Neural Network Analysis Research Articles

Application of Deep Learning for the Detection of Genetic Variations: Its Implementation in Classifying Alzheimer's Disease

Deep learning emerges as a promising technique, utilizing nonlinear transformations for feature extraction from high-dimensional datasets. However, its application encounters challenges in genome-wide association studies (GWAS) dealing with high-dimensional genomic data. This study introduces an innovative three-step method termed SWAT-CNN for the identification of genetic variants. This approach employs deep learning to pinpoint phenotype-related single nucleotide polymorphisms (SNPs), facilitating the development of precise disease classification models. In the first step, the entire genome undergoes division into non overlapping fragments of an optimal size. Subsequently, convolutional neural network (CNN) analysis is conducted on each fragment to identify phenotype-associated segments. The second step, employs a Sliding Window Association Test (SWAT), where CNN is utilized on the selected fragments to compute phenotype influence scores (PIS) and detect phenotype-associated SNPs based on these scores. The third step involves running CNN on all identified SNPs to construct a comprehensive classification model. Validation of the proposed approach utilized GWAS data from the Alzheimer’s disease Neuroimaging Initiative (ADNI), encompassing 981 subjects, including cognitively normal older adults (CN) and individuals with Alzheimer's disease (AD). Notably, the method successfully identified the widely recognized APOE region as the most significant genetic locus for AD. The resulting classification model exhibited an area under the curve (AUC) of 0.82, demonstrating compatibility with traditional machine learning approaches such as random forest and XGBoost. SWAT-CNN, as a groundbreaking deep learning-based genome-wide methodology, not only identified AD-associated SNPs but also presented a robust classification model for Alzheimer's disease, suggesting potential applications across diverse biomedical domains.

Regression and Neural Network Analysis of the System of Cardiorespiratory Synchronism Parameters to Predict the Minimum Limit of the Synchronization Range

In order to improve the existing step-by-step principle of determining the limits of the range of cardiorespiratory synchronization (CRS) and the regulatory and adaptive capabilities of the human body, a regression analysis of measurements of the CRS parameters system was carried out. Statistical prediction of the minimum limit of the synchronization range in the CRS method was performed using linear, nonlinear multiple regression models, as well as neural network regression. The following parameters were used as the basis for predicting the minimum limit of the synchronization range: initial heart rate, initial respiratory rate, age and gender. The variable initial heart rate contributed the most to all regression models. A comparative analysis of the quality indicators of all models showed that the prediction of the minimum limit of the synchronization range can be carried out both by linear multiple regression and using neural network regression implemented by means of a 3-4-1 multilayer perceptrons. However, having already had a NN-regression forecast of maximum limit of the synchronization range, as well as a NN-classification of the level of regulatory-adaptive status, the use of NN-regression will be proposed as the main forecasting method for its implementation in the software of the CRS measurement system.

Optimizing bone and biomass co-torrefaction parameters: High-performance arsenic removal from wastewater via co-torrefied bone char

This study aimed to investigate bone char's physicochemical transformations through co-torrefaction and co-pyrolysis processes with biomass. Additionally, it aimed to analyze the carbon sequestration process during co-torrefaction of bone and biomass and optimize the process parameters of co-torrefaction. Finally, the study sought to evaluate the arsenic sorption capacity of both torrefied and co-torrefied bone char. Bone and biomass co-torrefaction was conducted at 175 °C–300 °C. An orthogonal array of Taguchi techniques and artificial neural networks (ANN) were employed to investigate the influence of various torrefaction parameters on carbon dioxide sequestration within torrefied bone char. A co-torrefied bone char, torrefied at a reaction temperature of 300 °C, a heating rate of 15 °C·min−1, and mixed with 5 g m of biomass (wood dust), was selected for the arsenic (III) sorption experiment due to its elevated carbonate content. The results revealed a higher carbonate fraction (21%) in co-torrefied bone char at 300 °C compared to co-pyrolyzed bone char (500–700 °C). Taguchi and artificial neural network (ANN) analyses indicated that the relative impact of process factors on carbonate substitution in bone char followed the order of co-torrefaction temperature (38.8%) > heating rate (31.06%) > addition of wood biomass (30.1%). Co-torrefied bone chars at 300 °C exhibited a sorption capacity of approximately 3 mg g−1, surpassing values observed for pyrolyzed bone chars at 900 °C in the literature. The findings suggest that co-torrefied bone char could serve effectively as a sorbent in filters for wastewater treatment and potentially fulfill roles such as a remediation agent, pH stabilizer, or valuable source of biofertilizer in agricultural applications.

