检索结果-南通市图书馆

Deep Learning-Based Speech Emotion recognition: Leveraging Diverse Datasets and Augmentation Techniques for robust Modeling

在线全文

学校读者我要写书评

暂无评论

Journal of Harbin Institute of Technology(New series) 2024年

作者： Ayush porwal praveen Kumar Tyagi Ajay Sharma Dheeraj Kumar agarwal Department of Electronics and Instrumentation Engineering， Shri G.S.Institute of Technology and Science Department of Electronics and Communication Engineering， Maulana Azad National Institute of Technology School of Computing Science and Engineering， VIT Bhopal University

In recent years， Speech Emotion recognition (SEr) has developed into an essential instrument for interpreting human emotions from auditory data. The proposed research focused on the development of a SEr system employing deep learning and multiple datasets containing samples of emotive speech. The primary objective of this research endeavor is to investigate the utilization of Convolutional Neural Networks (CNNs) in the process of sound feature extraction. Stretching， pitch manipulation， and noise injection are a few of the techniques utilized in this study to improve the data quality. The investigation includes coverage of these methods. Feature extraction methods including Zero Crossing rate， Chroma＿stft， MFCC， rMS， and MelSpectogram are used to train a model. By using these techniques， audio signals can be transformed into recognized features that can be utilized to train the model. Ultimately， the study will produce a thorough evaluation of the model’s performance. When this method was applied， the model achieved an impressive accuracy of 94.57% on the test dataset. proposed work also validated on EMO-BD and IEMOCAp dataset. These consist of further data augmentation， feature engineering， and hyperparameter optimization. By following these development paths， SEr systems will be able to be implemented in real-world scenarios with greater accuracy and resilience.

关键词：

Tuberculosis in kidney transplant candidates and recipients

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

World Journal of Transplantation 2024年第3期14卷 59-69页

作者： pallavi prasad Sourabh Sharma Subashri Mohanasundaram Anupam agarwal Himanshu Verma Department of Nephrology Vardhman Mahavir Medical College and Safdarjung HospitalNew Delhi 110029DelhiIndia Department of Nephrology Prashanth HospitalChennai 600042Tamil NāduIndia

Tuberculosis(TB)is the leading cause of infectious mortality and morbidity in the world,second only to coronavirus disease *** with chronic kidney disease and kidney transplant recipients are at a higher risk of developing TB than the general *** TB is difficult to diagnose in this population due to close *** transplant candidates should be screened for latent TB infection and given TB *** who develop active TB pre-or post-trans-plantation should receive multidrug combination therapy of antitubercular therapy for the recommended duration with optimal dose modification as per glomerular filtration rate.

关键词： Tuberculosis Latent Antitubercular therapy Kidney transplantation Donor Interferon-gamma release assays Tuberculin rifampicin Isoniazid

Verification and Validation of High-resolution Inviscid and Viscous Conical Nozzle Flows

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Communications on Applied Mathematics and Computation 2024年第1期6卷 533-549页

作者： Luciano K.Araki rafael B.de r.Borges Nicholas Dicati p.da Silva Chi-Wang Shu Department of Mechanical Engineering Federal University of ParanáCuritiba81531-980ParanáBrazil Mathematics and Statistics Institute Rio de Janeiro State UniversityRio de Janeiro20550-900Rio de JaneiroBrazil Department of Mechanical Engineering State University of MaringáMaringá87020-900ParanáBrazil Division of Applied Mathematics Brown UniversityProvidenceRI02912USA

