Network Pattern Recognition Research Articles

Light-to-Spike Encoding Using Indium-Gallium-Zinc Oxide Phototransistor for all-Color Image Recognition with Dynamic Range and Precision Tunability.

To enhance the efficiency of machine vision system, physical hardware capable of sensing and encoding is essential. However, sensing and encoding color information has been overlooked. Therefore, this work utilizes an indium-gallium-zinc oxide (IGZO) phototransistor to detect varying densities of red, green, and blue (RGB) light, converting them into corresponding drain current (ID) states. By applying stochastic gate voltage (VG) pulses to the IGZO phototransistor, the fluctuations are generated in these ID states. When the ID exceeds the threshold current (ITC), a spike signal is generated. This approach enables the conversion of light densities into spike signals, achieving spike-rate encoding. Moreover, adjusting the standard deviation (σ) of the VG pulses controls the range of light densities converted into spike rates, while altering the mean (μ) of the VG pulses changes the baseline level of spike rates. Remarkably, separate RGB channels offer a tunable encoding process, which can emphasize individual colors and correct color bias. The encoded spike rates are also fed into a spiking neural network (SNN) for CIFAR-10 pattern recognition, achieving an accuracy of 86%. The method allows the operation of SNN and shows the tunability in the process of light-to-spike encoding, opening possibilities for color image processing.

Integration of Sensory Memory Process Display System for Gait Recognition

AbstractGait is among the most dependable, accurate, and secure methods of biometric identification. However, high power consumption and low computing capability are two major obstacles on wearable sensors‐based gait recognition system. In this work, an integrated system is reported combining a triboelectric nanogenerator (TENG), a memristor (Ag/HfOx/Pt), and perovskite‐based multicolor LEDs (PMCLED) for the visualization and recognition of foot patterns through signal‐on‐none and multi‐wavelength on‐device preprocessing. The flexible TENG acts as a sensory receptor, generating voltage based on the duration and intensity of pressure, which in turn promotes voltage‐triggered synaptic plasticity in the memristor. The PMCLED, with its threshold switching and multi‐wavelength emission characteristics, enables nonlinear filtering and amplification of the synaptic signal from the memristor, resulting in a simplified system design and reduced background noise. Additionally, the effectiveness of on‐device preprocessing is validated based on a 5 × 5 array of integrated devices and software‐built neural network for foot pattern visualization and recognition. The proposed system is able to recognize the on‐device preprocessed images with high accuracy, indicating great potentials in both healthcare monitoring and human‐machine interaction.

Neurons to heartbeats: spiking neural networks for electrocardiogram pattern recognition

Neurons to heartbeats: spiking neural networks for electrocardiogram pattern recognition

Medical diagnosis based on artificial intelligence and decision support system in the management of health development.

Medical diagnosis plays a critical role in our daily lives. Every day, over 10 billion cases of both mental and physical health disorders are diagnosed and reported worldwide. To diagnose these disorders, medical practitioners and health professionals employ various assessment tools. However, these tools often face scrutiny due to their complexity, prompting researchers to increase their experimental parameters to provide accurate justifications. Additionally, it is essential for professionals to properly justify, interpret, and analyse the results from these prediction tools. This research paper explores the use of artificial intelligence and advanced analytics in developing Clinical Decision Support Systems (CDSS). These systems are capable of diagnosing and detecting patterns of various medical disorders. Various machine learning algorithms contribute to building these assessment tools, with the Network Pattern Recognition (NEPAR) algorithm being the first to aid in developing CDSS. Over time, researchers have recognised the value of machine learning-based prediction models in successfully justifying medical diagnoses. The proposed CDSS models have demonstrated the ability to diagnose mental disorders with an accuracy of up to 89% using only 28 questions, without requiring human input. For physical health issues, additional parameters are used to enhance the accuracy of CDSS models. Consequently, medical professionals are increasingly relying on these machine learning-based CDSS models, utilising these tools to improve medical diagnosis and assist in decision-making. The different cross-validation values are considered to remove the data biasness.

