Neural Network Pattern Research Articles

Sparse Convolution FPGA Accelerator Based on Multi-Bank Hash Selection.

Reconfigurable processor-based acceleration of deep convolutional neural network (DCNN) algorithms has emerged as a widely adopted technique, with particular attention on sparse neural network acceleration as an active research area. However, many computing devices that claim high computational power still struggle to execute neural network algorithms with optimal efficiency, low latency, and minimal power consumption. Consequently, there remains significant potential for further exploration into improving the efficiency, latency, and power consumption of neural network accelerators across diverse computational scenarios. This paper investigates three key techniques for hardware acceleration of sparse neural networks. The main contributions are as follows: (1) Most neural network inference tasks are typically executed on general-purpose computing devices, which often fail to deliver high energy efficiency and are not well-suited for accelerating sparse convolutional models. In this work, we propose a specialized computational circuit for the convolutional operations of sparse neural networks. This circuit is designed to detect and eliminate the computational effort associated with zero values in the sparse convolutional kernels, thereby enhancing energy efficiency. (2) The data access patterns in convolutional neural networks introduce significant pressure on the high-latency off-chip memory access process. Due to issues such as data discontinuity, the data reading unit often fails to fully exploit the available bandwidth during off-chip read and write operations. In this paper, we analyze bandwidth utilization in the context of convolutional accelerator data handling and propose a strategy to improve off-chip access efficiency. Specifically, we leverage a compiler optimization plugin developed for Vitis HLS, which automatically identifies and optimizes on-chip bandwidth utilization. (3) In coefficient-based accelerators, the synchronous operation of individual computational units can significantly hinder efficiency. Previous approaches have achieved asynchronous convolution by designing separate memory units for each computational unit; however, this method consumes a substantial amount of on-chip memory resources. To address this issue, we propose a shared feature map cache design for asynchronous convolution in the accelerators presented in this paper. This design resolves address access conflicts when multiple computational units concurrently access a set of caches by utilizing a hash-based address indexing algorithm. Moreover, the shared cache architecture reduces data redundancy and conserves on-chip resources. Using the optimized accelerator, we successfully executed ResNet50 inference on an Intel Arria 10 1150GX FPGA, achieving a throughput of 497 GOPS, or an equivalent computational power of 1579 GOPS, with a power consumption of only 22 watts.

Quantifying information stored in synaptic connections rather than in firing patterns of neural networks.

A cornerstone of our understanding of both biological and artificial neural networks is that they store information in the strengths of connections among the constituent neurons. However, in contrast to the well-established theory for quantifying information encoded by the firing patterns of neural networks, little is known about quantifying information encoded by its synaptic connections. Here, we develop a theoretical framework using continuous Hopfield networks as an exemplar for associative neural networks, and data that follow mixtures of broadly applicable multivariate log-normal distributions. Specifically, we analytically derive the Shannon mutual information between the data and singletons, pairs, triplets, quadruplets, and arbitrary n-tuples of synaptic connections within the network. Our framework corroborates well-established insights about storage capacity of, and distributed coding by, neural firing patterns. Strikingly, it discovers synergistic interactions among synapses, revealing that the information encoded jointly by all the synapses exceeds the 'sum of its parts'. Taken together, this study introduces an interpretable framework for quantitatively understanding information storage in neural networks, one that illustrates the duality of synaptic connectivity and neural population activity in learning and memory.

An Automated Approach for Discriminating Hole Cleaning Efficiency While Predicting Penetration Rate in Egyptian Western Desert Wells

