Nonfragile Impulsive State Estimation for Complex Networks With Markovian Switching Topologies Subject to Limited Bit Rate Constraints.

In this article, we consider the impulsive estimation problem for a specific category of discrete-time complex networks (CNs) characterized by Markovian switching topologies. The measurement outputs of the underlying CNs, transmitted to the observer over wireless networks, are subject to bit rate constraints. To effectively reduce the estimation error and enhance estimation performance, a mode-dependent impulsive observer is proposed that employs the impulse mechanism. The application of stochastic analysis techniques leads to the derivation of a sufficient condition for ensuring the mean-square boundedness of the estimation error dynamics. The upper bound of the error is then analyzed by iteratively exploring the Lyapunov relation at both impulsive and non-impulsive instants. Moreover, an optimization algorithm is presented for handling the bit rate allocation, which is coupled with the design of desired observer gains using the linear matrix inequality (LMI) approach. Within this theoretical framework, the relationship between the mean-square estimation performance and the bit rate allocation protocol is further elucidated. Finally, a simulation example is provided to demonstrate the validity and effectiveness of the proposed impulsive estimation approach.

Unscented-Kalman-Filter-Based Remote State Estimation for Complex Networks With Quantized Measurements and Amplify-and-Forward Relays.

In this article, the remote estimation problem is addressed for a class of discrete-time complex networks under the influence of probabilistic quantization and amplify-and-forward (AF) relays. The underlying complex network model, which is inherently nonlinear and stochastic, is affected by additive process and measurement noises. Owing to the limited bandwidth of the transmission channel, the measurement outputs are quantized by a probabilistic quantizer prior to transmission. To enhance the signal quality over long-distance transmissions, the quantized measurements are sent to AF relays and subsequently forwarded to the estimator. Utilizing the unscented Kalman filter approach, a novel state estimator is designed to minimize an upper bound on the estimation error covariance. Moreover, sufficient conditions are derived to ensure that the estimation error is exponentially bounded in the mean-square sense. Lastly, the efficacy of the proposed scheme is illustrated through numerical simulations.

Versatile Framework for Drug-Target Interaction Prediction by Considering Domain-Specific Features.

Predicting drug-target interactions (DTIs) is one of the crucial tasks in drug discovery, but traditional wet-lab experiments are costly and time-consuming. Recently, deep learning has emerged as a promising tool for accelerating DTI prediction due to its powerful performance. However, the models trained on limited known DTI data struggle to generalize effectively to novel drug-target pairs. In this work, we propose a strategy to train an ensemble of models by capturing both domain-generic and domain-specific features (E-DIS) to learn diverse domain features and adapt them to out-of-distribution data. Multiple experts were trained on different domains to capture and align domain-specific information from various distributions without accessing any data from unseen domains. E-DIS provides a comprehensive representation of proteins and ligands by capturing diverse features. Experimental results on four benchmark data sets in both in-domain and cross-domain settings demonstrated that E-DIS significantly improved model performance and domain generalization compared to existing methods. Our approach presents a significant advancement in DTI prediction by combining domain-generic and domain-specific features, enhancing the generalization ability of the DTI prediction model.

S3T-Net: A novel electroencephalogram signals-oriented emotion recognition model.

In this paper, a novel skipping spatial-spectral-temporal network (S3T-Net) is developed to handle intra-individual differences in electroencephalogram (EEG) signals for accurate, robust, and generalized emotion recognition. In particular, aiming at the 4D features extracted from the raw EEG signals, a multi-branch architecture is proposed to learn spatial-spectral cross-domain representations, which benefits enhancing the model generalization ability. Time dependency among different spatial-spectral features is further captured via a bi-directional long-short term memory module, which employs an attention mechanism to integrate context information. Moreover, a skip-change unit is designed to add another auxiliary pathway for updating model parameters, which alleviates the vanishing gradient problem in complex spatial-temporal network. Evaluation results show that the proposed S3T-Net outperforms other advanced models in terms of the emotion recognition accuracy, which yields an performance improvement of 0.23% , 0.13%, and 0.43% as compared to the sub-optimal model in three test scenes, respectively. In addition, the effectiveness and superiority of the key components of S3T-Net are demonstrated from various experiments. As a reliable and competent emotion recognition model, the proposed S3T-Net contributes to the development of intelligent sentiment analysis in human-computer interaction (HCI) realm.