Artificial neural network analysis of the flow of nanofluids in a variable porous gap between two inclined cylinders for solar applications

Copper (Cu) nanoparticles (NPs) and polyvinyl alcohol (PVA) are utilized to enhance heat transfer (HT) which is used in the efficiency of solar energy systems. Copper nanoparticles have excellent thermal conductivity (TC) properties that enable them to conduct heat efficiently. In this arrangement, the gap between the two cylindrical channels is settled for the hybrid nanofluids (HNFs) flow in an inclined position that is favourable to sunlight. The nanomaterials consist of a mixture of PVA and Cu nanoparticles (NPs), to execute HNFs. Solar radiation is present on the hot side of the system. The porous gap between the two channels is considered variable which plays a crucial role in enhancing heat transfer and energy conversion. The varying permeability of the gap is adjusted to control the flow resistance and improve the stability of the system. It is observed that higher porosity allows for better convective heat transfer and reduced pressure drop. The transformed equations are solved through an artificial neural network (ANN) while the control volume finite element method (CVFEM) is also used to handle the governing equations. The Cu and PVA (HNF) improves solar radiation absorption and protects components, ultimately increasing the performance and efficiency of the systems.

Mittag‐Leffler stability of neural networks with Caputo–Hadamard fractional derivative

In this paper, a Hopfield‐type neural network system with Caputo–Hadamard fractional derivative is discussed. The importance of the existence of the equilibrium point in the analysis of artificial neural networks is well known. Another important investigation is the stability properties. So, the stability of a neural network system is dealt with in the present paper. First, a theorem that asserts the existence and uniqueness of the equilibrium point of the system is proven. Later, the conditions that ensure the Mittag‐Leffler stability of the equilibrium point is obtained by using the Lyapunov's direct method. In addition, an example is given with numerical simulations to show the effectiveness of our theoretical results.

Predicting travel mode choice with a robust neural network and Shapley additive explanations analysis

AbstractPredicting and understanding travellers’ mode choices is crucial to developing urban transportation systems and formulating traffic demand management strategies. Machine learning (ML) methods have been widely used as promising alternatives to traditional discrete choice models owing to their high prediction accuracy. However, a significant body of ML methods, especially the branch of neural networks, is constrained by overfitting and a lack of model interpretability. This study employs a neural network with feature selection for predicting travel mode choices and Shapley additive explanations (SHAP) analysis for model interpretation. A dataset collected in Chengdu, China was used for experimentation. The results reveal that the neural network achieves commendable prediction performance, with a 12% improvement over the traditional multinomial logit model. Also, feature selection using a combined result from two embedded methods can alleviate the overfitting tendency of the neural network, while establishing a more robust model against redundant or unnecessary variables. Additionally, the SHAP analysis identifies factors such as travel expenditure, age, driving experience, number of cars owned, individual monthly income, and trip purpose as significant features in our dataset. The heterogeneity of mode choice behaviour is significant among demographic groups, including different age, car ownership, and income levels.

Dissipativity Analysis of Memristive Inertial Competitive Neural Networks with Mixed Delays

Without altering the inertial system into the two first-order differential systems, this paper primarily works over the global exponential dissipativity (GED) of memristive inertial competitive neural networks (MICNNs) with mixed delays. For this purpose, a novel differential inequality is primarily established around the discussed system. Then, by applying the founded inequality and constructing some novel Lyapunov functionals, the GED criteria in the algebraic form and the linear matrix inequality (LMI) form are given, respectively. Furthermore, the estimation of the global exponential attractive set (GEAS) is furnished. Finally, a specific illustrative example is analyzed to check the correctness and feasibility of the obtained findings.

