The effective and timely diagnosis and treatment of ocular diseases are key to the rapid recovery of patients.Today,the mass disease that needs attention in this context is cataracts.Although deep learning has signifi...The effective and timely diagnosis and treatment of ocular diseases are key to the rapid recovery of patients.Today,the mass disease that needs attention in this context is cataracts.Although deep learning has significantly advanced the analysis of ocular disease images,there is a need for a probabilistic model to generate the distributions of potential outcomes and thusmake decisions related to uncertainty quantification.Therefore,this study implements a Bayesian Convolutional Neural Networks(BCNN)model for predicting cataracts by assigning probability values to the predictions.It prepares convolutional neural network(CNN)and BCNN models.The proposed BCNN model is CNN-based in which reparameterization is in the first and last layers of the CNN model.This study then trains them on a dataset of cataract images filtered from the ocular disease fundus images fromKaggle.The deep CNN model has an accuracy of 95%,while the BCNN model has an accuracy of 93.75% along with information on uncertainty estimation of cataracts and normal eye conditions.When compared with other methods,the proposed work reveals that it can be a promising solution for cataract prediction with uncertainty estimation.展开更多
For the diagnostics and health management of lithium-ion batteries,numerous models have been developed to understand their degradation characteristics.These models typically fall into two categories:data-driven models...For the diagnostics and health management of lithium-ion batteries,numerous models have been developed to understand their degradation characteristics.These models typically fall into two categories:data-driven models and physical models,each offering unique advantages but also facing limitations.Physics-informed neural networks(PINNs)provide a robust framework to integrate data-driven models with physical principles,ensuring consistency with underlying physics while enabling generalization across diverse operational conditions.This study introduces a PINN-based approach to reconstruct open circuit voltage(OCV)curves and estimate key ageing parameters at both the cell and electrode levels.These parameters include available capacity,electrode capacities,and lithium inventory capacity.The proposed method integrates OCV reconstruction models as functional components into convolutional neural networks(CNNs)and is validated using a public dataset.The results reveal that the estimated ageing parameters closely align with those obtained through offline OCV tests,with errors in reconstructed OCV curves remaining within 15 mV.This demonstrates the ability of the method to deliver fast and accurate degradation diagnostics at the electrode level,advancing the potential for precise and efficient battery health management.展开更多
The ability to accurately predict urban traffic flows is crucial for optimising city operations.Consequently,various methods for forecasting urban traffic have been developed,focusing on analysing historical data to u...The ability to accurately predict urban traffic flows is crucial for optimising city operations.Consequently,various methods for forecasting urban traffic have been developed,focusing on analysing historical data to understand complex mobility patterns.Deep learning techniques,such as graph neural networks(GNNs),are popular for their ability to capture spatio-temporal dependencies.However,these models often become overly complex due to the large number of hyper-parameters involved.In this study,we introduce Dynamic Multi-Graph Spatial-Temporal Graph Neural Ordinary Differential Equation Networks(DMST-GNODE),a framework based on ordinary differential equations(ODEs)that autonomously discovers effective spatial-temporal graph neural network(STGNN)architectures for traffic prediction tasks.The comparative analysis of DMST-GNODE and baseline models indicates that DMST-GNODE model demonstrates superior performance across multiple datasets,consistently achieving the lowest Root Mean Square Error(RMSE)and Mean Absolute Error(MAE)values,alongside the highest accuracy.On the BKK(Bangkok)dataset,it outperformed other models with an RMSE of 3.3165 and an accuracy of 0.9367 for a 20-min interval,maintaining this trend across 40 and 60 min.Similarly,on the PeMS08 dataset,DMST-GNODE achieved the best performance with an RMSE of 19.4863 and an accuracy of 0.9377 at 20 min,demonstrating its effectiveness over longer periods.The Los_Loop dataset results further emphasise this model’s advantage,with an RMSE of 3.3422 and an accuracy of 0.7643 at 20 min,consistently maintaining superiority across all time intervals.These numerical highlights indicate that DMST-GNODE not only outperforms baseline models but also achieves higher accuracy and lower errors across different time intervals and datasets.展开更多
The increasing popularity of the Internet and the widespread use of information technology have led to a rise in the number and sophistication of network attacks and security threats.Intrusion detection systems are cr...The increasing popularity of the Internet and the widespread use of information technology have led to a rise in the number and sophistication of network attacks and security threats.Intrusion detection systems are crucial to network security,playing a pivotal role in safeguarding networks from potential threats.However,in the context of an evolving landscape of sophisticated and elusive attacks,existing intrusion detection methodologies often overlook critical aspects such as changes in network topology over time and interactions between hosts.To address these issues,this paper proposes a real-time network intrusion detection method based on graph neural networks.The proposedmethod leverages the advantages of graph neural networks and employs a straightforward graph construction method to represent network traffic as dynamic graph-structured data.Additionally,a graph convolution operation with a multi-head attention mechanism is utilized to enhance the model’s ability to capture the intricate relationships within the graph structure comprehensively.Furthermore,it uses an integrated graph neural network to address dynamic graphs’structural and topological changes at different time points and the challenges of edge embedding in intrusion detection data.The edge classification problem is effectively transformed into node classification by employing a line graph data representation,which facilitates fine-grained intrusion detection tasks on dynamic graph node feature representations.The efficacy of the proposed method is evaluated using two commonly used intrusion detection datasets,UNSW-NB15 and NF-ToN-IoT-v2,and results are compared with previous studies in this field.The experimental results demonstrate that our proposed method achieves 99.3%and 99.96%accuracy on the two datasets,respectively,and outperforms the benchmark model in several evaluation metrics.展开更多
Foot reflexology is a non-invasive and safe complementary therapy that works by massaging the reflex zones of the feet and exerts systemic or whole-body regulation through meridian nerve conduction.This therapy is com...Foot reflexology is a non-invasive and safe complementary therapy that works by massaging the reflex zones of the feet and exerts systemic or whole-body regulation through meridian nerve conduction.This therapy is commonly used in the treatment of various conditions such as autism and Parkinson's disease.However,there is limited reporting on the use of foot reflexology therapy for infants with sensorineural hearing loss(SNHL).Currently,there is no definitive conclusion on how foot reflexology therapy can influence hearing.This editorial holds some guiding significance regarding this clinical issue.The aim is to present physiological evidence of how foot reflexology therapy can impact infants with SNHL,thereby enhancing clinician’s awareness of foot reflexology in treating infants with SNHL.展开更多