Transformative insights: Image-based breast cancer detection and severity assessment through advanced AI techniques

Abstract In the realm of image-based breast cancer detection and severity assessment, this study delves into the revolutionary potential of sophisticated artificial intelligence (AI) techniques. By investigating image processing, machine learning (ML), and deep learning (DL), the research illuminates their combined impact on transforming breast cancer diagnosis. This integration offers insights into early identification and precise characterization of cancers. With a foundation in 125 research articles, this article presents a comprehensive overview of the current state of image-based breast cancer detection. Synthesizing the transformative role of AI, including image processing, ML, and DL, the review explores how these technologies collectively reshape the landscape of breast cancer diagnosis and severity assessment. An essential aspect highlighted is the synergy between advanced image processing methods and ML algorithms. This combination facilitates the automated examination of medical images, which is crucial for detecting minute anomalies indicative of breast cancer. The utilization of complex neural networks for feature extraction and pattern recognition in DL models further enhances diagnostic precision. Beyond diagnostic improvements, the abstract underscores the substantial influence of AI-driven methods on breast cancer treatment. The integration of AI not only increases diagnostic precision but also opens avenues for individualized treatment planning, marking a paradigm shift toward personalized medicine in breast cancer care. However, challenges persist, with issues related to data quality and interpretability requiring continued research efforts. Looking forward, the abstract envisions future directions for breast cancer identification and diagnosis, emphasizing the adoption of explainable AI techniques and global collaboration for data sharing. These initiatives promise to propel the field into a new era characterized by enhanced efficiency and precision in breast cancer care.

A robust myoelectric pattern recognition framework based on individual motor unit activities against electrode array shifts

Background and objectiveElectrode shift is always one of the critical factors to compromise the performance of myoelectric pattern recognition (MPR) based on surface electromyogram (SEMG). However, current studies focused on the global features of SEMG signals to mitigate this issue but it is just an oversimplified description of the human movements without incorporating microscopic neural drive information. The objective of this work is to develop a novel method for calibrating the electrode array shifts toward achieving robust MPR, leveraging individual motor unit (MU) activities obtained through advanced SEMG decomposition. MethodsAll of the MUs from decomposition of SEMG data recorded at the original electrode array position were first initialized to train a neural network for pattern recognition. A part of decomposed MUs could be tracked and paired with MUs obtained at the original position based on spatial distribution of their MUAP waveforms, so as to determine the shift vector (describing both the orientation and distance of the shift) implicated consistently by these multiple MU pairs. Given the known shift vector, the features of the after-shift decomposed MUs were corrected accordingly and then fed into the network to finalize the MPR task. The performance of the proposed method was evaluated with data recorded by a 16 × 8 electrode array placed over the finger extensor muscles of 8 subjects performing 10 finger movement patterns. ResultsThe proposed method achieved a shift detection accuracy of 100 % and a pattern recognition accuracy approximating to 100 %, significantly outperforming the conventional methods with lower shift detection accuracies and lower pattern recognition accuracies (p < 0.05). ConclusionsOur method demonstrated the feasibility of using decomposed MUAP waveforms’ spatial distributions to calibrate electrode shift. This study provides a new tool to enhance the robustness of myoelectric control systems via microscopic neural drive information at an individual MU level.

Fall-related gait pattern recognition based on surface electromyography using a hybrid neural network with transfer learning

Gait analysis across different individuals presents significant challenges due to individual variability. In this study, we developed a gait analysis system to enable continuous decoding of gait patterns based on surface electromyography (EMG). Our objective was to address the long-term challenges associated with neural networks, particularly their limited generalization capability and the heavy training burden on new subjects in gait pattern recognition. Fifteen healthy subjects were recruited to imitate two fall-related gait patterns and collect myoelectric signals from eight lower limb muscles. We proposed a transfer learning-based framework and investigated eight adaptation schemes using varying quantities of target training samples. In the cross-subject evaluation, our framework demonstrated robust generalization capability and reduced training burden. Even with a limited number of target training samples, the framework outperformed state-of-the-art machine learning models and discriminative neural networks that have previously shown excellent recognition performance in subject-specific evaluation. This study highlights the promising potential of using transfer learning in conjunction with neural networks for gait pattern recognition across different individuals. It also provides valuable insights into the diagnosis, monitoring, and treatment of various medical conditions using surface EMG-based gait analysis systems.

Enhancing Heart Attack Prediction Accuracy through Optimized Machine Learning and Deep Learning: A Survey

Heart attacks are a leading cause of mortality worldwide. Early and accurate prediction can significantly improve patient outcomes. This paper explores the potential of optimized machine learning and deep learning techniques to enhance heart attack prediction accuracy. We discuss the challenges associated with traditional methods and propose a framework that leverages advancements in machine learning and deep learning. The framework may involve techniques like data pre-processing, feature selection, hyperparameter tuning, and ensemble methods to optimize the performance of machine learning algorithms like random forests and support vector machines. Additionally, it may explore deep learning architectures like convolutional neural networks or recurrent neural networks for feature extraction and pattern recognition from complex medical data. The goal is to achieve a more robust and accurate prediction model for heart attacks. This can empower healthcare professionals to identify high-risk individuals, allowing for preventative measures and early intervention, ultimately saving lives. The paper will delve into the specific methods employed for optimization, evaluate their effectiveness, and discuss the potential impact on improving heart attack prediction accuracy

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.