AbstractHigher rate of penetration (ROP) indicates successful drilling operation but is not the only drilling success measure. However, Conventional ROP prediction methods focus on increasing ROP and neglect the hole cleaning state, which can be altered by ROP changes. Higher ROP in vertical and deviated wells may increase cutting concentration, leading to hole cleaning problems such as overpulling and stuck pipe. With this problem in mind, this paper utilized geological, rheological, and drilling data of 31 vertical wells across four oil fields located in the Egyptian Western Desert, developed intelligent ROP prediction model through back propagation neural network (BPNN), and compared the proposed BPNN results with an empirical model. Finally, the pattern recognition algorithms including discriminant analysis, support vector machines, and neural network pattern recognition (NNPR) were implemented to discriminate hole cleaning efficiency following the ROP prediction process. Recognition models were developed based on predicted ROP, bit wear rate, specific energy, and drilling fluid carrying capacity index to evaluate hole cleaning. The accuracy of the multi-strategy classifier was evaluated using area under curve, confusion matrix, and receiver operating characteristic. The BPNN model outperformed the empirical model in terms of linear correlation coefficient (R = 98.6%) and average absolute error (AAE = 5.5%). Additionally, the best classification performance was achieved using the NNPR algorithm with 91% accuracy and a cross-validation error equal to zero. For validity, the proposed approach predicted ROP and classified hole cleaning efficiency for new vertical well in adjacent oil field, resulting in a 6% improvement in ROP.

IMPLEMENTATION OF BACKPROPAGATION AND HYBRID ARIMA-NN METHODS IN PREDICTING ACCURACY LEVELS OF RAINFALL IN MAKASSAR CITY

Hybrid ARIMA-NN is a combined approach of the ARIMA model used to capture linear patterns in time series data and Artificial Neural Networks (ANN) to handle non-linear and stochastic patterns. Using a gradient descent algorithm, backpropagation adjusts synaptic weights based on the error between the network's prediction and actual training data values. In this study, a comparison was made between the Backpropagation method and Hybrid ARIMA-NN in forecasting rainfall in Makassar City. Rainfall data in Makassar City uses data from the rainfall measuring station at the Paotere Maritime Meteorological Station in Makassar. The activation functions used are ReLU and Leaky ReLU with epoch parameters set at 350, and learning rates of 0.01, 0.001, 0.0001, and 0.00001. The two best methods selected for further evaluation are Backpropagation with architecture 12-32-16-8-1 and Hybrid ARIMA-NN (ARIMA [4,0,1]-NN 12-256-128-64-1). The ARIMA model (4,0,1) with AIC values of 1303.4 and RMSE 162,369 is the best compared to other models, which aligns with the advantages of backpropagation architecture. The results showed that the Backpropagation method excelled with an RMSE value of 137.320 or 0.1149, indicating high accuracy in forecasting changes in seasonal trends and patterns. Hybrid ARIMA-NN gives good results with RMSE 145.834, as residues contain better nonlinearity compared to ARIMA models (4,0,1), although it shows a slightly higher error rate compared to Backpropagation.

Efficient microplastic identification by hyperspectral imaging: A comparative study of spatial resolutions, spectral ranges and classification models to define an optimal analytical protocol

Microplastics (MPs) pollution is a global and challenging issue, necessitating the development of efficient analytical strategies for their detection to monitor their environmental impact.This study aims to define an optimal analytical protocol for characterizing MPs by hyperspectral imaging (HSI), comparing different setups based on spatial resolution, spectral range and classification models. The investigated MPs include polymers commonly found in the environment, such as polystyrene (PS), polypropylene (PP) and high-density polyethylene (HDPE), subdivided in three size classes (1000–2000 μm, 500–1000 μm, 250–500 μm). Furthermore, MP particles with diameters ranging from 30 to 250 μm were assessed to determine the limit of detection (LOD) in the different configurations. Hyperspectral images were acquired with two spatial resolutions, 150 and 30 μm/pixel, and two spectral ranges, 1000–1700 nm (NIR) and 1000–2500 nm (SWIR). Three classification models, Partial Least Square-Discriminant Analysis (PLS-DA), Error Correction Output Coding-Support Vector Machine (ECOC-SVM) and Neural Network Pattern Recognition (NNPR) were tested on the acquired images. The correctness of these models was evaluated by prediction maps and statistical parameters (Recall, Specificity and Accuracy). The results demonstrated that for MP particles larger than 250 μm, the optimal setup is a spatial resolution of 150 μm/pixel and a spectral range of 1000–1700 nm, utilizing a linear classification model like PLS-DA. This approach offers accurate predictions while being time- and cost-efficient. For MPs smaller than 250 μm, a higher spatial resolution of 30 μm/pixel with a spectral range of 1000–2500 nm and a non-linear classification method like ECOC-SVM is preferable. The LOD is 250 μm for the 150 μm/pixel resolution and ranges from 100 to 200 μm for the 30 μm/pixel resolution. These findings provide a valuable guide for selecting the appropriate HSI acquisition conditions and data processing methods to optimally characterize MPs of different sizes.