Neural organoids and confocal microscopy have the potential to play an important role in microconnectome research to understand neural patterns.We present PLayer,a plug-and-play embedded neural system,which demonstrat...Neural organoids and confocal microscopy have the potential to play an important role in microconnectome research to understand neural patterns.We present PLayer,a plug-and-play embedded neural system,which demonstrates the utilization of sparse confocal microscopy layers to interpolate continuous axial resolution.With an embedded system focused on neural network pruning,image scaling,and post-processing,PLayer achieves high-performance metrics with an average structural similarity index of 0.9217 and a peak signal-to-noise ratio of 27.75 dB,all within 20 s.This represents a significant time saving of 85.71%with simplified image processing.By harnessing statistical map estimation in interpolation and incorporating the Vision Transformer–based Restorer,PLayer ensures 2D layer consistency while mitigating heavy computational dependence.As such,PLayer can reconstruct 3D neural organoid confocal data continuously under limited computational power for the wide acceptance of fundamental connectomics and pattern-related research with embedded devices.展开更多
Numerical simulation is dominant in solving partial differential equations(PDEs),but balancing fine-grained grids with low computational costs is challenging.Recently,solving PDEs with neural networks(NNs)has gained i...Numerical simulation is dominant in solving partial differential equations(PDEs),but balancing fine-grained grids with low computational costs is challenging.Recently,solving PDEs with neural networks(NNs)has gained interest,yet cost-effectiveness and high accuracy remain a challenge.This work introduces a novel paradigm for solving PDEs,called multi-scale neural computing(MSNC),considering spectral bias of NNs and local approximation properties in the finite difference method(FDM).The MSNC decomposes the solution with a NN for efficient capture of global scale and the FDM for detailed description of local scale,aiming to balance costs and accuracy.Demonstrated advantages include higher accuracy(10 times for 1D PDEs,20 times for 2D PDEs)and lower costs(4 times for 1D PDEs,16 times for 2D PDEs)than the standard FDM.The MSNC also exhibits stable convergence and rigorous boundary condition satisfaction,showcasing the potential for hybrid of NN and numerical method.展开更多
The complexity and intricacy of the brain,which is composed of billions of neurons,pose significant challenges to its study.Understanding neural connections and communication at the single-cell level is crucial for un...The complexity and intricacy of the brain,which is composed of billions of neurons,pose significant challenges to its study.Understanding neural connections and communication at the single-cell level is crucial for unraveling the brain’s functions.This study presents a novel strategy that utilizes magnetic nanoparticles(MNPs)and magnetic fields to manipulate neurons,thereby creating customized small-scale neural circuits for studying neural connections.To establish the feasibility of this approach,the effects of MNPs on neurons were initially investigated,demonstrating their low toxicity.Subsequently,a micromagnet array(MMA)chip was employed to manipulate the neurons,facilitating their precise arrangement on the electrodes.Over several days,the neurons extended their axons and established connections with neighboring cells,forming small-scale circular neural circuits.These artificially engineered circuits offer a simplified and controlled environment for studying neural networks in contrast to naturally occurring biological networks.Furthermore,electrophysiological recordings were conducted to investigate the connections between the manipulated neurons.This study introduces a customized small-scale neural circuit platform with electrode-specific recording and stimulating capabilities,enabling the study of neuron-to-neuron interactions at the single-cell level.By leveraging MNPs and an MMA chip,this research offers a powerful tool for studying neural connections and advancing our understanding of the brain’s intricate workings.展开更多
Spiking Neural Network(SNN)inspired by the biological triggering mechanism of neurons to provide a novel solution for plant disease detection,offering enhanced performance and efficiency in contrast to Artificial Neur...Spiking Neural Network(SNN)inspired by the biological triggering mechanism of neurons to provide a novel solution for plant disease detection,offering enhanced performance and efficiency in contrast to Artificial Neural Networks(ANN).Unlike conventional ANNs,which process static images without fully capturing the inherent temporal dynamics,our approach represents the first implementation of SNNs tailored explicitly for agricultural disease classification,integrating an encoding method to convert static RGB plant images into temporally encoded spike trains.Additionally,while Bernoulli trials and standard deep learning architectures likeConvolutionalNeuralNetworks(CNNs)and Fully Connected Neural Networks(FCNNs)have been used extensively,our work is the first to integrate these trials within an SNN framework specifically for agricultural applications.This integration not only refines spike regulation and reduces computational overhead by 30%but also delivers superior accuracy(93.4%)in plant disease classification,marking a significant advancement in precision agriculture that has not been previously explored.Our approach uniquely transforms static plant leaf images into time-dependent representations,leveraging SNNs’intrinsic temporal processing capabilities.This approach aligns with the inherent ability of SNNs to capture dynamic,timedependent patterns,making them more suitable for detecting disease activations in plants than conventional ANNs that treat inputs as static entities.Unlike prior works,our hybrid encoding scheme dynamically adapts to pixel intensity variations(via threshold),enabling robust feature extraction under diverse agricultural conditions.The dual-stage preprocessing customizes the SNN’s behavior in two ways:the encoding threshold is derived from pixel distributions in diseased regions,and Bernoulli trials selectively reduce redundant spikes to ensure energy efficiency on low-power devices.We used a comprehensive dataset of 87,000 RGB images of plant leaves,which included 38 distinct classes of healthy and unhealthy leaves.To train and evaluate three distinct neural network architectures,DeepSNN,SimpleCNN,and SimpleFCNN,the dataset was rigorously preprocessed,including stochastic rotation,horizontal flip,resizing,and normalization.Moreover,by integrating Bernoulli trials to regulate spike generation,ourmethod focuses on extracting themost relevant featureswhile reducingcomputational overhead.Using a comprehensivedatasetof87,000RGB images across 38 classes,we rigorously preprocessed the data and evaluated three architectures:DeepSNN,SimpleCNN,and SimpleFCNN.The results demonstrate that DeepSNN outperforms the other models,achieving superior accuracy,efficient feature extraction,and robust spike management,thereby establishing the potential of SNNs for real-time,energy-efficient agricultural applications.展开更多
Neural architecture search(NAS)optimizes neural network architectures to align with specific data and objectives,thereby enabling the design of high-performance models without specialized expertise.However,a significa...Neural architecture search(NAS)optimizes neural network architectures to align with specific data and objectives,thereby enabling the design of high-performance models without specialized expertise.However,a significant limitation of NAS is that it requires extensive computational resources and time.Consequently,performing a comprehensive architectural search for each new dataset is inefficient.Given the continuous expansion of available datasets,there is an urgent need to predict the optimal architecture for the previously unknown datasets.This study proposes a novel framework that generates architectures tailored to unknown datasets by mapping architectures that have demonstrated effectiveness on the existing dataset into a latent feature space.As NAS is inherently represented as graph structures,we employed an encoder-decoder transformation model based on variational graph auto-encoders to perform this latent feature mapping.The encoder-decoder transformation model demonstrates strong capability in extracting features from graph structures,making it particularly well-suited for mapping NAS architectures.By training variational graph auto-encoders on existing high-quality architectures,the proposed method constructs a latent space and facilitates the design of optimal architectures for diverse datasets.Furthermore,to effectively define similarity amongarchitectures,wepropose constructing the latent spaceby incorporatingbothdataset andtaskfeatures.Experimental results indicate that our approach significantly enhances search efficiency and outperforms conventional methods in terms of model performance.展开更多