Capturing elaborated flow structures and phenomena is required for well-solved numerical *** finite difference methods allow simple discretization of mesh and model ***,they need simpler meshes,e.g.,*** inverse Lax-Wendroff(ILW)procedure can handle complex geometries for rectangular ***-resolution and high-order methods can capture elaborated flow structures and *** also have strong mathematical and physical backgrounds,such as positivity-preserving,jump conditions,and wave propagation *** perceive an effort toward direct numerical simulation,for instance,regarding weighted essentially non-oscillatory(WENO)***,we propose to solve a challenging engineering application without turbulence *** aim to verify and validate recent high-resolution and high-order *** check the solver accuracy,we solved vortex and Couette ***,we solved inviscid and viscous nozzle flows for a conical *** employed the finite difference method,positivity-preserving Lax-Friedrichs splitting,high-resolution viscous terms discretization,fifth-order multi-resolution WENO,ILW,and third-order strong stability preserving *** showed the solver is high-order and captured elaborated flow structures and *** can see oblique shocks in both nozzle *** the viscous flow,we also captured a free-shock separation,recirculation,entrainment region,Mach disk,and the diamond-shaped pattern of nozzle flows.

关键词： High-resolution Compressible Navier-Stokes Free-shock separation Nozzle flow

On the functions of astrocyte-mediated neuronal slow inward currents

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Neural regeneration research 2024年第12期19卷 2602-2612页

作者： Balázs pál Department of physiology Faculty of MedicineUniversity of DebrecenDebrecenHungary

Slow inward currents are known as neuronal excitatory currents mediated by glutamate release and activation of neuronal extra synaptic N-met hyl-D-aspartate receptors with the contribution of *** events are significantly slower than the excitatory postsynaptic *** of slow inward currents are determined by seve ral factors including the mechanisms of astrocytic activation and glutamate release,as well as the diffusion pathways from the release site towards the extra synaptic recepto *** are stimulated by neuronal network activity,which in turn excite neurons,forming an astrocyte-neuron feedback *** as a consequence of brain edema,astrocytic swelling can also induce slow inward currents under pathological *** is a growing body of evidence on the roles of slow inward currents on a single neuron or local network *** events often occur in synchro ny on neurons located in the same astrocytic *** synchronization of neuronal excitability,slow inward currents also set synaptic strength via eliciting timing-dependent synaptic *** addition,slow inward currents are also subject to non-synaptic plasticity triggered by long-la sting stimulation of the excitatory *** note,there might be important regionspecific differences in the roles and actions triggering slow inward *** greater networks,the pathophysiological roles of slow inward currents can be better understood than physiological *** inward currents are identified in the pathophysiological background of autism,as slow inward currents drive early hypersynchrony of the neural *** inward currents are significant contributors to paroxysmal depolarizational shifts/interictal *** events are related to epilepsy,but also found in Alzheimer's disease,parkinson's disease,and stroke,leading to the decline of cognitive *** with features overlapping with slow inward currents(excitatory,N-methyl-Dasp

关键词： astrocyte cortical spreading depolarization gliotransmission glutamate neural synchronization NMDA receptor paroxysmal depolarizational shift slow inward current

An Efficient Approach to Escalate the Speed of Training Convolution Neural Networks

在线全文

学校读者我要写书评

暂无评论

China Communications 2024年第2期21卷 258-269页

作者： p pabitha Anusha Jayasimhan Department of Computer Technology Anna UniversityMadras Institute of Technology CampusChennai 600044India

Deep neural networks excel at image identification and computer vision applications such as visual product search, facial recognition, medical image analysis, object detection, semantic segmentation,instance segmentation, and many others. In image and video recognition applications, convolutional neural networks(CNNs) are widely employed. These networks provide better performance but at a higher cost of computation. With the advent of big data, the growing scale of datasets has made processing and model training a time-consuming operation, resulting in longer training times. Moreover, these large scale datasets contain redundant data points that have minimum impact on the final outcome of the model. To address these issues, an accelerated CNN system is proposed for speeding up training by eliminating the noncritical data points during training alongwith a model compression method. Furthermore, the identification of the critical input data is performed by aggregating the data points at two levels of granularity which are used for evaluating the impact on the model *** experiments are conducted using the proposed method on CIFAr-10 dataset on resNet models giving a 40% reduction in number of FLOps with a degradation of just 0.11% accuracy.

