Network Approach Research Articles

Mental health and its consequences in people living with HIV: A network approach.

Psychiatric symptoms occur frequently in people living with human immunodeficiency virus (PLWH), which may affect quality of life, sexual risk behavior, and adherence to antiretroviral therapy (ART). Data from large cohorts are limited, and symptoms are often analyzed in isolation. Therefore, we applied a network analysis to assess the interrelatedness of mental health indicators in a large cohort of PLWH. We included 1615 PLWH on ART. Participants reported on the severity of depression, anxiety, impulsivity, substance use, quality of life, sexual risk behavior, and ART adherence. An Ising network model was constructed to analyze interrelations between mental health indicators and connections with clinical consequences. Our network analysis revealed that symptoms of depression, anxiety, and indicators of impulsivity were interrelated. Substance use was prevalent and strongly connected with sexual risk behavior. Quality of life was most strongly connected with symptoms of depression. Unexpectedly, ART adherence did not display connections with any of the mental health indicators. In PLWH, the interrelatedness between symptoms of depression and anxiety and indicators of impulsivity is high. Mainly, depressive symptoms seem to impact quality of life, which warrants attention for depression in PLWH. We did not observe evidence for the common assumption that patients suffering from psychiatric symptoms are less adherent to HIV treatment.

Information flow in the FTX bankruptcy: A network approach

This paper investigates the cryptocurrency network of the FTX exchange during the collapse of its native token, FTT, to understand how network structures adapt to significant financial disruptions, by exploiting vertex centrality measures. Using proprietary data on the transactional relationships between various cryptocurrencies, we construct the filtered correlation matrix to identify the most significant relations in the FTX and Binance markets. By using suitable centrality measures - closeness and information centrality - we assess network stability during FTX’s bankruptcy. The findings document the appropriateness of such vertex centralities in understanding the resilience and vulnerabilities of financial networks. By tracking the changes in centrality values before and during the FTX crisis, this study provides useful insights into the structural dynamics of the cryptocurrency market. Results reveal how different cryptocurrencies experienced shifts in their network roles due to the crisis. Moreover, our findings highlight the interconnectedness of cryptocurrency markets and how the failure of a single entity can lead to widespread repercussions that destabilize other nodes of the network.

Statistical analysis of proteins families: a network and random matrix approach

Statistical analysis of proteins families: a network and random matrix approach

Cognitive, affective, and normative factors affecting digital insurance adoption among persons with disabilities: A two-stage SEM-ANN analysis

This study evaluates the determinants of digital insurance adoption among persons with disabilities (PWDs) using the cognitive, affective, and normative (CAN) model. The study considers (i) cognitive factors such as perceived credibility, perceived knowledge, perceived usefulness, perceived complexity, and facilitating conditions; (ii) affective factors including technology anxiety and technology pleasure; and (iii) normative factors encompassing social influence. Moreover, it explores the relationship between perceived complexity and behavioral intention (BI) to adopt digital insurance among PWDs, as mediated by perceived knowledge. This study employs a two-stage hybrid structural equation modeling–artificial neural network (SEM-ANN) approach to test the hypothesis, and data from 323 physically challenged participants were collected. Empirical results show that all factors, except for perceived complexity and technological anxiety, significantly predict BI adoption of digital insurance among PWDs, whereas perceived usefulness was found to have the highest impact on BI. Although perceived complexity affects perceived knowledge, it does not significantly mediate the relationship between complexity and BI. This study expands on the CAN model and provides practical insights for PWDs in adopting digital insurance.