Reverse design of highly GeO2-doped silica optical fibers with broadband and flat dispersion profiles is proposed using a neural network(NN) combined with a particle swarm optimization(PSO) algorithm.Firstly,the NN mo...Reverse design of highly GeO2-doped silica optical fibers with broadband and flat dispersion profiles is proposed using a neural network(NN) combined with a particle swarm optimization(PSO) algorithm.Firstly,the NN model designed to predict optical fiber dispersion is trained with an appropriate choice of hyperparameters,achieving a root mean square error(RMSE) of 9.47×10-7on the test dataset,with a determination coefficient(R2) of 0.999.Secondly,the NN is combined with the PSO algorithm for the inverse design of dispersion-flattened optical fibers.To expand the search space and avoid particles becoming trapped in local optimal solutions,the PSO algorithm incorporates adaptive inertia weight updating and a simulated annealing algorithm.Finally,by using a suitable fitness function,the designed fibers exhibit flat group velocity dispersion(GVD) profiles at 1 400—2 400 nm,where the GVD fluctuations and minimum absolute GVD values are below 18 ps·nm-1·km-1and 7 ps·nm-1·km-1,respectively.展开更多
Dear Editor,This letter presents a novel latent factorization model for high dimensional and incomplete (HDI) tensor, namely the neural Tucker factorization (Neu Tuc F), which is a generic neural network-based latent-...Dear Editor,This letter presents a novel latent factorization model for high dimensional and incomplete (HDI) tensor, namely the neural Tucker factorization (Neu Tuc F), which is a generic neural network-based latent-factorization-of-tensors model under the Tucker decomposition framework.展开更多
In this paper,a sparse graph neural network-aided(SGNN-aided)decoder is proposed for improving the decoding performance of polar codes under bursty interference.Firstly,a sparse factor graph is constructed using the e...In this paper,a sparse graph neural network-aided(SGNN-aided)decoder is proposed for improving the decoding performance of polar codes under bursty interference.Firstly,a sparse factor graph is constructed using the encoding characteristic to achieve high-throughput polar decoding.To further improve the decoding performance,a residual gated bipartite graph neural network is designed for updating embedding vectors of heterogeneous nodes based on a bidirectional message passing neural network.This framework exploits gated recurrent units and residual blocks to address the gradient disappearance in deep graph recurrent neural networks.Finally,predictions are generated by feeding the embedding vectors into a readout module.Simulation results show that the proposed decoder is more robust than the existing ones in the presence of bursty interference and exhibits high universality.展开更多
This paper presents the variational physics-informed neural network(VPINN)as an effective tool for static structural analyses.One key innovation includes the construction of the neural network solution as an admissibl...This paper presents the variational physics-informed neural network(VPINN)as an effective tool for static structural analyses.One key innovation includes the construction of the neural network solution as an admissible function of the boundary-value problem(BVP),which satisfies all geometrical boundary conditions.We then prove that the admissible neural network solution also satisfies natural boundary conditions,and therefore all boundary conditions,when the stationarity condition of the variational principle is met.Numerical examples are presented to show the advantages and effectiveness of the VPINN in comparison with the physics-informed neural network(PINN).Another contribution of the work is the introduction of Gaussian approximation of the Dirac delta function,which significantly enhances the ability of neural networks to handle singularities,as demonstrated by the examples with concentrated support conditions and loadings.It is hoped that these structural examples are so convincing that engineers would adopt the VPINN method in their structural design practice.展开更多
Grains are the most important food consumed globally,yet their yield can be severely impacted by pest infestations.Addressing this issue,scientists and researchers strive to enhance the yield-to-seed ratio through eff...Grains are the most important food consumed globally,yet their yield can be severely impacted by pest infestations.Addressing this issue,scientists and researchers strive to enhance the yield-to-seed ratio through effective pest detection methods.Traditional approaches often rely on preprocessed datasets,but there is a growing need for solutions that utilize real-time images of pests in their natural habitat.Our study introduces a novel twostep approach to tackle this challenge.Initially,raw images with complex backgrounds are captured.In the subsequent step,feature extraction is performed using both hand-crafted algorithms(Haralick,LBP,and Color Histogram)and modified deep-learning architectures.We propose two models for this purpose:PestNet-EF and PestNet-LF.PestNet-EF uses an early fusion technique to integrate handcrafted and deep learning features,followed by adaptive feature selection methods such as CFS and Recursive Feature Elimination(RFE).PestNet-LF utilizes a late fusion technique,incorporating three additional layers(fully connected,softmax,and classification)to enhance performance.These models were evaluated across 15 classes of pests,including five classes each for rice,corn,and wheat.The performance of our suggested algorithms was tested against the IP102 dataset.Simulation demonstrates that the Pestnet-EF model achieved an accuracy of 96%,and the PestNet-LF model with majority voting achieved the highest accuracy of 94%,while PestNet-LF with the average model attained an accuracy of 92%.Also,the proposed approach was compared with existing methods that rely on hand-crafted and transfer learning techniques,showcasing the effectiveness of our approach in real-time pest detection for improved agricultural yield.展开更多
Neural machine interface technology is a pioneering approach that aims to address the complex challenges of neurological dysfunctions and disabilities resulting from conditions such as congenital disorders,traumatic i...Neural machine interface technology is a pioneering approach that aims to address the complex challenges of neurological dysfunctions and disabilities resulting from conditions such as congenital disorders,traumatic injuries,and neurological diseases.Neural machine interface technology establishes direct connections with the brain or peripheral nervous system to restore impaired motor,sensory,and cognitive functions,significantly improving patients'quality of life.This review analyzes the chronological development and integration of various neural machine interface technologies,including regenerative peripheral nerve interfaces,targeted muscle and sensory reinnervation,agonist–antagonist myoneural interfaces,and brain–machine interfaces.Recent advancements in flexible electronics and bioengineering have led to the development of more biocompatible and highresolution electrodes,which enhance the performance and longevity of neural machine interface technology.However,significant challenges remain,such as signal interference,fibrous tissue encapsulation,and the need for precise anatomical localization and reconstruction.The integration of advanced signal processing algorithms,particularly those utilizing artificial intelligence and machine learning,has the potential to improve the accuracy and reliability of neural signal interpretation,which will make neural machine interface technologies more intuitive and effective.These technologies have broad,impactful clinical applications,ranging from motor restoration and sensory feedback in prosthetics to neurological disorder treatment and neurorehabilitation.This review suggests that multidisciplinary collaboration will play a critical role in advancing neural machine interface technologies by combining insights from biomedical engineering,clinical surgery,and neuroengineering to develop more sophisticated and reliable interfaces.By addressing existing limitations and exploring new technological frontiers,neural machine interface technologies have the potential to revolutionize neuroprosthetics and neurorehabilitation,promising enhanced mobility,independence,and quality of life for individuals with neurological impairments.By leveraging detailed anatomical knowledge and integrating cutting-edge neuroengineering principles,researchers and clinicians can push the boundaries of what is possible and create increasingly sophisticated and long-lasting prosthetic devices that provide sustained benefits for users.展开更多