Numerical simulations on compression behaviors of the laminated shale based on the digital image technology and the discrete element method.

As an unconventional reservoir sedimentary rock, shale contains a series of layers and various microstructures that lead to complex mechanical properties, such as the anisotropy of stiffness and strength. This study is directed towards the anisotropy caused by the microstructures of the shale, employing the 2D particle flow code (PFC2D) to explore stiffness, strength, failure mode, and micro-crack evolution. More realistic microstructures and the calibration of microscopic parameters of the shale are reasonably considered through the computed tomography (CT) images and mineral analysis. The corresponding numerical simulation results are fully compared with the experimental results. In what follows, the sensitivity analysis is conducted on the key microscopic parameters and microstructure characteristics in numerical samples with laminated characteristics. The results show that the influence of microscopic parameters of the parallel bonding model on macroscopic parameters is related to the layering angle and the face type, and the microstructures and initial cracks of numerical samples can considerably affect the macroscopic mechanical behaviors of the laminated samples. Next, the effect of confining pressure on the mechanical properties of layered shale is also discussed based on the numerical results. These findings highlight the potential of this approach for applications in micro-scaled models and calibration of microscopic parameters to probe mechanical behaviors of the laminated rock.

State Estimation for Markovian Jump Neural Networks Under Probabilistic Bit Flips: Allocating Constrained Bit Rates.

In this article, the state estimation problem is studied for Markovian jump neural networks (MJNNs) within a digital network framework. The wireless communication channel with limited bandwidth is characterized by a constrained bit rate, and the occurrence of bit flips during wireless transmission is mathematically modeled. A transmission mechanism, which includes coding-decoding under bit-rate constraints and considers probabilistic bit flips, is introduced, providing a thorough characterization of the digital transmission process. A mode-dependent remote estimator is designed, which is capable of effectively capturing the internal state of the neural network. Furthermore, a sufficient condition is proposed to ensure the estimation error to remain bounded under challenging network conditions. Within this theoretical framework, the relationship between the neural network's estimation performance and the bit rate is explored. Finally, a simulation example is provided to validate the theoretical findings.

Choice Function-Based Hyper-Heuristics for Causal Discovery under Linear Structural Equation Models.

Causal discovery is central to human cognition, and learning directed acyclic graphs (DAGs) is its foundation. Recently, many nature-inspired meta-heuristic optimization algorithms have been proposed to serve as the basis for DAG learning. However, a single meta-heuristic algorithm requires specific domain knowledge and empirical parameter tuning and cannot guarantee good performance in all cases. Hyper-heuristics provide an alternative methodology to meta-heuristics, enabling multiple heuristic algorithms to be combined and optimized to achieve better generalization ability. In this paper, we propose a multi-population choice function hyper-heuristic to discover the causal relationships encoded in a DAG. This algorithm provides a reasonable solution for combining structural priors or possible expert knowledge with swarm intelligence. Under a linear structural equation model (SEM), we first identify the partial v-structures through partial correlation analysis as the structural priors of the next nature-inspired swarm intelligence approach. Then, through partial correlation analysis, we can limit the search space. Experimental results demonstrate the effectiveness of the proposed methods compared to the earlier state-of-the-art methods on six standard networks.

Visualization of Microcirculation at Acupoints in vivo of Alzheimer's Disease Animal Model with Photoacoustic Microscope: A Pilot Study.