Multi-task learning for IoT traffic classification: A comparative analysis of deep autoencoders

As a system allowing intra-network devices to automatically communicate over the Internet, the Internet of Things (IoT) faces increasing popularity in modern applications and security threats — particularly network intrusions that target both networks and devices. A major threat is network attacks that attempt to obtain unauthorised access and damage the networks or systems. To effectively safeguard against these attacks, it is essential to use efficient hybrid intrusion detection systems with advanced techniques. Deep learning-based algorithms, such as Autoencoders (AEs), have been recognised as efficient methods for intrusion detection. However, it is crucial to further analyse the impact of different structures on detection performance. This paper proposes a method combining AE for data reconstruction and anomaly detection and multi-task learning (MTL)-structured models for IoT traffic classification. Additionally, we suggest a hybrid resampling technique with the Synthetic Minority Over-Sampling Technique and Generative Adversarial Network (SMOTE-GAN) for enhanced training and conduct a comparative analysis of different neural networks with AEs. The experimental results on two publicly available datasets demonstrate the effectiveness and practicality of the proposed AE-MTL approach compared with single-task learning. Additionally, while the performance varies on different datasets, Long Short-Term Memory (LSTM), Gated Recurrent Units (GRU), and Convolutional Neural Network (CNN) have improved the detection of minor classes.

Reachable set performance and quantized sampled-data synchronization analysis of neural networks under random packet dropouts via enhanced looped functional

The study aims to present the synchronization concerns of neural networks with stochastic packet dropouts utilizing a quantized memory sampled-data control (QMSDC) scheme and the problem of reachable set analysis (RSA) for NNs under the influence of external disturbances are discussed. To do this, a novel free-weighting matrix integral inequality (FWMII) is proposed with an augmented Lyapunov-functional. Furthermore, at this time, we show that the proposed inequality is less conservative due to the delay information. In the meantime, a novel enhanced looped function is created that utilizes not only sampling information but also incorporates both transmission delay information and a quantized sampling pattern. Next, using the proposed FWMII technique, several sufficient conditions in terms of linear matrix inequalities (LMIs) are derived to suggest how an augmented closed-loop system can achieves stochastic mean-square exponential synchronization under random packet loss situations. Finally, the proposed approach demonstrates its usefulness and superiority by comparing it with existing methods through numerical simulation of the secure communications.

Application of Neural Network in Seismic Risk Survey

The accuracy of inversion data is determined by the reliability of linear inversion method. Therefore, the study of nonlinear seismic inversion method is a meaningful work. This paper mainly aims at the application of nonlinear neural network inversion method in Fujian Risk Survey, the following work has been accomplished: (1) The theory and algorithm of the conventional multilayer feedforward neural network and some problems in the design of the model are analyzed and summarized. Aiming at the gradient descent function search algorithm in the BP algorithm, a method for global optimization is proposed by combining simulated annealing (global search algorithm) with Conjugate gradient method (local search algorithm) , the principle and implementation of simulated annealing and conjugate gradient algorithm are analyzed in detail. (2) Based on the analysis of conventional multilayer feedforward neural networks, the model structure and basic principle of probabilistic neural networks are studied. The object function search algorithm and its realization are analyzed. The method is applied to the risk elimination prediction for the first time, and the results are satisfactory. (3) The method and principle of inversion of reservoir parameters by using various seismic attribute information and neural network technology are discussed. In order to make full use of the effective information contained in seismic signal, the method of seismic attribute extraction and optimization is studied, the principle and implementation method of attribute optimal ranking by stepwise recursive method and selecting the most important attribute combination by interactive verification method are presented.

Experimental, Price-Performance and Artificial Neural Network (ANN) Analysis of MWCNT-CuO/ Water-EG (50-50) Nanofluid as a Coolant/Antifreeze Working Fluid

Experimental, Price-Performance and Artificial Neural Network (ANN) Analysis of MWCNT-CuO/ Water-EG (50-50) Nanofluid as a Coolant/Antifreeze Working Fluid

Large-region image inpainting using adaptive Fourier Neural Network and space-frequency domain structural analysis

Large-region image inpainting using adaptive Fourier Neural Network and space-frequency domain structural analysis

Evaluating Reliability and Economics of EV Charging Configurations and Deep Reinforcement Learning in Robotics and Autonomy

Growing EV popularity drives companies to focus on reliable charging station designs despite challenges in maintaining reliability. A proposed 36-ported design combines uniform and non-uniform port arrangements, tested with 50-350 kW systems. Failure rates are estimated using MILHDBK217F and MILHBK-338B standards, assessing port reliability and station success rates through binomial distribution and cost analysis. This design improves voltage stability and reduces maintenance costs through enhanced port reliability. In robotics and autonomous systems, Deep Reinforcement Learning (DRL) excels but faces challenges from unsafe policies leading to hazardous decisions. This study introduces a reliability assessment framework for DRL-controlled systems, using formal neural network analysis. A two-level verification approach evaluates safety locally using reachability tools and globally by aggregating local safety metrics across tasks. Experimental validation confirms the framework's effectiveness in enhancing RAS safety.