Mesh-Free Solution of 2D Poisson Equation with High Frequency Charge Patterns Using Data-Free Physics Informed Neural Network

Abstract In this paper, we present a data-free physics-informed neural networks (PINNs) approach for solving two-dimensional (2D) Poisson equation, which is pivotal in fields such as electromagnetics, mechanical engginering, and thermodynamics. Traditional numerical method for solving this equation often require structured mesh generation such as Finite Element Method (FEM), which can be computationally expensive when dealing with high resolution Poisson Equation Solution. To address this challenge, we leverage the capabilities of PINNs capturing pattern of complex system by incorporating physical law and boundary condition as part of loss function on training model. While PINNs provide a robust framework for solving differential equations within boundary condition, they have struggle with capturing high-frequency pattern due to smooth nature of typical activation function used in neural networks. To evercome this issue, we enhance our model by incorporating Fourier Features Networks, which map inputs through a series of sinusoidal functions before feeding the input into the neural network. The result show that Fourier feature network can enhance convergence of training of PINNs model faster and obtained better result than PINNs without Fourier feature networks.

Advancing cancer driver gene identification through an integrative network and pathway approach

ObjectiveCancer is a complex genetic disease characterized by the accumulation of various mutations, with driver genes playing a crucial role in cancer initiation and progression. Distinguishing driver genes from passenger mutations is essential for understanding cancer biology and discovering therapeutic targets. However, the majority of existing methods ignore the mutational heterogeneity and commonalities among patients, which hinders the identification of driver genes more effectively. MethodsThis study introduces MCSdriver, a novel computational model that integrates network and pathway information to prioritize the identification of cancer driver genes. MCSdriver employs a bidirectional random walk algorithm to quantify the mutual exclusivity and functional relationships between mutated genes within patient cohorts. It calculates similarity scores based on a mutual exclusivity-weighted network and pathway coverage patterns, accounting for patient-specific heterogeneity and molecular profile similarity. ResultsThis approach enhances the accuracy and quality of driver gene identification. MCSdriver demonstrates superior performance in identifying cancer driver genes across four cancer types from The Cancer Genome Atlas, showing a higher F-score, Recall and Precision compared to existing ranking list-based and module-based models. ConclusionThe MCSdriver model not only outperforms other models in identifying known cancer driver genes but also effectively identifies novel driver genes involved in cancer-related biological processes. The model’s consideration of patient-specific heterogeneity and similarity in molecular profiles significantly enhances the accuracy and quality of driver gene identification. Validation through Gene Ontology enrichment analysis and literature mining further underscores its potential application value in personalized cancer therapy, offering a promising tool for advancing our understanding and treatment of cancer.

Individual-specific metabolic network based on 18F-FDG PET revealing multi-level aberrant metabolisms in Parkinson's disease.

Metabolic network analysis in Parkinson's disease (PD) based on 18F-FDG PET has revealed PD-related metabolic patterns. However, alterations at the systemic metabolic network level and at the connection level between different brain regions still remain unknown. This study aimed to explore metabolic network alterations at multiple network levels among PD patients using an individual-specific metabolic network (ISMN) approach. 18F-FDG-PET images of patients with PD (n = 34) and healthy subjects (n = 47) were collected. Healthy subjects were further separated into reference group (n = 28) and control group (n = 19) randomly. Standardized uptake value normalized by lean body mass ratio (SULr) maps was calculated from the PET images. ISMNs were constructed based on SULr maps for PD patients and controls with reference to the reference group. Comparisons of nodal and edge features were performed between PD and control groups. Correlation analysis was conducted between multilevel network properties and clinical scales in PD group. A linear classifier was trained based on nodal or edge features to distinguish PD from controls. The distance from each patient's ISMN to the group-level difference network showed a negative correlation with Hoehn and Yahr stage (r = -0.390, p = .023). Eight nodes from ISMN were identified which exhibited significantly increased nodal degree in PD patients compared to controls (p < .05). Eleven edges were observed which demonstrated significant distinctions in Z-score values in comparisons to the control group (p < .05). Furthermore, the nodal and edge features showed comparable performances in PD diagnosis compared to the traditional SULr values, with area under the receiver operating characteristic curve larger than 0.91. The proposed ISMN approach revealed systemic metabolic deviations, as well as nodal and edge distinctions in PD, which might be supplementary to the existing findings on PD-related metabolic patterns.