Background: Alzheimer's disease may be effectively treated with acupoint-based acupuncture, which is acknowledged globally. However, more research is needed to understand the alterations in acupoints that occur throughout the illness and acupuncture treatment. Objective: This research investigated the differences in acupoint microcirculation between normal mice and AD animals in vivo. This research also examined how acupuncture affected AD animal models and acupoint microcirculation. Methods: 6-month-old SAMP8 mice were divided into two groups: the AD group and the acupuncture group. Additionally, SAMR1 mice of the same month were included as the normal group. The study involved subjecting a group of mice to 28 consecutive days of acupuncture at the ST36 (Zusanli) and CV12 (Zhongwan) acupoints. Following this treatment, the Morris water maze test was conducted to assess the mice's learning and memory abilities; the acoustic-resolution photoacoustic microscope (AR-PAM) imaging system was utilized to observe the microcirculation in CV12 acupoint region and head-specific region of each group of mice. Results: In comparison to the control group, the mice in the AD group exhibited a considerable decline in their learning and memory capabilities (p < 0.01). In comparison to the control group, the vascular in the CV12 region and head-specific region in mice from the AD group exhibited a considerable reduction in length, distance, and diameter r (p < 0.01). The implementation of acupuncture treatment had the potential to enhance the aforementioned condition to a certain degree. Conclusions: These findings offered tangible visual evidence that supports the ongoing investigation into the underlying mechanisms of acupuncture's therapeutic effects.

Set-Membership State Estimation for Multirate Nonlinear Complex Networks Under FlexRay Protocols: A Neural-Network-Based Approach.

In this article, the set-membership state estimation problem is investigated for a class of nonlinear complex networks under the FlexRay protocols (FRPs). In order to address practical engineering requirements, the multirate sampling is taken into account which allows for different sampling periods of the system state and the measurement. On the other hand, the FRP is deployed in the communication network from sensors to estimators in order to alleviate the communication burden. The underlying nonlinearity studied in this article is of a general nature, and an approach based on neural networks is employed to handle the nonlinearity. By utilizing the convex optimization technique, sufficient conditions are established in order to restrain the estimation errors within certain ellipsoidal constraints. Then, the estimator gains and the tuning scalars of the neural network are derived by solving several optimization problems. Finally, a practical simulation is conducted to verify the validity of the developed set-membership estimation scheme.

Secure State Estimation for Artificial Neural Networks With Unknown-But-Bounded Noises: A Homomorphic Encryption Scheme.

This article is concerned with the secure state estimation problem for artificial neural networks (ANNs) subject to unknown-but-bounded noises, where sensors and the remote estimator are connected via open and bandwidth-limited communication networks. Using the encoding-decoding mechanism (EDM) and the Paillier encryption technique, a novel homomorphic encryption scheme (HES) is introduced, which aims to ensure the secure transmission of measurement information within communication networks that are constrained by bandwidth. Under this encoding-decoding-based HES, the data being transmitted can be encrypted into ciphertexts comprising finite bits. The emphasis of this research is placed on the development of a secure set-membership state estimation algorithm, which allows for the computation of estimates using encrypted data without the need for decryption, thereby ensuring data security throughout the entire estimation process. Taking into account the unknown-but-bounded noises, the underlying ANN, and the adopted HES, sufficient conditions are determined for the existence of the desired ellipsoidal set. The related secure state estimator gains are then derived by addressing optimization problems using the Lagrange multiplier method. Lastly, an example is presented to verify the effectiveness of the proposed secure state estimation approach.

Multi-view graph pooling with coarsened graph disentanglement.

Multi-view graph pooling utilizes information from multiple perspectives to generate a coarsened graph, exhibiting superior performance in graph-level tasks. However, existing methods mainly focus on the types of multi-view information to improve graph pooling operations, lacking explicit control over the pooling process and theoretical analysis of the relationships between views. In this paper, we rethink the current paradigm of multi-view graph pooling from an information theory perspective, subsequently introducing GDMGP, an innovative method for multi-view graph pooling derived from the principles of graph disentanglement. This approach effectively simplifies the original graph into a more structured, disentangled coarsened graph, enhancing the clarity and utility of the graph representation. Our approach begins with the design of a novel view mapper that dynamically integrates the node and topology information of the original graph. This integration enhances its information sufficiency. Next, we introduce a view fusion mechanism based on conditional entropy to accurately regulate the task-relevant information in the views, aiming to minimize information loss in the pooling process. Finally, to further enhance the expressiveness of the coarsened graph, we disentangle the fused view into task-relevant and task-irrelevant subgraphs through mutual information minimization, retaining the task-relevant subgraph for downstream tasks. We theoretically demonstrate that the performance of the coarsened graph generated by our GDMGP is superior to that of any single input view. The effectiveness of GDMGP is further validated by experimental results on seven public datasets.