Financial priorities and the growth border of the Russian economy

Subject. The article investigates the role of foreign exchange reserves in the development of the Russian economy, the securities market development, and the banking system stability. Objectives. The aim is to review the monetary policy options under international sanctions strengthening. Methods. I apply methods of statistical, neural network, and cluster analysis. Results. The study established that it is impossible to reduce currency risks by issuing shares listed in the domestic market in foreign currency. The preferred directions of Russia's monetary policy are as follows: strengthening the country's international investment position, increasing the volume of the Federal bond market, and increasing the issue of shares listed in the domestic market in rubles. Conclusions. The growth boundary of the Russian economy is determined by such parameters as the amount of funds in escrow accounts of individuals, the size of insurance and pension reserves.

Challenges for computer vision as a tool for screening urban trees through street-view images

Urban forests play a fundamental and irreplaceable role within cities through the ecosystem services they provide, such as carbon capture. However, inadequate management of urban trees can heighten the risks they pose to society. For instance, mechanical failures of tree components, such as branches, can cause harm to individuals and property. Regular assessments of tree conditions are necessary to mitigate these tree-related hazards, yet such evaluations are labor-intensive and currently lack automation. Previous studies have proposed utilizing street view images to alleviate tree inspection and shown the feasibility of visually inspecting trees. However, only a limited number of studies have addressed the automatic evaluation of urban trees, a challenge that can potentially be addressed using deep learning networks. Particularly in urban environments, there is a pressing need for increased automation in unresolved computer vision tasks. Therefore, this research presents a comprehensive analysis of neural networks and publicly available datasets that can aid arborists in automatically identifying urban trees. Specifically, we investigate the potential of deep learning networks in classifying tree genera and segmenting individual trees and their trunks. We emphasize the utilization of transfer learning strategies to enhance tree identification. The results demonstrate that neural networks can be considered practical tools for assisting arborists in tree recognition. Nevertheless, there are still gaps that remain and require attention in future research endeavors.

Assessing inflation and greenhouse gas emissions interplay via neural network analysis: a comparative study of energy use in the USA, EU, and China

This study examines the relationship between inflation and greenhouse gas (GHG) emissions in three major economies: the United States of America (USA), the European Union (EU), and China. The analysis spans from 1960 to 2021 for the USA and EU, and from 1971 to 2021 for China. A feedforward neural network model, optimized using the Levenberg–Marquardt backpropagation algorithm, was employed to predict GHG emissions based on annual inflation rates and fossil fuel energy consumption. The study integrates historical data on inflation trends with GHG emissions, measured in CO2 equivalents, and fossil fuel energy consumption, expressed as a percentage of total energy use. This multidimensional approach allows for a nuanced understanding of the economic-environmental interplay in these regions. Key findings indicate a nonlinear response of GHG emissions to inflation rates. In the USA, GHG emissions begin to decrease when inflation rates exceed 4.7%. Similarly, in the EU, a steep reduction in emissions is observed beyond a 7.5% inflation rate. China presents a more complex pattern, with two critical inflection points: the first at a 4.5% inflation rate, where GHG emissions start to decline sharply, and the second at a 7% inflation rate, beyond which further increases in inflation do not significantly reduce emissions. A critical global insight is the identification of a uniform inflation rate, around 4.4%, across all regions, at which GHG emissions consistently increase by 1%, hinting at a shared global economic behavior impacting the environment. This discovery is vital for policymakers, emphasizing the need for tailored regional strategies that consider unique economic structures, energy policies, and environmental regulations, alongside a coordinated global approach.

Comparative Study on Simulated Outdoor Navigation for Agricultural Robots.