Adaptive Artificial Intelligence Techniques for QoS and Congestion Management in 5G Wireless Networks: A Literature

In the rapidly evolving landscape of 5G wireless networks, ensuring Quality of Service (QoS) and managing congestion effectively are critical challenges that must be addressed to meet the stringent performance requirements of modern applications. This paper presents an in-depth study on the application of adaptive artificial intelligence (AI) techniques for QoS and congestion management in 5G wireless networks. By leveraging machine learning (ML) and deep learning (DL) algorithms, we propose an intelligent framework that dynamically adjusts network parameters in real-time to optimize performance and mitigate congestion. The proposed approach integrates reinforcement learning for adaptive decision-making, convolutional neural networks (CNNs) for traffic pattern recognition, and long short-term memory (LSTM) networks for predictive analysis of network congestion trends. Through extensive simulations and real-world testing, our framework demonstrates significant improvements in QoS metrics such as latency, throughput, and packet loss, while efficiently managing congestion across diverse network scenarios. The results indicate that adaptive AI techniques hold immense potential in enhancing the robustness and efficiency of 5G wireless networks, paving the way for more reliable and high-performance communication systems. KEYWORDS: artificial intelligence (AI), machine learning (ML), convolutional neural networks (CNNs), and long short-term memory (LSTM) networks

Predicting New Single/Multiphase-Structure High-Entropy Alloys Using a Pattern Recognition Network

Machine learning methods were employed to predict the phase structures of high-entropy alloys (HEAs). These alloys were classified into four categories: bcc (body-centered cubic), fcc (face-centered cubic), bcc+fcc (body-centered cubic and face-centered cubic) and others (containing intermetallic compounds and other structural alloys). The utilized algorithm was a Pattern Recognition Network (PRN) utilizing cross-entropy as the loss function, enabling the prediction of HEAs’ phase formation probability. The PRN algorithm demonstrated an accuracy exceeding 87% based on the test data. The PRN algorithm successfully predicted the transformation from fcc to fcc+bcc and subsequently to a bcc structure with the increase in Al content in AlxCoCu6Ni6Fe6 and AlxCoCrCuNiFe HEAs. In addition, AlxCoCu6Ni6Fe6 (x = 1, 3, 6, 9) HEAs were prepared using a vacuum arc furnace, and the microstructure of the as-cast alloy was tested by means of XRD, SEM, and EBSD, confirming the high consistency between the predicted and observed phase structures. This study showcases the efficacy of the PRN algorithm in predicting both single- and multiphase-structure high-entropy alloys, offering valuable insights into alloy design and development.

Bearing-Fault-Feature Enhancement and Diagnosis Based on Coarse-Grained Lattice Features.

In view of the frequent failures occurring in rolling bearings, the strong background noise present in signals, weak features, and difficulties associated with extracting fault characteristics, a method of enhancing and diagnosing rolling bearing faults based on coarse-grained lattice features (CGLFs) is proposed. First, the vibrational signals of bearings are subjected to adaptive filtering to eliminate background noise. Second, frequency-domain transformation is performed, and a coarse-grained approach is used to continuously segment the spectrum. Within each segment, amplitude-enhancement operations are executed, transforming the data into a CGLF graph that enhances fault characteristics. This graph is then fed into a Swin Transformer-based pattern-recognition network. Third and finally, a high-precision fault diagnosis model is constructed using fully connected layers and Softmax, enabling the diagnosis of bearing faults. The fault recognition accuracy reaches 98.30% and 98.50% with public datasets and laboratory data, respectively, thereby validating the feasibility and effectiveness of the proposed method. This research offers an efficient and feasible fault diagnosis approach for rolling bearings.

Fault Detection and Normal Operating Condition in Power Transformers via Pattern Recognition Artificial Neural Network

Aging, degradation, or damage to internal insulation materials often contribute to transformer failures. Furthermore, combustible gases can be produced when these insulation materials experience thermal or electrical stresses. This paper presents an artificial neural network for pattern recognition (PRN) to classify the operating conditions of power transformers (normal, thermal faults, and electrical faults) depending on the combustible gases present in them. Two network configurations were presented, one with five and the other with ten neurons in the hidden layer. The main advantage of applying this model through artificial neural networks is its ability to capture the nonlinear characteristics of the samples under study, thus avoiding the need for iterative procedures. The effectiveness and applicability of the proposed methodology were evaluated on 815 real data samples. Based on the results, the PRN performed well in both training and validation (for samples that were not part of the training), with a mean squared error (MSE) close to expected (0.001). The network was able to classify the samples with a 98% accuracy rate of the 815 samples presented and with 100% accuracy in validation, showing that the methodology developed is capable of acting as a tool for diagnosing the operability of power transformers.