Periostin Promotes the Proliferation, Differentiation and Mineralization of Osteoblasts from Ovariectomized Rats.

Perimenopausal period causes a significant amount of bone loss, which results in primary osteoporosis (OP). The Periostin (Postn) may play important roles in the pathogenesis of OP after ovariectomized (OVX) rats. To identify the roles of Postn in the bone marrow mesenchymal stem cell derived osteoblasts (BMSC-OB) in OVX rats, we investigated the expression of Wnt/ß-catenin signaling pathways in BMSC-OB and the effects of Postn on bone formation by development of BMSC-OB cultures. Twenty-four female Sprague-Dawley rats at 6 months were randomized into 3 groups: sham-operated (SHAM) group, OVX group and OVX+Postn group. The rats were killed after 3 months, and their bilateral femora and tibiae were collected for BMSC-OB culture, Micro-CT Analysis, Bone Histomorphometric Measurement, Transmission Electron Microscopy and Immunohistochemistry Staining. The dose/time-dependent effects of Postn on the proliferation, differentiation and mineralization of BMSC-OB and the expression of osteoblastic markers were measured in in vitro experiments. We found increased Postn increased bone mass, promoted bone formation of trabeculae, Wnt signaling and the osteogenic activity in osteoblasts in sublesional femur. Postn have the function to enhance cell proliferation, differentiation and mineralization at a proper concentration and incubation time. Interestingly, in BMSC-OB from OVX rats treated with the different dose of Postn, the osteoblastic markers expression and Wnt/ß-catenin signaling pathways were significantly promoted. The direct effect of Postn may lead to inhibit excessive bone resorption and increase bone formation through the Wnt/ß-catenin signaling pathways after OVX. Postn may play a very important role in the pathogenesis of OP after OVX.

Enhanced multi-label cardiology diagnosis with channel-wise recurrent fusion.

The timely detection of abnormal electrocardiogram (ECG) signals is vital for preventing heart disease. However, traditional automated cardiology diagnostic methods have the limitation of being unable to simultaneously identify multiple diseases in a segment of ECG signals, and do not consider the potential correlations between the 12-lead ECG signals. To address these issues, this paper presents a novel network architecture, denoted as Branched Convolution and Channel Fusion Network (BCCF-Net), designed for the multi-label diagnosis of ECG cardiology to achieve simultaneous identification of multiple diseases. Among them, the BCCF-Net incorporates the Channel-wise Recurrent Fusion (CRF) network, which is designed to enhance the ability to explore potential correlation information between 12 leads. Furthermore, the utilization of the squeeze and excitation (SE) attention mechanism maximizes the potential of the convolutional neural network (CNN). In order to efficiently capture complex patterns in space and time across various scales, the multi branch convolution (MBC) module has been developed. Through extensive experiments on two public datasets with seven subtasks, the efficacy and robustness of the proposed ECG multi-label classification framework have been comprehensively evaluated. The results demonstrate the superior performance of the BCCF-Net compared to other state-of-the-art algorithms. The developed framework holds practical application in clinical settings, allowing for the refined diagnosis of cardiac arrhythmias through ECG signal analysis.

First- and Second-Order Sliding Mode Control Design for Networked 2-D Systems Under Round-Robin Protocol.