Gene-diet interactions in carbonated sugar-sweetened beverage consumption and metabolic syndrome risk: A machine learning analysis in a large hospital-based cohort

Background and aimCarbonated sugar-sweetened beverages (CSSB) intake has been increasingly linked to metabolic diseases. To investigate the association between CSSB intake and metabolic syndrome (MetS) risk, and the interaction between genetic predisposition to CSSB intake and dietary patterns. MethodsWe examined a hospital-based cohort of 57,940 participants, categorized into low-CSSB (n=52,848) and high-CSSB (n=5,092) groups based on a 50 ml daily consumption cutoff. A genome-wide association study (GWAS) identified single-nucleotide polymorphisms (SNPs) associated with CSSB intake, and SNP-SNP/SNP-environment interactions were explored. Using XGBoost and deep neural network (DNN) approaches, we developed prediction models for CSSB intake. ResultsThe low- and high-CSSB groups daily consumed an average of 0.56 and 8.91 g sugar from the soda, respectively. The high-CSSB group had unhealthy dietary habits and a lower intake of carotenoids, folate, vitamins C and D, calcium, flavonoids, and phenols than the low-CSSB group, consistent with the results of the prediction models. A polygenic risk score (PRS) based on 6 selected SNPs, linked to genes involved in obesity, diabetes, and nervous system disorders, showed the strongest association with CSSB intake and insulin resistance. Notably, carbohydrate, fat, and Western-style diet (WSD) intake interacted with the PRS, with lower carbohydrate and higher fat and WSD intakes associated with a stronger PRS-sugar intake relationship. The prediction models by XGboost and DNN mainly included dietary factors to explain CSSB intake. ConclusionsA significant interplay between genetic predisposition and poor dietary habits, particularly increased CSSB intake associated with WSD, contributed to MetS risk. It suggested that personalized dietary interventions based on genetic profiles could mitigate MetS risk, especially in populations transitioning to Westernized diets.

On acoustic solitary waves in a multispecies degenerate relativistic magnetized plasma using physics informed neural networks

In this paper, we investigate the nonlinear electrostatic wave propagation in a two-dimensional magnetized plasma. The plasma consists of electron and positron components with relativistic degeneracy and stationary ions for neutralizing its background. Using the basic equations for this type of plasma in combination with the reductive perturbation method, we derived the Zakharov–Kuznetsov equation using the Lorentz transformation stretching method (LT). For the first time, we compared the results of the Galilean transformation stretching method (GT) and the LT method to investigate the effect of plasma parameters, such as the relativistic degeneracy parameter of electron particles (re0), the density ratio of ion to electrons (δ), and the normalized electron cyclotron (Ωe), on the amplitude and width of the wave solutions. The plasma parameters used in this research are representative of compact astrophysical objects. Numerical results showed that the amplitude of wave solutions obtained by the LT method is smaller than the GT method, but the width is greater. We provide a physical explanation for these differences. Furthermore, we present a physics-informed neural network (PINN) approach to directly recover the intrinsic nonlinear dynamics from spatiotemporal data. The PINN model uses a deep neural network constrained by the governing equations to learn the optimal parameters, with the aim of enhancing the predictive capabilities of the system. The results of this study provide valuable insight into the propagation of nonlinear waves in white dwarfs, where relativistic effects are significant. These findings could substantially advance the development of emerging machine learning applications in astrophysics.