Turbofan engine health status prediction with neural network pattern recognition and automated feature engineering

PurposeThis study aims to present the concept of aircraft turbofan engine health status prediction with artificial neural network (ANN) pattern recognition but augmented with automated features engineering (AFE).Design/methodology/approachThe main concept of engine health status prediction was based on three case studies and a validation process. The first two were performed on the engine health status parameters, namely, performance margin and specific fuel consumption margin. The third one was generated and created for the engine performance and safety data, specifically created for the final test. The final validation of the neural network pattern recognition was the validation of the proposed neural network architecture in comparison to the machine learning classification algorithms. All studies were conducted for ANN, which was a two-layer feedforward network architecture with pattern recognition. All case studies and tests were performed for both simple pattern recognition network and network augmented with automated feature engineering (AFE).FindingsThe greatest achievement of this elaboration is the presentation of how on the basis of the real-life engine operational data, the entire process of engine status prediction might be conducted with the application of the neural network pattern recognition process augmented with AFE.Practical implicationsThis research could be implemented into the engine maintenance strategy and planning. Engine health status prediction based on ANN augmented with AFE is an extremely strong tool in aircraft accident and incident prevention.Originality/valueAlthough turbofan engine health status prediction with ANN is not a novel approach, what is absolutely worth emphasizing is the fact that contrary to other publications this research was based on genuine, real engine performance operational data as well as AFE methodology, which makes the entire research very reliable. This is also the reason the prediction results reflect the effect of the real engine wear and deterioration process.

Graph features based classification of bronchial and pleural rub sound signals: the potential of complex network unwrapped.

The study presents a novel technique for lung auscultation based on graph theory, emphasizing the potential of graph parameters in distinguishing lung sounds and supporting earlier detection of various respiratory pathologies. The frequency spread and the component magnitudes are revealed from the analysis of eighty-five bronchial (BS) and pleural rub (PS) lung sounds employing the power spectral density (PSD) plot and wavelet scalogram. The low-frequency spread, and persistence of the high-intensity frequency components are visible in BS sounds emanating from the uniform cross-sectional area of the trachea. The frictional rub between the pleurae causes a higher frequency spread of low-intensity intermittent frequency components in PS signals. From the complex networks of BS and PS, the extracted graph features are - graph density ([Formula: see text], transitivity ([Formula: see text], degree centrality ([Formula: see text]), betweenness centrality ([Formula: see text], eigenvector centrality ([Formula: see text]), and graph entropy (En). The high values of [Formula: see text] and [Formula: see text] show a strong correlation between distinct segments of the BS signal originating from a consistent cross-sectional tracheal diameter and, hence, the generation of high-intense low-spread frequency components. An intermittent low-intense and a relatively greater frequency spread in PS signal appear as high [Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text] values. With these complex network parameters as input attributes, the supervised machine learning techniques- discriminant analyses, support vector machines, k-nearest neighbors, and neural network pattern recognition (PRNN)- classify the signals with more than 90% accuracy, with PRNN having 25 neurons in the hidden layer achieving the highest (98.82%).

#2534 Classification of patients with acute kidney damage on the base of blood pressure value as a risk factor using a neural network