This article investigates the sliding mode control (SMC) problem for a class of uncertain 2-D systems described by the Roesser models with a bounded disturbance. In order to reduce the communication usage between the controller and the actuators, it is supposed that only one actuator node can gain the access to the network at each sampling time along horizontal or vertical direction, where a proper 2-D round-robin protocol is designed to periodically regulate the access token and a set of zero-order holders (ZOHs) is employed to keep the other actuator nodes unchanged until the next renewed signal arrives. Based on a novel 2-D common sliding function, a token-dependent 2-D SMC scheme with first-order sliding mode is appropriately constructed to cope with the impacts from the periodic scheduling signal and the ZOHs. Furthermore, a novel super-twisting-like 2-D SMC scheme with second-order sliding mode is designed to improve the robustness against the bounded disturbance. By resorting to token-dependent Lyapunov-like function, sufficient conditions are obtained to guarantee the ultimate boundedness of the horizontal and vertical states as well as the 2-D common sliding function. For acquiring the optimized gain matrices, two searching algorithms are formulated to solve two optimization problems arising from finding optimized control performance. Finally, two comparative examples are exploited to demonstrate the effectiveness and the advantageous of the proposed first- and second-order 2-D SMC design schemes under round-robin scheduling mechanism.

Efficient Predefined-Time Adaptive Neural Networks for Computing Time-Varying Tensor Moore-Penrose Inverse.

This article proposes predefined-time adaptive neural network (PTANN) and event-triggered PTANN (ET-PTANN) models to efficiently compute the time-varying tensor Moore-Penrose (MP) inverse. The PTANN model incorporates a novel adaptive parameter and activation function, enabling it to achieve strongly predefined-time convergence. Unlike traditional time-varying parameters that increase over time, the adaptive parameter is proportional to the error norm, thereby better allocating computational resources and improving efficiency. To further enhance efficiency, the ET-PTANN model combines an event trigger with the evolution formula, resulting in the adjustment of step size and reduction of computation frequency compared to the PTANN model. By conducting mathematical derivations, the article derives the upper bound of convergence time for the proposed neural network models and determines the minimum execution interval for the event trigger. A simulation example demonstrates that the PTANN and ET-PTANN models outperform other related neural network models in terms of computational efficiency and convergence rate. Finally, the practicality of the PTANN and ET-PTANN models is demonstrated through their application for mobile sound source localization.

Joint State and Unknown Input Estimation for a Class of Artificial Neural Networks With Sensor Resolution: An Encoding-Decoding Mechanism.

This article is concerned with the joint state and unknown input (SUI) estimation for a class of artificial neural networks (ANNs) with sensor resolution (SR) under the encoding-decoding mechanisms. The consideration of SR, which is an important specification of sensors in the real world, caters to engineering practice. Furthermore, the implementation of the encoding-decoding mechanism in the communication network aims to accommodate the limited bandwidth. The objective of this study is to propose a set-membership estimation algorithm that accurately estimates the state of the ANN without being influenced by the unknown input while accounting for the SR and the encoding-decoding mechanism. First, a sufficient condition is derived to ensure an ellipsoidal constraint on the estimation error. Then, by addressing an optimization problem, the design of the estimator gains is accomplished, and the minimal ellipsoidal constraint on the state estimation error is obtained. Finally, an example is provided to confirm the validity of the proposed joint SUI estimation scheme.

Distributed Proportional-Integral Fuzzy State Estimation Over Sensor Networks Under Energy-Constrained Denial-of-Service Attacks.

This article deals with the distributed proportional-integral state estimation problem for nonlinear systems over sensor networks (SNs), where a number of spatially distributed sensor nodes are utilized to collect the system information. The signal transmissions among different sensor nodes are realized via their individual channels subject to energy-constrained Denial-of-Service (EC-DoS) cyber-attacks launched by the adversaries whose aim is to block the nodewise communications. Such EC-DoS attacks are characterized by a sequence of attack starting time-instants and a sequence of attack durations. Based on the measurement outputs of each node, a novel distributed fuzzy proportional-integral estimator is proposed that reflects the topological information of the SNs. The estimation error dynamics is shown to be regulated by a switching system under certain assumptions on the frequency and the duration of the EC-DoS attacks. Then, by resorting to the average dwell-time method, a unified framework is established to analyze the dynamical behaviors of the resultant estimation error system, and sufficient conditions are obtained to guarantee the stability as well as the weighted H∞ performance of the estimation error dynamics. Finally, a numerical example is given to verify the effectiveness of the proposed estimation scheme.