Microglial activation that occurs rapidly after closed head injury may play important and complex roles in neuroinflammation-associated neuronal damage and repair.We previously reported that induced neural stem cells ...Microglial activation that occurs rapidly after closed head injury may play important and complex roles in neuroinflammation-associated neuronal damage and repair.We previously reported that induced neural stem cells can modulate the behavior of activated microglia via CXCL12/CXCR4 signaling,influencing their activation such that they can promote neurological recovery.However,the mechanism of CXCR4 upregulation in induced neural stem cells remains unclear.In this study,we found that nuclear factor-κB activation induced by closed head injury mouse serum in microglia promoted CXCL12 and tumor necrosis factor-αexpression but suppressed insulin-like growth factor-1 expression.However,recombinant complement receptor 2-conjugated Crry(CR2-Crry)reduced the effects of closed head injury mouse serum-induced nuclear factor-κB activation in microglia and the levels of activated microglia,CXCL12,and tumor necrosis factor-α.Additionally,we observed that,in response to stimulation(including stimulation by CXCL12 secreted by activated microglia),CXCR4 and Crry levels can be upregulated in induced neural stem cells via the interplay among CXCL12/CXCR4,Crry,and Akt signaling to modulate microglial activation.In agreement with these in vitro experimental results,we found that Akt activation enhanced the immunoregulatory effects of induced neural stem cell grafts on microglial activation,leading to the promotion of neurological recovery via insulin-like growth factor-1 secretion and the neuroprotective effects of induced neural stem cell grafts through CXCR4 and Crry upregulation in the injured cortices of closed head injury mice.Notably,these beneficial effects of Akt activation in induced neural stem cells were positively correlated with the therapeutic effects of induced neural stem cells on neuronal injury,cerebral edema,and neurological disorders post–closed head injury.In conclusion,our findings reveal that Akt activation may enhance the immunoregulatory effects of induced neural stem cells on microglial activation via upregulation of CXCR4 and Crry,thereby promoting induced neural stem cell–mediated improvement of neuronal injury,cerebral edema,and neurological disorders following closed head injury.展开更多
Human brain development is a complex process,and animal models often have significant limitations.To address this,researchers have developed pluripotent stem cell-derived three-dimensional structures,known as brain-li...Human brain development is a complex process,and animal models often have significant limitations.To address this,researchers have developed pluripotent stem cell-derived three-dimensional structures,known as brain-like organoids,to more accurately model early human brain development and disease.To enable more consistent and intuitive reproduction of early brain development,in this study,we incorporated forebrain organoid culture technology into the traditional unguided method of brain organoid culture.This involved embedding organoids in matrigel for only 7 days during the rapid expansion phase of the neural epithelium and then removing them from the matrigel for further cultivation,resulting in a new type of human brain organoid system.This cerebral organoid system replicated the temporospatial characteristics of early human brain development,including neuroepithelium derivation,neural progenitor cell production and maintenance,neuron differentiation and migration,and cortical layer patterning and formation,providing more consistent and reproducible organoids for developmental modeling and toxicology testing.As a proof of concept,we applied the heavy metal cadmium to this newly improved organoid system to test whether it could be used to evaluate the neurotoxicity of environmental toxins.Brain organoids exposed to cadmium for 7 or 14 days manifested severe damage and abnormalities in their neurodevelopmental patterns,including bursts of cortical cell death and premature differentiation.Cadmium exposure caused progressive depletion of neural progenitor cells and loss of organoid integrity,accompanied by compensatory cell proliferation at ectopic locations.The convenience,flexibility,and controllability of this newly developed organoid platform make it a powerful and affordable alternative to animal models for use in neurodevelopmental,neurological,and neurotoxicological studies.展开更多
Deep neural networks(DNNs)are effective in solving both forward and inverse problems for nonlinear partial differential equations(PDEs).However,conventional DNNs are not effective in handling problems such as delay di...Deep neural networks(DNNs)are effective in solving both forward and inverse problems for nonlinear partial differential equations(PDEs).However,conventional DNNs are not effective in handling problems such as delay differential equations(DDEs)and delay integrodifferential equations(DIDEs)with constant delays,primarily due to their low regularity at delayinduced breaking points.In this paper,a DNN method that combines multi-task learning(MTL)which is proposed to solve both the forward and inverse problems of DIDEs.The core idea of this approach is to divide the original equation into multiple tasks based on the delay,using auxiliary outputs to represent the integral terms,followed by the use of MTL to seamlessly incorporate the properties at the breaking points into the loss function.Furthermore,given the increased training dificulty associated with multiple tasks and outputs,we employ a sequential training scheme to reduce training complexity and provide reference solutions for subsequent tasks.This approach significantly enhances the approximation accuracy of solving DIDEs with DNNs,as demonstrated by comparisons with traditional DNN methods.We validate the effectiveness of this method through several numerical experiments,test various parameter sharing structures in MTL and compare the testing results of these structures.Finally,this method is implemented to solve the inverse problem of nonlinear DIDE and the results show that the unknown parameters of DIDE can be discovered with sparse or noisy data.展开更多
Depressive disorder is a chronic,recurring,and potentially life-endangering neuropsychiatric disease.According to a report by the World Health Organization,the global population suffering from depression is experienci...Depressive disorder is a chronic,recurring,and potentially life-endangering neuropsychiatric disease.According to a report by the World Health Organization,the global population suffering from depression is experiencing a significant annual increase.Despite its prevalence and considerable impact on people,little is known about its pathogenesis.One major reason is the scarcity of reliable animal models due to the absence of consensus on the pathology and etiology of depression.Furthermore,the neural circuit mechanism of depression induced by various factors is particularly complex.Considering the variability in depressive behavior patterns and neurobiological mechanisms among different animal models of depression,a comparison between the neural circuits of depression induced by various factors is essential for its treatment.In this review,we mainly summarize the most widely used behavioral animal models and neural circuits under different triggers of depression,aiming to provide a theoretical basis for depression prevention.展开更多
基金Saudi Arabia for funding this work through Small Research Group Project under Grant Number RGP.1/316/45.