Abstract Background and Aims In acute kidney damage, the monitoring of hemodynamic status is best achieved by determining the blood pressure value. It is considered that the values of diastolic blood pressure in young and healthy people are constant from the aorta to the periphery, and the values of systolic and pulse pressure increase from the periphery to the aorta. The frequency of hypertension correlates with the presence of a cardiovascular precipitating risk factor in acute kidney injury, hypertension is also a modifying risk factor in the second and third stages of acute kidney injury, while the presence of hypotension on admission can be a clinical indicator of its cause. In patients with acute kidney damage and in patients with sepsis, the mechanism of autoregulation of mean arterial pressure, which is affected by stroke volume and heart rate, is not effective. The aim of the paper is to examine the classification of hospitalized patients with acute kidney injury based on blood pressure as a risk factor. Method The study included 86 patients over 18 years of age of both sexes (39 men 45.3% and 47 women 54.7%) with an average age of 66.85 ± 15.30 years who were diagnosed with acute kidney injury on admission to hospital treatment. Patients were divided into groups according to the stages of renal insufficiency. The study was designed as a cross-sectional comparative study. The assessment of the existence of acute kidney damage, as well as the determination of the stage of the disease, was based on the diagnostic criteria proposed by the K/DIGO expert group for acute kidney damage. The study did not include patients who had disease progression after acute kidney failure and were treated with dialysis for more than three months. Basic clinical, biochemical and hemodynamic parameters were determined in all subjects at the beginning of treatment, except for the analysis of nitrogen status, which was determined both before and at the end of treatment. During the formation of the neural network (pattern recognition Mat Lab), the parameters of systolic arterial pressure, diastolic arterial pressure normalized by min-max normalization to values from 0 to 1 were included. The data were randomly distributed into three categories: 80% for training, 10% for test, 10% for validation, which means that the parameters of 68 patients were used for training, 9 patients for validation and another 9 patients for the test. The number of input parameters is 33 to 39, while the number of output parameters is 2 in relation to treatment. Formed neural networks with different included blood pressure parameters, structures of 25 neurons in the hidden layer, with 55 and 15 epochs and with accuracy of 97.7% and 90.7% of patient classification. Results Our results show that 40.3% of patients in the third stage of acute kidney injury have hypertension. The lowest average values of systolic (108.44 ± 22.81 mmHg) and diastolic (65.56 ± 15.29 mmHg) blood pressure were experienced by patients in the first stage of acute kidney damage, and the lowest values of mean arterial pressure existed in the second stage of acute kidney damage (68.20 ± 44.96 mmHg) .Statistically significant risk factors for the development of the third stage of acute kidney damage in the multivariate analysis were patients older than 65 years, and mean arterial pressure values less than 65 mmHg. The confusion matrices showed that the classifiers adapted to the minimum and maximum values of systolic, diastolic and mean arterial blood pressure are well adapted and achieve satisfactory accuracy with 97.7% and 90.7%. Conclusion By comparing blood pressure parameters in hospitalized patients in relation to the stage of acute kidney damage, no statistically significant deviations were found. Univariate logistic regression analysis showed that age over 65 was a significant independent risk factor for stage three acute kidney injury, while the multivariate model showed that risk factors for the development of stage three acute kidney injury were patients same age and middle arterial blood pressure less than 65 mmHg. Satisfactory accuracy of formed neural networks when classifying patients during validation training and test was also demonstrated.

Indoor Infrastructure Maintenance Framework Using Networked Sensors, Robots, and Augmented Reality Human Interface

Sensing and cognition by homeowners and technicians for home maintenance are prime examples of human–building interaction. Damage, decay, and pest infestation present signals that humans interpret and then act upon to remedy and mitigate. The maintenance cognition process has direct effects on sustainability and economic vitality, as well as the health and well-being of building occupants. While home maintenance practices date back to antiquity, they readily submit to augmentation and improvement with modern technologies. This paper describes the use of networked smart technologies embedded with machine learning (ML) and presented in electronic formats to better inform homeowners and occupants about safety and maintenance issues, as well as recommend courses of remedial action. The demonstrated technologies include robotic sensing in confined areas, LiDAR scans of structural shape and deformation, moisture and gas sensing, water leak detection, network embedded ML, and augmented reality interfaces with multi-user teaming capabilities. The sensor information passes through a private local dynamic network to processors with neural network pattern recognition capabilities to abstract the information, which then feeds to humans through augmented reality and conventional smart device interfaces. This networked sensor system serves as a testbed and demonstrator for home maintenance technologies, for what can be termed Home Maintenance 4.0.

Predicting transcriptional activation domain function using Graph Neural Networks.