A novel fixed-time error-monitoring neural network for solving dynamic quaternion-valued Sylvester equations.

This paper addresses the dynamic quaternion-valued Sylvester equation (DQSE) using the quaternion real representation and the neural network method. To transform the Sylvester equation in the quaternion field into an equivalent equation in the real field, three different real representation modes for the quaternion are adopted by considering the non-commutativity of quaternion multiplication. Based on the equivalent Sylvester equation in the real field, a novel recurrent neural network model with an integral design formula is proposed to solve the DQSE. The proposed model, referred to as the fixed-time error-monitoring neural network (FTEMNN), achieves fixed-time convergence through the action of a state-of-the-art nonlinear activation function. The fixed-time convergence of the FTEMNN model is theoretically analyzed. Two examples are presented to verify the performance of the FTEMNN model with a specific focus on fixed-time convergence. Furthermore, the chattering phenomenon of the FTEMNN model is discussed, and a saturation function scheme is designed. Finally, the practical value of the FTEMNN model is demonstrated through its application to image fusion denoising.

ELTS-Net: An enhanced liver tumor segmentation network with augmented receptive field and global contextual information.

The liver is one of the organs with the highest incidence rate in the human body, and late-stage liver cancer is basically incurable. Therefore, early diagnosis and lesion location of liver cancer are of important clinical value. This study proposes an enhanced network architecture ELTS-Net based on the 3D U-Net model, to address the limitations of conventional image segmentation methods and the underutilization of image spatial features by the 2D U-Net network structure. ELTS-Net expands upon the original network by incorporating dilated convolutions to increase the receptive field of the convolutional kernel. Additionally, an attention residual module, comprising an attention mechanism and residual connections, replaces the original convolutional module, serving as the primary components of the encoder and decoder. This design enables the network to capture contextual information globally in both channel and spatial dimensions. Furthermore, deep supervision modules are integrated between different levels of the decoder network, providing additional feedback from deeper intermediate layers. This constrains the network weights to the target regions and optimizing segmentation results. Evaluation on the LiTS2017 dataset shows improvements in evaluation metrics for liver and tumor segmentation tasks compared to the baseline 3D U-Net model, achieving 95.2% liver segmentation accuracy and 71.9% tumor segmentation accuracy, with accuracy improvements of 0.9% and 3.1% respectively. The experimental results validate the superior segmentation performance of ELTS-Net compared to other comparison models, offering valuable guidance for clinical diagnosis and treatment.

EEGProgress: A fast and lightweight progressive convolution architecture for EEG classification.

Because of the intricate topological structure and connection of the human brain, extracting deep spatial features from electroencephalograph (EEG) signals is a challenging and time-consuming task. The extraction of topological spatial information plays a crucial role in EEG classification, and the architecture of the spatial convolution greatly affects the performance and complexity of convolutional neural network (CNN) based EEG classification models. In this study, a progressive convolution CNN architecture named EEGProgress is proposed, aiming to efficiently extract the topological spatial information of EEG signals from multi-scale levels (electrode, brain region, hemisphere, global) with superior speed. To achieve this, the raw EEG data is permuted using the empirical topological permutation rule, integrating the EEG data with numerous topological properties. Subsequently, the spatial features are extracted by a progressive feature extractor including prior, electrode, region, and hemisphere convolution blocks, progressively extracting the deep spatial features with reduced parameters and speed. Finally, the comparison and ablation experiments under both cross-subject and within-subject scenarios are conducted on a public dataset to verify the performance of the proposed EEGProgress and the effectiveness of the topological permutation. The results demonstrate the superior feature extraction ability of the proposed EEGProgress, with an average increase of 4.02% compared to other CNN-based EEG classification models under both cross-subject and within-subject scenarios. Furthermore, with the obtained average testing time, FLOPs, and parameters, the proposed EEGProgress outperforms other comparison models in terms of model complexity.