关键词： CNN deep learning image classification model compression

Spastin and alsin protein interactome analyses begin to reveal key canonical pathways and suggest novel druggable targets

在线全文

学校读者我要写书评

暂无评论

Neural regeneration research 2025年第3期20卷 725-739页

作者： Benjamin r.Helmold Angela Ahrens Zachary Fitzgerald p.Hande Ozdinler Department of Neurology Feinberg School of MedicineNorthwestern UniversityChicagoILUSA Center for Molecular Innovation and Drug Discovery Center for Developmental TherapeuticsChemistry of Life Processes InstituteNorthwestern UniversityEvanstonILUSA Mesulam Center for Cognitive Neurology and Alzheimer’s Disease Feinberg School of MedicineNorthwestern UniversityChicagoILUSA Feinberg School of Medicine Les Turner ALS Center at Northwestern UniversityChicagoILUSA

Developing effective and long-term treatment strategies for rare and complex neurodegenerative diseases is challenging. One of the major roadblocks is the extensive heterogeneity among patients. This hinders understanding the underlying disease-causing mechanisms and building solutions that have implications for a broad spectrum of patients. One potential solution is to develop personalized medicine approaches based on strategies that target the most prevalent cellular events that are perturbed in patients. Especially in patients with a known genetic mutation, it may be possible to understand how these mutations contribute to problems that lead to neurodegeneration. protein–protein interaction analyses offer great advantages for revealing how proteins interact, which cellular events are primarily involved in these interactions, and how they become affected when key genes are mutated in patients. This line of investigation also suggests novel druggable targets for patients with different mutations. Here, we focus on alsin and spastin, two proteins that are identified as “causative” for amyotrophic lateral sclerosis and hereditary spastic paraplegia, respectively, when mutated. Our review analyzes the protein interactome for alsin and spastin, the canonical pathways that are primarily important for each protein domain, as well as compounds that are either Food and Drug Administration–approved or are in active clinical trials concerning the affected cellular pathways. This line of research begins to pave the way for personalized medicine approaches that are desperately needed for rare neurodegenerative diseases that are complex and heterogeneous.

关键词： ALS2 alsin amyotrophic lateral sclerosis hereditary spastic paraplegia neurodegenerative diseases personalized medicine precision medicine protein interactome protein-protein interactions SpAST spastin

Kinematic analysis of flexible bipedal robotic systems

在线全文

学校读者我要写书评

暂无评论

Applied Mathematics and Mechanics(English Edition) 2024年第5期45卷 795-818页

作者： r.FAZEL A.M.SHAFEI S.r.NEKOO Department of Mechanical Engineering Shahid Bahonar University of KermanKerman ***Iran Departamento de Ingenieria de Sistemas y Automatica Escuela Técnica Superior de IngenieriaUniversidad de SevillaSeville 41092Spain

In spite of its intrinsic complexities,the passive gait of bipedal robots on a sloping ramp is a subject of interest for numerous *** distinguishes the present research from similar works is the consideration of flexibility in the constituent links of this type of robotic *** is not a far-fetched assumption because in the transient(impact)phase,due to the impulsive forces which are applied to the system,the likelihood of exciting the vibration modes increases ***,the human leg bones that are involved in walking are supported by viscoelastic muscles and ***,for achieving more exact results,it is essential to model the robot links with viscoelastic *** this end,the Gibbs-Appell formulation and Newton's kinematic impact law are used to derive the most general form of the system's dynamic equations in the swing and transient phases of *** most important issue in the passive walking motion of bipedal robots is the determination of the initial robot configuration with which the system could accomplish a periodic and stable gait solely under the effect of gravitational *** extremely unstable nature of the system studied in this paper and the vibrations caused by the impulsive forces induced by the impact of robot feet with the inclined surface are some of the very serious challenges encountered for achieving the above-mentioned *** overcome such challenges,an innovative method that uses a combination of the linearized equations of motion in the swing phase and the algebraic motion equations in the transition phase is presented in this paper to obtain an eigenvalue *** solving this problem,the suitable initial conditions that are necessary for the passive gait of this bipedal robot on a sloping surface are *** effects of the characteristic parameters of elastic links including the modulus of elasticity and the Kelvin-Voigt coefficient on the walking stability of this type of