文摘The effective and timely diagnosis and treatment of ocular diseases are key to the rapid recovery of patients.Today,the mass disease that needs attention in this context is cataracts.Although deep learning has significantly advanced the analysis of ocular disease images,there is a need for a probabilistic model to generate the distributions of potential outcomes and thusmake decisions related to uncertainty quantification.Therefore,this study implements a Bayesian Convolutional Neural Networks(BCNN)model for predicting cataracts by assigning probability values to the predictions.It prepares convolutional neural network(CNN)and BCNN models.The proposed BCNN model is CNN-based in which reparameterization is in the first and last layers of the CNN model.This study then trains them on a dataset of cataract images filtered from the ocular disease fundus images fromKaggle.The deep CNN model has an accuracy of 95%,while the BCNN model has an accuracy of 93.75% along with information on uncertainty estimation of cataracts and normal eye conditions.When compared with other methods,the proposed work reveals that it can be a promising solution for cataract prediction with uncertainty estimation.
基金supported by the Beijing Natural Science Foundation(Grant No.L223013)。
文摘For the diagnostics and health management of lithium-ion batteries,numerous models have been developed to understand their degradation characteristics.These models typically fall into two categories:data-driven models and physical models,each offering unique advantages but also facing limitations.Physics-informed neural networks(PINNs)provide a robust framework to integrate data-driven models with physical principles,ensuring consistency with underlying physics while enabling generalization across diverse operational conditions.This study introduces a PINN-based approach to reconstruct open circuit voltage(OCV)curves and estimate key ageing parameters at both the cell and electrode levels.These parameters include available capacity,electrode capacities,and lithium inventory capacity.The proposed method integrates OCV reconstruction models as functional components into convolutional neural networks(CNNs)and is validated using a public dataset.The results reveal that the estimated ageing parameters closely align with those obtained through offline OCV tests,with errors in reconstructed OCV curves remaining within 15 mV.This demonstrates the ability of the method to deliver fast and accurate degradation diagnostics at the electrode level,advancing the potential for precise and efficient battery health management.
文摘The ability to accurately predict urban traffic flows is crucial for optimising city operations.Consequently,various methods for forecasting urban traffic have been developed,focusing on analysing historical data to understand complex mobility patterns.Deep learning techniques,such as graph neural networks(GNNs),are popular for their ability to capture spatio-temporal dependencies.However,these models often become overly complex due to the large number of hyper-parameters involved.In this study,we introduce Dynamic Multi-Graph Spatial-Temporal Graph Neural Ordinary Differential Equation Networks(DMST-GNODE),a framework based on ordinary differential equations(ODEs)that autonomously discovers effective spatial-temporal graph neural network(STGNN)architectures for traffic prediction tasks.The comparative analysis of DMST-GNODE and baseline models indicates that DMST-GNODE model demonstrates superior performance across multiple datasets,consistently achieving the lowest Root Mean Square Error(RMSE)and Mean Absolute Error(MAE)values,alongside the highest accuracy.On the BKK(Bangkok)dataset,it outperformed other models with an RMSE of 3.3165 and an accuracy of 0.9367 for a 20-min interval,maintaining this trend across 40 and 60 min.Similarly,on the PeMS08 dataset,DMST-GNODE achieved the best performance with an RMSE of 19.4863 and an accuracy of 0.9377 at 20 min,demonstrating its effectiveness over longer periods.The Los_Loop dataset results further emphasise this model’s advantage,with an RMSE of 3.3422 and an accuracy of 0.7643 at 20 min,consistently maintaining superiority across all time intervals.These numerical highlights indicate that DMST-GNODE not only outperforms baseline models but also achieves higher accuracy and lower errors across different time intervals and datasets.
文摘The increasing popularity of the Internet and the widespread use of information technology have led to a rise in the number and sophistication of network attacks and security threats.Intrusion detection systems are crucial to network security,playing a pivotal role in safeguarding networks from potential threats.However,in the context of an evolving landscape of sophisticated and elusive attacks,existing intrusion detection methodologies often overlook critical aspects such as changes in network topology over time and interactions between hosts.To address these issues,this paper proposes a real-time network intrusion detection method based on graph neural networks.The proposedmethod leverages the advantages of graph neural networks and employs a straightforward graph construction method to represent network traffic as dynamic graph-structured data.Additionally,a graph convolution operation with a multi-head attention mechanism is utilized to enhance the model’s ability to capture the intricate relationships within the graph structure comprehensively.Furthermore,it uses an integrated graph neural network to address dynamic graphs’structural and topological changes at different time points and the challenges of edge embedding in intrusion detection data.The edge classification problem is effectively transformed into node classification by employing a line graph data representation,which facilitates fine-grained intrusion detection tasks on dynamic graph node feature representations.The efficacy of the proposed method is evaluated using two commonly used intrusion detection datasets,UNSW-NB15 and NF-ToN-IoT-v2,and results are compared with previous studies in this field.The experimental results demonstrate that our proposed method achieves 99.3%and 99.96%accuracy on the two datasets,respectively,and outperforms the benchmark model in several evaluation metrics.
基金Supported by the Fundamental Research Funds for the Central Universities,No.2022CDJYGRH-004.
文摘Foot reflexology is a non-invasive and safe complementary therapy that works by massaging the reflex zones of the feet and exerts systemic or whole-body regulation through meridian nerve conduction.This therapy is commonly used in the treatment of various conditions such as autism and Parkinson's disease.However,there is limited reporting on the use of foot reflexology therapy for infants with sensorineural hearing loss(SNHL).Currently,there is no definitive conclusion on how foot reflexology therapy can influence hearing.This editorial holds some guiding significance regarding this clinical issue.The aim is to present physiological evidence of how foot reflexology therapy can impact infants with SNHL,thereby enhancing clinician’s awareness of foot reflexology in treating infants with SNHL.
基金supported by the National Key R&D Program of China(Grant No.2021YFA1001000)the National Natural Science Foundation of China(Grant Nos.82111530212,U23A20282,and 61971255)+2 种基金the Natural Science Founda-tion of Guangdong Province(Grant No.2021B1515020092)the Shenzhen Bay Laboratory Fund(Grant No.SZBL2020090501014)the Shenzhen Science,Technology and Innovation Commission(Grant Nos.KJZD20231023094659002,JCYJ20220530142809022,and WDZC20220811170401001).