Analysis of factors that lead to the functionality of transcriptional activation domains remains a crucial and yet challenging task owing to the significant diversity in their sequences and their intrinsically disordered nature. Almost all existing methods that have aimed to predict activation domains have involved traditional machine learning approaches, such as logistic regression, that are unable to capture complex patterns in data or plain convolutional neural networks and have been limited in exploration of structural features. However, there is a tremendous potential in the inspection of the structural properties of activation domains, and an opportunity to investigate complex relationships between features of residues in the sequence. To address these, we have utilized the power of graph neural networks which can represent structural data in the form of nodes and edges, allowing nodes to exchange information among themselves. We have experimented with two kinds of graph formulations, one involving residues as nodes and the other assigning atoms to be the nodes. A logistic regression model was also developed to analyze feature importance. For all the models, several feature combinations were experimented with. The residue-level GNN model with amino acid type, residue position, acidic/basic/aromatic property and secondary structure feature combination gave the best performing model with accuracy, F1 score and AUROC of 97.9%, 71% and 97.1% respectively which outperformed other existing methods in the literature when applied on the dataset we used. Among the other structure-based features that were analyzed, the amphipathic property of helices also proved to be an important feature for classification. Logistic regression results showed that the most dominant feature that makes a sequence functional is the frequency of different types of amino acids in the sequence. Our results consistent have shown that functional sequences have more acidic and aromatic residues whereas basic residues are seen more in non-functional sequences.

CNN-Based Pattern Classifiers for Precise Identification of Perinatal EEG Biomarkers of Brain Injury in Preterm Neonates

Electroencephalographic (EEG) monitoring is important for the diagnosis of hypoxic-ischemic (HI) brain injury in high-risk preterm infants. EEG monitoring is limited by the reliance on expert clinical observation. However, high-risk preterm infants often do not present observable symptoms due to their frailty. Thus, there is an urgent need to find better ways to automatically quantify changes in the EEG these high-risk babies. This article is a first step towards this goal. This innovative study demonstrates the effectiveness of deep Convolutional Neural Networks (CNN) pattern classifiers, trained on spectrally-detailed Wavelet Scalograms (WS) images derived from neonatal EEG sharp waves—a potential translational HI biomarker, at birth. The WS-CNN classifiers exhibit outstanding performance in identifying HI sharp waves within an exclusive clinical EEG recordings dataset of preterm infants immediately after birth. The work has impact as it demonstrates exceptional high accuracy of 99.34 ± 0.51% cross-validated across 13,624 EEG patterns over 48 h raw EEG at low 256 Hz clinical sampling rates. Furthermore, the WS-CNN pattern classifier is able to accurately identify the sharp-waves within the most critical first hours of birth (n = 8, 4:36 ± 1:09 h), regardless of potential morphological changes influenced by different treatments/drugs or the evolutionary ‘timing effects’ of the injury. This underscores its reliability as a tool for the identification and quantification of clinical EEG sharp-wave biomarkers at bedside.

An APSO-based TPA-BLSTM model for predicting ship motion in irregular waves using wave-series input

ABSTRACT Combining the bidirectional long short-term memory neural network and temporal pattern attention mechanism, a machine learning model based on multi-feature inputs is developed to predict ship motion in irregular waves. An adaptive particle swarm optimisation algorithm is proposed to adapt the parameters of neural network to different feature inputs. The study indicates that the improved particle swarm optimisation algorithm avoids the problem of poor adaptability of empirical parameters to different advance time. The combination of bidirectional long short-term memory neural network and temporary pattern attention mechanism further enhances the data mining ability. Compared with the prediction mode solely based on the ship motion input, the inclusion of wave elevation input can effectively improve the accuracy and advance time for ship’s motion prediction. Importantly, the prediction accuracy will improve with an increased number of inputs of wave probes, which are arranged along the wave propagation direction.

