features from the RF data of drones. Existing deep learning-based methods are either high computational burdens or have low accuracy. In this paper, we propose a deep complex-valued convolutional neural network (DC-CNN) method based on RF fingerprinting for recognizing different drones. Compared with existing recognition methods, the DC-CNN method has a high recognition accuracy, fast running time, and small network complexity. Nine algorithm models and two datasets are used to represent the superior performance of our system. Experimental results show that our proposed DC-CNN can achieve recognition accuracies of 99.5% and 74.1%, respectively, on four and eight classes of RF drone datasets.

This study was conducted to determine the effect of dietary sanguinarine supplementation on serum metabolites in growing pigs. A total of 240 (Yorkshire '[Duroc' Landrace]) pigs at 6 wk of age were randomly allotted to 3 dietary treatments, with 8 replicates per treatment and 10 piglets per replicate. The dietary treatments were as follows: the basal diet (BD) control group, the sanguinarine group (BD + 0.04 g/kg), and the antibiotic group (BD + 0.2 g/kg colistin). On d 7 and 14 after initiation of the treatment, BW, ADG, ADFI, feed efficiency (F/G), and serum levels of amino acids were measured. Dietary sanguinarine and antibiotic supplementation increased ADG, ADFI, and BW compared with the control group (P < 0.05). The data showed that the serum contents of amino acids, such as Gly, Ile, Lys, Met, Arg, Ala, and Thr, increased (P < 0.05) in the sanguinarine group compared with those in the colistin group and/or control group. These findings indicate a beneficial effect of dietary sanguinarine supplementation on the serum concentrations of some essential and nonessential amino acids, which improves growth performance of growing pigs. [ABSTRACT FROM AUTHOR]