Visualized in-sensor computing

In artificial nervous systems, conductivity changes indicate synaptic weight updates, but they provide limited information compared to living organisms. We present the pioneering design and production of an electrochromic neuromorphic transistor employing color updates to represent synaptic weight for in-sensor computing. Here, we engineer a specialized mechanism for adaptively regulating ion doping through an ion-exchange membrane, enabling precise control over color-coded synaptic weight, an unprecedented achievement. The electrochromic neuromorphic transistor not only enhances electrochromatic capabilities for hardware coding but also establishes a visualized pattern-recognition network. Integrating the electrochromic neuromorphic transistor with an artificial whisker, we simulate a bionic reflex system inspired by the longicorn beetle, achieving real-time visualization of signal flow within the reflex arc in response to environmental stimuli. This research holds promise in extending the biomimetic coding paradigm and advancing the development of bio-hybrid interfaces, particularly in incorporating color-based expressions.

Using machine learning to predict artistic styles: an analysis of trends and the research agenda

AbstractIn the field of art, machine learning models have been used to predict artistic styles in paintings. The foregoing is somewhat advantageous for analysts, as these tools can provide more valuable results and help reduce bias in the results and conclusions provided. Therefore, the objective of this research was to examine research trends in the use of machine learning to predict artistic styles from a bibliometric review based on the PRISMA methodology. From the search equations, 268 documents were found, out of which, following the application of inclusion and exclusion criteria, 128 documents were analyzed. Through quantitative analysis, a growing research interest in the subject is evident, progressing from user perception approaches to the utilization of tools like deep learning for art studies. Among the main results, it is possible to identify that one of the most used techniques in the field has been neural networks for pattern recognition. Also, a large part of the research focuses on the use of design software for image creation and manipulation. Finally, it is found that the number of studies focused on contemporary modern art is still limited, this is due to the fact that a large part of the investigations has focused on historical artistic styles.

Model compression of deep neural network architectures for visual pattern recognition: Current status and future directions

Visual Pattern Recognition Networks (VPRNs) are widely used in various visual data based applications such as computer vision and edge AI. VPRNs help to enhance a machine's learning by extracting underlying patterns from input images and videos. VPRNs can be implemented using statistical, syntactical, or Machine Learning (ML) based approaches. With the popularity of Deep Neural Networks (DNNs) in the recent past, DNN based VPRNs have become an inevitable choice due to their suitability for handling such high dimensional data.However, such DNN based VPRNs bring along a curse of dimensionality, leading to intricate computations, substantial memory demands, and increased energy requirements. This impedes their practical deployment in resource constrained and strict latency required environments. Such overheads impose a demand for compression of VPRNs without impairing their performance. This research presents an exhaustive survey on compression techniques used for sequential, non-sequential and advanced DNN architectures from the perspective of Visual Pattern Recognition (VPR). Research findings of this study reveal potential scope to compress non-sequential, sequential and advanced DNN architectures for VPR applications. The study also clarifies challenges, reported and unreported issues as well as research gaps and provides directions for further research in this domain.

Sequence Modeling with Recurrent Neural Networks (RNNs) for Student Learning Behavior Pattern Recognition in a Flipped Classroom

The flipped classroom model has become increasingly popular in education, altering the traditional methods of teaching. It is important to understand and acknowledge how students learn within this framework in order to optimize instructional strategies and promote personalized learning. This study investigates the use of Sequence Modeling with Recurrent Neural Networks (RNNs) to identify patterns in student learning behavior within a flipped classroom setting. The proposed deep learning architecture utilizes RNNs to analyze sequential patterns in students' interactions with the flipped classroom materials, while also incorporating attention mechanisms to better detect important patterns and temporal dynamics in the learning process. Multimodal learning techniques are also employed, combining data from various sources to gain a comprehensive understanding of student behavior. Additionally, clustering techniques using autoencoders are explored to group students with similar learning behaviors. Predictive models, such as RNN or LSTM networks, are developed to forecast future learning behaviors and provide insights into potential challenges or successes for individual students. The effectiveness of this framework is evaluated using real-world data from flipped classroom implementations, with performance metrics like recall, precision, and accuracy used to assess the success of the sequence modeling approach in recognizing and predicting student behavior patterns. Overall, the application of deep learning methods, specifically sequence modeling with RNNs, demonstrates potential for improving personalized learning experiences and facilitating proactive interventions to support diverse student needs.