文摘Neural organoids and confocal microscopy have the potential to play an important role in microconnectome research to understand neural patterns.We present PLayer,a plug-and-play embedded neural system,which demonstrates the utilization of sparse confocal microscopy layers to interpolate continuous axial resolution.With an embedded system focused on neural network pruning,image scaling,and post-processing,PLayer achieves high-performance metrics with an average structural similarity index of 0.9217 and a peak signal-to-noise ratio of 27.75 dB,all within 20 s.This represents a significant time saving of 85.71%with simplified image processing.By harnessing statistical map estimation in interpolation and incorporating the Vision Transformer–based Restorer,PLayer ensures 2D layer consistency while mitigating heavy computational dependence.As such,PLayer can reconstruct 3D neural organoid confocal data continuously under limited computational power for the wide acceptance of fundamental connectomics and pattern-related research with embedded devices.
基金supported by the National Natural Science Foundation of China(Grant No.92152301).
文摘Numerical simulation is dominant in solving partial differential equations(PDEs),but balancing fine-grained grids with low computational costs is challenging.Recently,solving PDEs with neural networks(NNs)has gained interest,yet cost-effectiveness and high accuracy remain a challenge.This work introduces a novel paradigm for solving PDEs,called multi-scale neural computing(MSNC),considering spectral bias of NNs and local approximation properties in the finite difference method(FDM).The MSNC decomposes the solution with a NN for efficient capture of global scale and the FDM for detailed description of local scale,aiming to balance costs and accuracy.Demonstrated advantages include higher accuracy(10 times for 1D PDEs,20 times for 2D PDEs)and lower costs(4 times for 1D PDEs,16 times for 2D PDEs)than the standard FDM.The MSNC also exhibits stable convergence and rigorous boundary condition satisfaction,showcasing the potential for hybrid of NN and numerical method.
基金supported by Westlake Universitythe Research Center for Industries of the Future of Westlake University (No. WU2022C040).
文摘The complexity and intricacy of the brain,which is composed of billions of neurons,pose significant challenges to its study.Understanding neural connections and communication at the single-cell level is crucial for unraveling the brain’s functions.This study presents a novel strategy that utilizes magnetic nanoparticles(MNPs)and magnetic fields to manipulate neurons,thereby creating customized small-scale neural circuits for studying neural connections.To establish the feasibility of this approach,the effects of MNPs on neurons were initially investigated,demonstrating their low toxicity.Subsequently,a micromagnet array(MMA)chip was employed to manipulate the neurons,facilitating their precise arrangement on the electrodes.Over several days,the neurons extended their axons and established connections with neighboring cells,forming small-scale circular neural circuits.These artificially engineered circuits offer a simplified and controlled environment for studying neural networks in contrast to naturally occurring biological networks.Furthermore,electrophysiological recordings were conducted to investigate the connections between the manipulated neurons.This study introduces a customized small-scale neural circuit platform with electrode-specific recording and stimulating capabilities,enabling the study of neuron-to-neuron interactions at the single-cell level.By leveraging MNPs and an MMA chip,this research offers a powerful tool for studying neural connections and advancing our understanding of the brain’s intricate workings.
基金supported in part by the Basic Science Research Program through the National Research Foundation of Korea(NRF),funded by the Ministry of Education(NRF-2021R1A6A1A03039493).
文摘Spiking Neural Network(SNN)inspired by the biological triggering mechanism of neurons to provide a novel solution for plant disease detection,offering enhanced performance and efficiency in contrast to Artificial Neural Networks(ANN).Unlike conventional ANNs,which process static images without fully capturing the inherent temporal dynamics,our approach represents the first implementation of SNNs tailored explicitly for agricultural disease classification,integrating an encoding method to convert static RGB plant images into temporally encoded spike trains.Additionally,while Bernoulli trials and standard deep learning architectures likeConvolutionalNeuralNetworks(CNNs)and Fully Connected Neural Networks(FCNNs)have been used extensively,our work is the first to integrate these trials within an SNN framework specifically for agricultural applications.This integration not only refines spike regulation and reduces computational overhead by 30%but also delivers superior accuracy(93.4%)in plant disease classification,marking a significant advancement in precision agriculture that has not been previously explored.Our approach uniquely transforms static plant leaf images into time-dependent representations,leveraging SNNs’intrinsic temporal processing capabilities.This approach aligns with the inherent ability of SNNs to capture dynamic,timedependent patterns,making them more suitable for detecting disease activations in plants than conventional ANNs that treat inputs as static entities.Unlike prior works,our hybrid encoding scheme dynamically adapts to pixel intensity variations(via threshold),enabling robust feature extraction under diverse agricultural conditions.The dual-stage preprocessing customizes the SNN’s behavior in two ways:the encoding threshold is derived from pixel distributions in diseased regions,and Bernoulli trials selectively reduce redundant spikes to ensure energy efficiency on low-power devices.We used a comprehensive dataset of 87,000 RGB images of plant leaves,which included 38 distinct classes of healthy and unhealthy leaves.To train and evaluate three distinct neural network architectures,DeepSNN,SimpleCNN,and SimpleFCNN,the dataset was rigorously preprocessed,including stochastic rotation,horizontal flip,resizing,and normalization.Moreover,by integrating Bernoulli trials to regulate spike generation,ourmethod focuses on extracting themost relevant featureswhile reducingcomputational overhead.Using a comprehensivedatasetof87,000RGB images across 38 classes,we rigorously preprocessed the data and evaluated three architectures:DeepSNN,SimpleCNN,and SimpleFCNN.The results demonstrate that DeepSNN outperforms the other models,achieving superior accuracy,efficient feature extraction,and robust spike management,thereby establishing the potential of SNNs for real-time,energy-efficient agricultural applications.
基金funded by the New Energy and Industrial Technology Development Organization(NEDO),grant number JPNP18002.
文摘Neural architecture search(NAS)optimizes neural network architectures to align with specific data and objectives,thereby enabling the design of high-performance models without specialized expertise.However,a significant limitation of NAS is that it requires extensive computational resources and time.Consequently,performing a comprehensive architectural search for each new dataset is inefficient.Given the continuous expansion of available datasets,there is an urgent need to predict the optimal architecture for the previously unknown datasets.This study proposes a novel framework that generates architectures tailored to unknown datasets by mapping architectures that have demonstrated effectiveness on the existing dataset into a latent feature space.As NAS is inherently represented as graph structures,we employed an encoder-decoder transformation model based on variational graph auto-encoders to perform this latent feature mapping.The encoder-decoder transformation model demonstrates strong capability in extracting features from graph structures,making it particularly well-suited for mapping NAS architectures.By training variational graph auto-encoders on existing high-quality architectures,the proposed method constructs a latent space and facilitates the design of optimal architectures for diverse datasets.Furthermore,to effectively define similarity amongarchitectures,wepropose constructing the latent spaceby incorporatingbothdataset andtaskfeatures.Experimental results indicate that our approach significantly enhances search efficiency and outperforms conventional methods in terms of model performance.
