This paper presents the design of a machine learning-based classifier for differentiation between Schizophrenia patients and healthy controls using features extracted from electroencephalograph(EEG) signals based on event related potential(ERP). A number are an online EEG dataset with 81 subjects, including 32 49 patients. The preprocessed since is relatively small, random forest learning algorithm chosen to be applied different combinations feature sets classification. It found that...

Abstract Named Entity Recognition (NER) plays a crucial role in the field of Natural Language Processing, holding significant value applications such as information extraction, knowledge graphs, and question–answering systems. However, Chinese NER faces challenges semantic complexity, uncertain entity boundaries, nested structures. To address these issues, this study proposes an innovative approach, namely Multi-Granularity BERT Adapter Efficient Global Pointer (MGBERT-Pointer). The encoding...

Electrocardiogram (ECG) heartbeat classification plays a vital role in early diagnosis and effective treatment, which provide opportunities for earlier prevention intervention. In an effort to continuously monitor detect abnormalities patients’ ECG signals on portable devices, this paper present lightweight method based spiking neural network (SNN), relatively shallow SNN model integrated with channel-wise attentional module. We further explore the best-optimized architecture, benefits from...

Establishing a deep learning model for transformer fault diagnosis using oil chromatogram data requires large number of samples. The lack and imbalance can lead to overfitting, representativeness the model, unsatisfactory prediction results on test set data, making it difficult accurately diagnose faults. A conditional Wasserstein generative adversarial network with gradient penalty optimization (CWGAN-GP) is adopted in this paper, which based expand chromatography samples 500 sets 5 types...

Summary Aims Studies showed fastigial nucleus stimulation ( FNS ) reduced brain damage, but the mechanisms of neuroprotection induced by were not entirely understood; Micro RNA s are noncoding molecules that regulate gene expression in a posttranscriptional manner, their functional consequence response to ischemia–reperfusion IR remains unknown. We investigated role micro ‐29c rat. Methods The rat models conducted 1 day after . Besides, miR‐29c antagomir (or agomir or control) was infused...

Chaotic systems can be synchronized and used for secure communication to transmit video, audio text files. Field Programmable Gate Arrays (FPGAs) are beneficial the implementation of high speed, low cost power embedded systems. In this paper, a model-based design approach is presented FPGA optimization chaotic generators. Lorenz attractor has its significance in studying as subject paper. The conceptual model built using MATLAB Simulink, equivalent hardware created Xilinx System Generator...

This paper presents an Artificial Neural Network (ANN) design for a chaotic generator, and the training performances three layer ANN architecture with different number of hidden neurons. Chaotic systems can be synchronized used secure communication. such as Lorenz attractor, Rossler attractor Chen's system are generally implemented directly based on their definitions represented by unique group ordinary differential equations (ODEs). An feed forward trained using output values system. The...

This paper presents the hardware implementation of single-neuron models with three types activation functions using fixed-point data format on Field Programmable Gate Arrays (FPGA). Activation function defines transfer behavior a neuron model and consequently Artificial Neural Network (ANN) constructed it. compared single designed bipolar ramp, threshold sigmoid functions. It is also demonstrated that FPGA performance can be significantly improved by 16-bit instead 32-bit floating-point for function.

This paper presents the preliminary work of a multidisciplinary brain research program. The goal this program is to generate accurate and effective signals for non-invasive stimulation, deliver hardware prototype monitor treat motion related mental disease such as Parkinson's Epilepsy. It was shown in previous that Electroencephalogram (EEG) captured from activities demonstrate chaotic features. Artificial neural network (ANN) resembles biological can be used simulate system. trained ANN...

Abstract To improve the performance of named entity recognition in lack well-annotated data, a transfer learning-based Chinese model is proposed this paper. The specific tasks are as follows: (1) first/, data method based on features proposed. By calculating similarity feature distribution between low resource and high most representative selected for mapping, distance two domains calculated to make up gap then trained by data. (2) Then, an boundary detection This utilizes BiLSTM+CRF main...

This proposed research explores a novel approach to image classification by deploying complex-valued neural network (CVNN) on Field-Programmable Gate Array (FPGA), specifically for classifying 2D images transformed into polar form. The aim of this is address the limitations existing models in terms energy and resource efficiency, exploring potential FPGA-based hardware acceleration conjunction with advanced architectures like CVNNs. methodological innovation lies Cartesian transformation...

Spiking Neural Network (SNN) is a particular Artificial Networks (ANN) form. An SNN has similar features as an ANN, but different information system that will allow to have higher energy efficiency than ANN. This paper presents the design and implementation of on FPGA. The model designed be lower power consumption existing models in aspect FPGA accuracy loss training method part algorithm. coding scheme proposed this rate scheme. introduces conversion directly map trained parameters from ANN...

The rheological behavior of viscoelastic fluids was investigated with regard to the specialty tertiary oil recovery. Aqueous polyacrylamide solutions at different concentrations were selected simulate polymer systems, and Haake RS 150 type rheometer used measure behavior. experimental results showed that viscoelasticity positively influenced by solution negatively affected temperatures. coefficient first normal stress deference decreased increasing shear rate. In addition, relationship...

This study identified potential biomarkers in urine, plasma and feces of high fructose-fed rats using¹H NMR-based metabonomics.

Neural networks (NNs) have been demonstrated to be useful in a broad range of applications, such as image recognition, automatic translation, and advertisement recommendation. State-of-the-art NNs are known both computationally memory intensive, due the ever-increasing deep structure, i.e., multiple layers with massive neurons connections (i.e., synapses). Sparse emerged an effective solution reduce amount computation required. Though existing NN accelerators able efficiently process dense...

In this paper we investigate the performance of statistical modeling digital mammograms by means wavelet domain hidden Markov trees for its inclusion to a computer‐aided diagnostic prompting system. The system is designed detecting clusters microcalcifications. Their further discrimination as benign or malignant be done radiologists. model used segmenting images based on maximum likelihood classifier enhanced weighting technique. Further classification incorporates spatial filtering single...

Solar energy is abundant, and technological advances have made solar systems more affordable than ever before. Using photovoltaic (PV) could significantly reduce our reliance on fossil fuels, facilitate sustainable uses. power utilities, such as self-compacting disposal bins be used to enhance waste management processes. This particularly important in Canada, where $3.3 billion was spent 2016. In this study, irradiance climatic conditions at eight locations a University campus Regina,...

We design and fabricate a novel ring-core fiber with modulated refractive index profile to suppress the micro-bending induced modal coupling.Two OAM mode-group transmission over 50-km without using MIMO equalization is also experimentally demonstrated.

Variational dropout (VD) is a generalization of Gaussian dropout, which aims at inferring the posterior network weights based on log-uniform prior them to learn these as well rate simultaneously. The not only interprets regularization capacity in training, but also underpins inference such posterior. However, an improper (i.e., its integral infinite), causes be ill-posed, thus restricting performance VD. To address this problem, we present new termed variational Bayesian (VBD), turns exploit...

We have demonstrated direct generation of high-order orbital angular momentum (OAM) modes by chiral long-period gratings (CLPGs) written in a few mode fiber (FMF) via CO2 laser. The core (LP11, LP21, and LP31 modes) were realized the CLPGs through high order diffraction with efficiencies more than 99.69%. -1 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">st</sup> , -2 xmlns:xlink="http://www.w3.org/1999/xlink">nd</sup> -3...