This research presents a comprehensive comparative analysis of SLAM algorithms and Deep Neural Network (DNN)-based Behavior Cloning (BC) navigation in outdoor agricultural environments. The study categorizes SLAM algorithms into laser-based and vision-based approaches, addressing the specific challenges posed by uneven terrain and the similarity between aisles in an orchard farm. The DNN-based BC navigation technique proves efficient, exhibiting reduced human intervention and providing a viable alternative for agricultural navigation. Despite the DNN-based BC navigation approach taking more time to reach its target due to a constant throttle limit for steady speed, the overall performance in terms of driving deviation and human intervention is notable compared to conventional SLAM algorithms. We provide comprehensive evaluation criteria for selecting optimal techniques for outdoor agricultural navigations. The algorithms were tested in three different scenarios: Precision, Speed, and Autonomy. Our proposed performance metric, P, is weighted and normalized. The DNN-based BC algorithm showed the best performance among the others, with a performance of 0.92 in the Precision and Autonomy scenarios. When Speed is more important, the RTAB-Map showed the best score with 0.96. In a case where Autonomy has a higher priority, Gmapping also showed a comparable performance of 0.92 with the DNN-based BC.

Plasma microRNA expression in adolescents and young adults with endometriosis: the importance of hormone use.

Prior studies have investigated the diagnostic potential of microRNA (miRNA) expression profiles for endometriosis. However, the vast majority of previous studies have only included adult women. Therefore, we sought to investigate differential expression of miRNAs among adolescents and young adults with endometriosis. The Women's Health Study: from Adolescence to Adulthood (A2A) is an ongoing WERF EPHect compliant longitudinal cohort. Our analysis included 64 patients with surgically-confirmed endometriosis (96% rASRM stage I/II) and 118 females never diagnosed with endometriosis frequency matched on age (median = 21 years) and hormone use at blood draw. MicroRNA measurement was separated into discovery (10 cases and 10 controls) and internal replication (54 cases and 108 controls) phases. The levels of 754 plasma miRNAs were assayed in the discovery phase using PCR with rigorous internal control measures, with the relative expression of miRNA among cases vs. controls calculated using the 2-ΔΔCt method. miRNAs that were significant in univariate analyses stratified by hormone use were included in the internal replication phase. The internal replication phase was split 2:1 into a training and testing set and utilized FirePlex miRNA assay to assess 63 miRNAs in neural network analyses. The testing set of the validation phase was utilized to calculate the area under the curve (AUC) of the best fit models from the training set including hormone use as a covariate. In the discovery phase, 49 miRNAs were differentially expressed between endometriosis cases and controls. The associations of the 49 miRNAs differed by hormone use at the time of blood draw. Neural network analysis in the testing set of the internal replication phase determined a final model comprising 5 miRNAs (miR-542-3p, let-7b-3p, miR-548i, miR-769-5p, miR-30c-1-3p), yielding AUC = 0.77 (95% CI: 0.67-0.87, p < 0.001). Sensitivity in the testing dataset improved (83.3% vs. 72.2%) while the specificity decreased (58.3% vs. 72.2%) compared to the training set. The results suggest that miR-542-3p, let-7b-3p, miR-548i, miR-769-5p, miR-30c-1-3p may be dysregulated among adolescent and young adults with endometriosis. Hormone use was a significant modifier of miRNA dysregulation and should be considered rigorously in miRNA diagnostic studies.

Possibilities of neural network analysis in determining the severity form of new coronavirus infection in school-age children

Possibilities of neural network analysis in determining the severity form of new coronavirus infection in school-age children

Designing a set of criteria for evaluating artificial neural networks trained with physics-based data to replicate molecular dynamics and other particle method trajectories

This article presents an in-depth analysis and evaluation of artificial neural networks (ANNs) when applied to replicate trajectories in molecular dynamics (MD) simulations or other particle methods. This study focuses on several architectures—feedforward neural networks (FNNs), convolutional neural networks (CNNs), recurrent neural networks (RNNs), time convolutions (TCs), self-attention (SA), graph neural networks (GNNs), neural ordinary differential equation (ODENets), and an example of physics-informed machine learning (PIML) model—assessing their effectiveness and limitations in understanding and replicating the underlying physics of particle systems. Through this analysis, this paper introduces a comprehensive set of criteria designed to evaluate the capability of ANNs in this context. These criteria include the minimization of losses, the permutability of particle indices, the ability to predict trajectories recursively, the conservation of particles, the model’s handling of boundary conditions, and its scalability. Each network type is systematically examined to determine its strengths and weaknesses in adhering to these criteria. While, predictably, none of the networks fully meets all criteria, this study extends beyond the simple conclusion that only by integrating physics-based models into ANNs is it possible to fully replicate complex particle trajectories. Instead, it probes and delineates the extent to which various neural networks can “understand” and interpret aspects of the underlying physics, with each criterion targeting a distinct aspect of this understanding.