基金supported by the Fundamental Research Funds for the Central Universities (No.2024JBZY021)the National Natural Science Foundation of China (No.61575018)。
文摘Reverse design of highly GeO2-doped silica optical fibers with broadband and flat dispersion profiles is proposed using a neural network(NN) combined with a particle swarm optimization(PSO) algorithm.Firstly,the NN model designed to predict optical fiber dispersion is trained with an appropriate choice of hyperparameters,achieving a root mean square error(RMSE) of 9.47×10-7on the test dataset,with a determination coefficient(R2) of 0.999.Secondly,the NN is combined with the PSO algorithm for the inverse design of dispersion-flattened optical fibers.To expand the search space and avoid particles becoming trapped in local optimal solutions,the PSO algorithm incorporates adaptive inertia weight updating and a simulated annealing algorithm.Finally,by using a suitable fitness function,the designed fibers exhibit flat group velocity dispersion(GVD) profiles at 1 400—2 400 nm,where the GVD fluctuations and minimum absolute GVD values are below 18 ps·nm-1·km-1and 7 ps·nm-1·km-1,respectively.
基金supported by the National Natural Science Foundation of China(62272078)Chongqing Natural Science Foundation(CSTB2023NSCQ-LZX0069)the Science and Technology Research Program of Chongqing Municipal Education Commission(KJQN202300210)
文摘Dear Editor,This letter presents a novel latent factorization model for high dimensional and incomplete (HDI) tensor, namely the neural Tucker factorization (Neu Tuc F), which is a generic neural network-based latent-factorization-of-tensors model under the Tucker decomposition framework.
文摘In this paper,a sparse graph neural network-aided(SGNN-aided)decoder is proposed for improving the decoding performance of polar codes under bursty interference.Firstly,a sparse factor graph is constructed using the encoding characteristic to achieve high-throughput polar decoding.To further improve the decoding performance,a residual gated bipartite graph neural network is designed for updating embedding vectors of heterogeneous nodes based on a bidirectional message passing neural network.This framework exploits gated recurrent units and residual blocks to address the gradient disappearance in deep graph recurrent neural networks.Finally,predictions are generated by feeding the embedding vectors into a readout module.Simulation results show that the proposed decoder is more robust than the existing ones in the presence of bursty interference and exhibits high universality.
基金supported by the National Natural Science Foundation of China(Nos.12072118 and12372029)。
文摘This paper presents the variational physics-informed neural network(VPINN)as an effective tool for static structural analyses.One key innovation includes the construction of the neural network solution as an admissible function of the boundary-value problem(BVP),which satisfies all geometrical boundary conditions.We then prove that the admissible neural network solution also satisfies natural boundary conditions,and therefore all boundary conditions,when the stationarity condition of the variational principle is met.Numerical examples are presented to show the advantages and effectiveness of the VPINN in comparison with the physics-informed neural network(PINN).Another contribution of the work is the introduction of Gaussian approximation of the Dirac delta function,which significantly enhances the ability of neural networks to handle singularities,as demonstrated by the examples with concentrated support conditions and loadings.It is hoped that these structural examples are so convincing that engineers would adopt the VPINN method in their structural design practice.
基金supported in part by the Basic Science Research Program through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(NRF-2021R1A6A1A03039493)in part by the NRF grant funded by the Korean government(MSIT)(NRF-2022R1A2C1004401).
文摘Grains are the most important food consumed globally,yet their yield can be severely impacted by pest infestations.Addressing this issue,scientists and researchers strive to enhance the yield-to-seed ratio through effective pest detection methods.Traditional approaches often rely on preprocessed datasets,but there is a growing need for solutions that utilize real-time images of pests in their natural habitat.Our study introduces a novel twostep approach to tackle this challenge.Initially,raw images with complex backgrounds are captured.In the subsequent step,feature extraction is performed using both hand-crafted algorithms(Haralick,LBP,and Color Histogram)and modified deep-learning architectures.We propose two models for this purpose:PestNet-EF and PestNet-LF.PestNet-EF uses an early fusion technique to integrate handcrafted and deep learning features,followed by adaptive feature selection methods such as CFS and Recursive Feature Elimination(RFE).PestNet-LF utilizes a late fusion technique,incorporating three additional layers(fully connected,softmax,and classification)to enhance performance.These models were evaluated across 15 classes of pests,including five classes each for rice,corn,and wheat.The performance of our suggested algorithms was tested against the IP102 dataset.Simulation demonstrates that the Pestnet-EF model achieved an accuracy of 96%,and the PestNet-LF model with majority voting achieved the highest accuracy of 94%,while PestNet-LF with the average model attained an accuracy of 92%.Also,the proposed approach was compared with existing methods that rely on hand-crafted and transfer learning techniques,showcasing the effectiveness of our approach in real-time pest detection for improved agricultural yield.
基金supported in part by the National Natural Science Foundation of China,Nos.81927804(to GL),82260456(to LY),U21A20479(to LY)Science and Technology Planning Project of Shenzhen,No.JCYJ20230807140559047(to LY)+3 种基金Key-Area Research and Development Program of Guangdong Province,No.2020B0909020004(to GL)Guangdong Basic and Applied Research Foundation,No.2023A1515011478(to LY)the Science and Technology Program of Guangdong Province,No.2022A0505090007(to GL)Ministry of Science and Technology,Shenzhen,No.QN2022032013L(to LY)。
文摘Neural machine interface technology is a pioneering approach that aims to address the complex challenges of neurological dysfunctions and disabilities resulting from conditions such as congenital disorders,traumatic injuries,and neurological diseases.Neural machine interface technology establishes direct connections with the brain or peripheral nervous system to restore impaired motor,sensory,and cognitive functions,significantly improving patients'quality of life.This review analyzes the chronological development and integration of various neural machine interface technologies,including regenerative peripheral nerve interfaces,targeted muscle and sensory reinnervation,agonist–antagonist myoneural interfaces,and brain–machine interfaces.Recent advancements in flexible electronics and bioengineering have led to the development of more biocompatible and highresolution electrodes,which enhance the performance and longevity of neural machine interface technology.However,significant challenges remain,such as signal interference,fibrous tissue encapsulation,and the need for precise anatomical localization and reconstruction.The integration of advanced signal processing algorithms,particularly those utilizing artificial intelligence and machine learning,has the potential to improve the accuracy and reliability of neural signal interpretation,which will make neural machine interface technologies more intuitive and effective.These technologies have broad,impactful clinical applications,ranging from motor restoration and sensory feedback in prosthetics to neurological disorder treatment and neurorehabilitation.This review suggests that multidisciplinary collaboration will play a critical role in advancing neural machine interface technologies by combining insights from biomedical engineering,clinical surgery,and neuroengineering to develop more sophisticated and reliable interfaces.By addressing existing limitations and exploring new technological frontiers,neural machine interface technologies have the potential to revolutionize neuroprosthetics and neurorehabilitation,promising enhanced mobility,independence,and quality of life for individuals with neurological impairments.By leveraging detailed anatomical knowledge and integrating cutting-edge neuroengineering principles,researchers and clinicians can push the boundaries of what is possible and create increasingly sophisticated and long-lasting prosthetic devices that provide sustained benefits for users.