关键词： bipedal robot flexible link swing phase transient phase eigenvalue problem Kelvin-Voigt coefficient

Holographic dark energy in non-conserved gravity theory

在线全文

学校读者我要写书评

暂无评论

Communications in Theoretical physics 2024年第4期76卷 126-133页

作者： H r Fazlollahi Institute of Gravitation and Cosmology Peoples Friendship University of Russia(RUDN University)6 Miklukho-Maklaya StMoscow117198Russia

recently, reconsidering the rastall idea T_(μ;νν)^(v)=α_(,μ) through relativistic thermodynamics gives a new form for the scalar field α which led us to construct modern modified theory of gravity debugged ‘non-conserved gravity theory’ Fazlollahi 2023 Euro. phys. J. C 83 923. This theory unlike other modified theories of gravity cannot directly explain the current acceleration expansion in the absence of the cosmological constant and or existence of other forms of dark energy. Hence, in this study we have reinvestigated holographic dark energy Ρ_(X)~H^(2) in the non-conserved theory of gravity. In this context, the density and pressure of dark energy depend on the non-conserved term and density of the dust matter field. As shown, due to non-conservation effects on large-scale structures, unlike the original holographic model, our model onsets an acceleration epoch for the current Universe satisfies observations. Moreover, the interaction and viscous scenarios are studied for this model.

关键词： non-conserved gravity covariant form of thermodynamics holographic dark energy

Unmixing detrital zircon U-pb ages reveals tectonic and climatic depositional influences on the Carboniferous Ansilta Formation,Calingasta-Uspallata Basin,Western Argentina

在线全文

学校读者我要写书评

暂无评论

Geoscience Frontiers 2024年第4期15卷 61-77页

作者： J.r.Malone J.E.Malone J.L.Isbell D.H.Malone J.p.Craddock K.N.pauls Department of Geological Sciences Jackson School of GeosciencesUniversity of Texas at AustinAustinTX 78712USA Iowa Geological Survey University of IowaIowa CityIA 52240USA Department of Geosciences University of Wisconsin-MilwaukeeMilwaukeeWI 53211USA Department of Geography Geology and the EnvironmentIllinois State UniversityNormalIL 61790-4400USA

The Late paleozoic Ice Age(LpIA)was a principal control of sedimentation across Gondwana from the late Devonian through early *** assess the hypothesis that glacial to interglacial transitions in western Argentina were the primary control influencing sediment routing patterns among the various Carboniferous-permian basins in western *** Carboniferous Ansilta Formation consists of glaciomarine,nearshore,and fluvial systems deposited during the LpIA along the eastern margin of the Calingasta-Uspallata Basin in *** lower,glacially influenced succession of the Ansilta Formation records at least five glacial advances;the upper succession of consists of progradational shallow marine,deltaic,and fluvial *** combine 1225 new U-pb zircon ages from six samples of the Carboniferous Ansilta Formation in the Calingasta-Uspallata Basin with 5864 U-pb ages from 147 pub-lished samples in the detrital pymix forward mixture model to characterize provenance *** the glacially influenced lower Ansilta Formation,sediment was derived locally from the protoprecordillera,which was a prominent highland with alpine glaciers flowing west and east into the Calingasta-Uspallata and paganzo basins,***,there was little or no connection between these two basins during Serpukhovian-Bashkirian *** fluvial/deltaic upper Ansilta had distal sediment sources in the Sierras ***,our results support the collapse of the protoprecordillera topographic barrier,enabling drainage patterns connecting the paganzo and Calingasta-Uspallata basins by late pennsylvanian-early permian time.

关键词： Late paleozoic Ice Age Argentina Calingasta-Uspallata Basin Ansilta Formation Detrital zircon provenance