Efficient water-related failure detection in PEM fuel cells: Combining a PEMFCs fractional order impedance model with FFT-PWM techniques and artificial neural network classification

Water management and early detection of faults in proton exchange membrane fuel cells (PEMFCs) are among the most critical constraints that limit the optimal spread of this type of energy. Consequently, it is necessary to enhance the reliability and durability of PEMFCs by developing an approach to diagnose and identify water failure modes. This paper proposes an effective and simple method to detect, diagnose, and classify various water failure modes in PEMFCs using a hybrid diagnostic approach. This approach combines the PEMFC fractional order impedance model (FOIM) with fast Fourier transform pulse width modulation (FFT-PWM) techniques and artificial neural network pattern recognition (ANN-PR) classification. The results show an accurate match between the electrochemical impedance spectroscopy (EIS) experimental data, the Nyquist impedance spectra of FOIM, and the FFT-PWM algorithm as a proposed alternative technique to EIS measurements. Learning of ANN-PR was performed using the frequency spectrum amplitude (FSA) database of the voltage and current signals produced by the PEMFCs FOIM DC/DC boost converter, which was generated using the FFT-PWM algorithm. The ANN-PR achieved low values for error accuracy, with the Low Square Error and Learning Error reaching 6.676 × 10−19 and 1.888 × 10−16, respectively. The elements inside the confusion matrix and the rest of the matrices confirm that the proposed model's accuracy, precision, recall, and high F1 score reached 100%. Furthermore, all predictions made by the ANN-PR model were consistently accurate across all areas of failure detection. Overall, the proposed method helps in analyzing, diagnosing, and classifying fuel cell failure modes such as flooding and drying, which may simplify the health assessment of PEMFC.

Mitigating Information Interruptions by COVID-19 Face Masks: A Three-Stage Speech Enhancement Scheme

The coronavirus disease 2019 (COVID-19) preventive measures have resulted in significant lifestyle changes. One of the COVID-19 new normal is the usage of face masks for protection against airborne aerosol which creates distractions and interruptions in voice communication. It has a different influence on speech than the standard concept of noise affecting speech communication. Furthermore, it has varied effects on speech in different frequency bands. To provide a solution to this problem, a three-stage adaptive speech enhancement (SE) scheme is developed in this article. In the first stage, the tunable <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$</tex-math> </inline-formula> -factor wavelet transform (TQWT) features are extracted by properly setting the quality factor values and the number of levels from the input speech signal. In the second stage, the adjustable parameters of the preemphasis filter and modified multiband spectral subtraction (MBSS) are determined using bio-inspired techniques for different masking and signal-to-noise ratio (SNR) conditions. In the third stage, the weights, center values, standard deviation of the Gaussian radial basis functions, and input patterns of the radial basis function neural networks (RBFNNs) are updated to predict the optimized parameters from the input TQWT-based cepstral features (TQCFs). In the end, the performance of the proposed algorithm is compared with the standard SE algorithms using two speech datasets.

Improving the Accuracy of Neural Network Pattern Recognition by Fractional Gradient Descent

Improving the Accuracy of Neural Network Pattern Recognition by Fractional Gradient Descent

Diagnostic Method for the Disease Severity of Powdery Mildew in Pepper Leaves Based on Feature Transfer Learning Model

HighlightsPowdery mildew diagnostic system has been established for identifying diseased leaves in the complex backgrounds.The degree of disease has been quantified to improve the computational efficiency of the recognition model.A diagnostic system for early powdery mildew was constructed by combining feature transfer learning algorithms.The recognition system of disease degree realizes the precise diagnosis of early powdery mildew in the field.Abstract. Combining image processing technology to achieve real-time and accurate diagnosis of the powdery mildew can effectively improve the treatment efficiency of pepper diseases. Due to the influence of weeds and soil, capturing pepper disease information analyzed by traditional image processing algorithms in the field is laborious. From this perspective, the diagnostic system of powdery mildew has been proposed to automatically identify the different disease degrees in the initial stage of powdery mildew. The image segmentation algorithm of four-dimensional adaptive clustering was used to detect pepper leaves in complex backgrounds. The impact of each disease region on the accuracy of the recognition model was evaluated based on energy, entropy, contrast, and correlation. Combined with a neural network pattern recognition system and a quantitative conjugate gradient training algorithm, the fusion algorithm was evaluated on a dataset of 35577 images taken in the field with different distinct time intervals, angles, and lighting conditions. The diagnosis model was used to diagnose the disease grade of two peppers with different disease degrees. The collected images of the whole pepper plants were scanned and measured by a linear array. The experimental results showed that 66.34% of the leaves of the early disease plants were classified as disease level 5, and 65.88% of the leaves of the late disease plants were classified as disease level 10. In conclusion, this method can realize the intelligent recognition of pepper disease degree efficiently and accurately. Keywords: Diagnosis model, Feature transfer learning, Fusion algorithm, Powdery mildew.