基金supported by the National Natural Science Foundation of China,Nos.82271397(to MG),82001293(to MG),82171355(to RX),81971295(to RX),and 81671189(to RX)。
文摘Microglial activation that occurs rapidly after closed head injury may play important and complex roles in neuroinflammation-associated neuronal damage and repair.We previously reported that induced neural stem cells can modulate the behavior of activated microglia via CXCL12/CXCR4 signaling,influencing their activation such that they can promote neurological recovery.However,the mechanism of CXCR4 upregulation in induced neural stem cells remains unclear.In this study,we found that nuclear factor-κB activation induced by closed head injury mouse serum in microglia promoted CXCL12 and tumor necrosis factor-αexpression but suppressed insulin-like growth factor-1 expression.However,recombinant complement receptor 2-conjugated Crry(CR2-Crry)reduced the effects of closed head injury mouse serum-induced nuclear factor-κB activation in microglia and the levels of activated microglia,CXCL12,and tumor necrosis factor-α.Additionally,we observed that,in response to stimulation(including stimulation by CXCL12 secreted by activated microglia),CXCR4 and Crry levels can be upregulated in induced neural stem cells via the interplay among CXCL12/CXCR4,Crry,and Akt signaling to modulate microglial activation.In agreement with these in vitro experimental results,we found that Akt activation enhanced the immunoregulatory effects of induced neural stem cell grafts on microglial activation,leading to the promotion of neurological recovery via insulin-like growth factor-1 secretion and the neuroprotective effects of induced neural stem cell grafts through CXCR4 and Crry upregulation in the injured cortices of closed head injury mice.Notably,these beneficial effects of Akt activation in induced neural stem cells were positively correlated with the therapeutic effects of induced neural stem cells on neuronal injury,cerebral edema,and neurological disorders post–closed head injury.In conclusion,our findings reveal that Akt activation may enhance the immunoregulatory effects of induced neural stem cells on microglial activation via upregulation of CXCR4 and Crry,thereby promoting induced neural stem cell–mediated improvement of neuronal injury,cerebral edema,and neurological disorders following closed head injury.
基金supported by the National Key R&D Program of China,No.2019YFA0110300(to ZG)the National Natural Science Foundation of China,Nos.81773302(to YF),32070862(to ZG).
文摘Human brain development is a complex process,and animal models often have significant limitations.To address this,researchers have developed pluripotent stem cell-derived three-dimensional structures,known as brain-like organoids,to more accurately model early human brain development and disease.To enable more consistent and intuitive reproduction of early brain development,in this study,we incorporated forebrain organoid culture technology into the traditional unguided method of brain organoid culture.This involved embedding organoids in matrigel for only 7 days during the rapid expansion phase of the neural epithelium and then removing them from the matrigel for further cultivation,resulting in a new type of human brain organoid system.This cerebral organoid system replicated the temporospatial characteristics of early human brain development,including neuroepithelium derivation,neural progenitor cell production and maintenance,neuron differentiation and migration,and cortical layer patterning and formation,providing more consistent and reproducible organoids for developmental modeling and toxicology testing.As a proof of concept,we applied the heavy metal cadmium to this newly improved organoid system to test whether it could be used to evaluate the neurotoxicity of environmental toxins.Brain organoids exposed to cadmium for 7 or 14 days manifested severe damage and abnormalities in their neurodevelopmental patterns,including bursts of cortical cell death and premature differentiation.Cadmium exposure caused progressive depletion of neural progenitor cells and loss of organoid integrity,accompanied by compensatory cell proliferation at ectopic locations.The convenience,flexibility,and controllability of this newly developed organoid platform make it a powerful and affordable alternative to animal models for use in neurodevelopmental,neurological,and neurotoxicological studies.
文摘Deep neural networks(DNNs)are effective in solving both forward and inverse problems for nonlinear partial differential equations(PDEs).However,conventional DNNs are not effective in handling problems such as delay differential equations(DDEs)and delay integrodifferential equations(DIDEs)with constant delays,primarily due to their low regularity at delayinduced breaking points.In this paper,a DNN method that combines multi-task learning(MTL)which is proposed to solve both the forward and inverse problems of DIDEs.The core idea of this approach is to divide the original equation into multiple tasks based on the delay,using auxiliary outputs to represent the integral terms,followed by the use of MTL to seamlessly incorporate the properties at the breaking points into the loss function.Furthermore,given the increased training dificulty associated with multiple tasks and outputs,we employ a sequential training scheme to reduce training complexity and provide reference solutions for subsequent tasks.This approach significantly enhances the approximation accuracy of solving DIDEs with DNNs,as demonstrated by comparisons with traditional DNN methods.We validate the effectiveness of this method through several numerical experiments,test various parameter sharing structures in MTL and compare the testing results of these structures.Finally,this method is implemented to solve the inverse problem of nonlinear DIDE and the results show that the unknown parameters of DIDE can be discovered with sparse or noisy data.
基金supported by the Brain&Behavior Research Foundation(30233).
文摘Depressive disorder is a chronic,recurring,and potentially life-endangering neuropsychiatric disease.According to a report by the World Health Organization,the global population suffering from depression is experiencing a significant annual increase.Despite its prevalence and considerable impact on people,little is known about its pathogenesis.One major reason is the scarcity of reliable animal models due to the absence of consensus on the pathology and etiology of depression.Furthermore,the neural circuit mechanism of depression induced by various factors is particularly complex.Considering the variability in depressive behavior patterns and neurobiological mechanisms among different animal models of depression,a comparison between the neural circuits of depression induced by various factors is essential for its treatment.In this review,we mainly summarize the most widely used behavioral animal models and neural circuits under different triggers of depression,aiming to provide a theoretical basis for depression prevention.