Adaptive fuzzy energy management strategy for range‐extended electric vehicles integrated with deep learning

AbstractA novel framework for adaptive energy management, rooted in deep learning principles, is proposed to minimize fuel consumption in extended‐range electric vehicles amidst intricate driving scenarios. This innovative approach integrates a long short‐term memory (LSTM) network for pattern recognition across three driving patterns and an adaptive fuzzy controller. To mitigate the impact of poor hyperparameter selection on recognition accuracy, Gray Wolf Optimization is employed to optimize the hidden layer nodes, training times, and learning rate of the LSTM. Simultaneously, a genetic algorithm is utilized to optimize the vertex coordinates of the fuzzy control membership function, enabling the adaptive adjustment of parameters in the fuzzy energy management strategy. The condition recognition model accurately identifies the vehicle's driving status and seamlessly transitions to an energy management strategy tailored to the present conditions. This ensures optimal operation, enhancing overall fuel efficiency and performance. The simulation results robustly validate the efficacy of this approach: the GWO‐LSTM network achieves an impressive 97.7% accuracy in recognizing working conditions, surpassing the 88.9% accuracy of the traditional LSTM network. Furthermore, the fuel consumption reduction achieved by the adaptive fuzzy energy management strategy amounts to 11.9% compared with the conventional fuzzy energy management approach. This outcome underscores the tangible enhancement in vehicle fuel economy resulting from the seamless integration of deep learning techniques.

An AI-Enabled Blockchain Algorithm: A Novel Approach to Counteract Blockchain Network Security Attacks

INTRODUCTION: In this research, we present a novel method for strengthening the security of blockchain networks through the use of AI-driven technology. Blockchain has emerged as a game-changing technology across industries, but its security flaws, particularly in relation to Sybil and Distributed Denial of Service (DDoS) attacks, are a major cause for worry. To defend the blockchain from these sophisticated attacks, our research centres on creating a strong security solution that combines networks of Long Short-Term Memory (LSTM) and Self-Organizing Maps (SOM).&#x0D; OBJECTIVES: The main goal of this project is to create and test an AI-driven blockchain algorithm that enhances blockchain security by utilising LSTM and SOM networks. These are the objectives that the research hopes to achieve: In order to assess the shortcomings and weaknesses of existing blockchain security mechanisms. The goal is to create a new approach that uses LSTM sequence learning and SOM pattern recognition to anticipate and stop security breaches. In order to see how well this integrated strategy works in a simulated blockchain setting against different types of security risks.&#x0D; METHODS: The methods used in our study are based on social network analysis. A combination of support vector machines (SOM) for pattern recognition and long short-term memory (LSTM) networks for learning and event sequence prediction using historical data constitutes the methodology. The steps involved in conducting research are: The current state of blockchain security mechanisms is examined in detail. Creating a virtual blockchain and incorporating the SOM+LSTM algorithm.&#x0D; Putting the algorithm through its paces in order to see how well it detects and defends against different security risks.&#x0D; RESULTS: Significant enhancements to blockchain network security are the primary outcomes of this study. Important results consist of: Using the SOM+LSTM technique, we were able to increase the detection rates of possible security risks, such as Sybil and DDoS attacks. Enhanced reaction times when compared to conventional security techniques for attack prediction and prevention. Demonstrated ability of the algorithm to adapt and learn from new patterns of attacks, assuring long-term sustainability.&#x0D; CONCLUSION: This paper's findings highlight the efficacy of enhancing blockchain security through the integration of artificial intelligence technologies such as LSTM and SOM networks. In addition to improving blockchain technology's detection and forecasting capabilities, the SOM+LSTM algorithm helps advance the platform toward greater security and reliability. This study provides a solid answer to the increasing worries about cyber dangers in the modern era and opens the door to more sophisticated AI uses in blockchain security.

Метод синтеза данных для повышения эффективности обучения нейронных сетей

The article is devoted to the formation of the initial data set for training neural networks. A proven method for creating synthetic data based on a graphics processor using a graphics pipeline is presented. A distinctive feature of this method is its modular architecture, which makes it easy to modify, delete or add individual stages to the synthetic image generation pipeline. A one-stage automatic detector based on a convolutional neural network of the Yolo type has been trained, the quality of the trained model and the operation of the recognition algorithm were also evaluated. Conclusions are drawn regarding the possibility of using this approach when creating representative samples of a large volume and their further use for training neural networks in pattern recognition.