Interpreting Neural Network Pattern With Pruning for PolSAR Target Recognition

Interpreting Neural Network Pattern With Pruning for PolSAR Target Recognition

Postsynaptic cell type and synaptic distance do not determine efficiency of monosynaptic rabies virus spread measured at synaptic resolution.

Retrograde monosynaptic tracing using glycoprotein-deleted rabies virus is an important component of the toolkit for investigation of neural circuit structure and connectivity. It allows for the identification of first-order presynaptic connections to cell populations of interest across both the central and peripheral nervous system, helping to decipher the complex connectivity patterns of neural networks that give rise to brain function. Despite its utility, the factors that influence the probability of transsynaptic rabies spread are not well understood. While it is well established that expression levels of rabies glycoprotein used to trans-complement G-deleted rabies can result in large changes in numbers of inputs labeled per starter cell (convergence index [CI]), it is not known how typical values of CI relate to the proportions of synaptic contacts or input neurons labeled. And it is not known whether inputs to different cell types, or synaptic contacts that are more proximal or distal to the cell body, are labeled with different probabilities. Here, we use a new rabies virus construct that allows for the simultaneous labeling of pre- and postsynaptic specializations to quantify the proportion of synaptic contacts labeled in mouse primary visual cortex. We demonstrate that with typical conditions about 40% of first-order presynaptic excitatory synapses to cortical excitatory and inhibitory neurons are labeled. We show that using matched tracing conditions there are similar proportions of labeled contacts onto L4 excitatory pyramidal, somatostatin (Sst) inhibitory, and vasoactive intestinal peptide (Vip) starter cell types. Furthermore, we find no difference in the proportions of labeled excitatory contacts onto postsynaptic sites at different subcellular locations.

Neural Network Classification in Javanese Handwriting Recognition using Projection Profile Histogram and Local Binary Pattern Histogram

Indonesia consists of various regional tribes, where each tribe has cultural diversity and some even have their own regional letters, like Javanese tribe has Javanese characters. Javanese letters consist of 20 basic letters called Nglegena script. Subject about Javanese language is delivered to elementary student until now aims to preserve Indonesian culture especially the Javanese. In this study, we present two feature extraction methods are Local Binary Pattern (LBP) and Profile Projection (PP). Neural Network (NN) chosen as classifiers for classifying 20 javanese letters Nglegena. Some digital image processing processes are carried out, are image inversion, dilation, denoising and skeletoning. The Javanese script dataset is taken from the Kaggle database with the name Aksara Jawa: Aksara Jawa Custom Dataset, consists of 2154 train images and 480 test images. The experiment were carried out in two models, Projection Profile Histogram - Neural Network (PPH-NN) and Local Binary Pattern Histogram - Neural Network (LBPH-NN). The experiment show that both feature extraction methods have very good performance, 99.98% PPH-NN and 89.6% LBPH-NN on average.

The Enhanced Performance of Neuromorphic Computing Hardware in an ITO/ZnO/HfOx/W Bilayer-Structured Memory Device.

This study discusses the potential application of ITO/ZnO/HfOx/W bilayer-structured memory devices in neuromorphic systems. These devices exhibit uniform resistive switching characteristics and demonstrate favorable endurance (>102) and stable retention (>104 s). Notably, the formation and rupture of filaments at the interface of ZnO and HfOx contribute to a higher ON/OFF ratio and improve cycle uniformity compared to RRAM devices without the HfOx layer. Additionally, the linearity of potentiation and depression responses validates their applicability in neural network pattern recognition, and spike-timing-dependent plasticity (STDP) behavior is observed. These findings collectively suggest that the ITO/ZnO/HfOx/W structure holds the potential to be a viable memory component for integration into neuromorphic systems.