Optimal operation model of ecological flow of hydropower station based on genetic algorithm and neural network

Amid diverse water demands, the conflict between ecological flow adjustments and power generation flow adjustments in hydropower station water resource dispatching, coupled with an increasing diversity of constraints, complicates the issue of water resource optimization and dispatching. How to scientifically develop feedback regulation mechanisms to promote a virtuous ecological cycle, fine tune existing hydropower station scheduling plans, and achieve multi-objective optimization scheduling is the core issue in researching the comprehensive utilization of water resources. This study establishes a reservoir flow regulation model with the objective functions of maximizing power generation and achieving optimal average ecological protection. The model, constrained by water balance, reservoir capacity, unit output, and unit overflow, was optimized and solved using the genetic algorithm-backpropagation neural network (GA-BPNN) approach. This study establishes a reservoir flow regulation model with the GA-BPNN algorithm was applied in multi-objective optimization to identify an optimal solution that maximizes overall system performance while adhering to ecological flow constraints. The results indicate that the optimized plan has performed effectively in hydropower station operations. It not only satisfies ecological flow requirements, but also significantly enhances power generation efficiency. Specifically, power generation increased by 32,680 kw·h, a 23.85% growth rate. The optimized plan significantly enhances the comprehensive utilization efficiency of water resources. It offers a scientific basis for the rational planning, dispatching and utilization of water resources in hydropower stations and reservoirs.

Sample Feature Enhancement Model Based on Heterogeneous Graph Representation Learning for Few-shot Relation Classification

Few-Shot Relation Classification (FSRC) aims to predict novel relationships by learning from limited samples. Graph Neural Network (GNN) approaches for FSRC constructs data as graphs, effectively capturing sample features through graph representation learning. However, they often face several challenges: 1) They tend to neglect the interactions between samples from different support sets and overlook the implicit noise in labels, leading to sub-optimal sample feature generation. 2) They struggle to deeply mine the diverse semantic information present in FSRC data. 3) Over-smoothing and overfitting limit the model's depth and adversely affect overall performance. To address these issues, we propose a Sample Representation Enhancement model based on Heterogeneous Graph Neural Network (SRE-HGNN) for FSRC. This method leverages inter-sample and inter-class associations (i.e., label mutual attention) to effectively fuse features and generate more expressive sample representations. Edge-heterogeneous GNNs are employed to enhance sample features by capturing heterogeneous information of varying depths through different edge attentions. Additionally, we introduce an attention-based neighbor node culling method, enabling the model to stack higher levels and extract deeper inter-sample associations, thereby improving performance. Finally, experiments are conducted for the FSRC task, and SRE-HGNN achieves an average accuracy improvement of 1.84% and 1.02% across two public datasets.

Deep indoor illumination estimation based on spherical gaussian representation with scene prior knowledge

High dynamic range illumination estimation from a single low dynamic range image is a critical task in the fields of computer vision, graphics and augmented reality. However, directly learning the full HDR environment map or parametric lighting information from a single image is extremely difficult and inaccurate. As a result, we propose a two-stage network approach for illumination estimation that integrates spherical gaussian (SG) representation with scene prior knowledge. In the first stage, a convolutional neural network is utilized to generate material and geometric information about the scene, which serves as prior knowledge for lighting prediction. In the second stage, we model indoor environment illumination using 128 SG functions with fixed center direction and bandwidth, allowing only the amplitude to vary. Subsequently, a Transformer-based lighting parameter regressor is employed to capture the complex relationship between the input images with scene prior information and its SG illumination. Additionally, we introduce a hybrid loss function, which combines a masked loss for high-frequency illumination with a rendering loss for improving the visual quality. By training and evaluating the lighting model on the created SG illumination dataset, the proposed method achieves competitive results in both quantitative metrics and visual quality, outperforming state-of-the-art methods.

Short interbirth interval and adverse pregnancy outcomes: a Bayesian network approach

BackgroundInterbirth interval (IBI), the time between consecutive births, has been tied to perinatal outcomes. ObjectiveTo analyze adverse perinatal events following short IBI in a large South American sample. Study designObservational, retrospective, hospital-based study including malformed and nonmalformed live- and stillbirths. Outcomes were preterm birth (PTB), low birth weight (LBW), and specific birth defects. Logistic regressions were used to evaluate the risk of selected variables for short IBI and for adverse outcomes after short IBI, adjusting by confounders. Bayesian networks exhibited relationships among short IBI, outcomes, and variables. ResultsShort IBI rate was 2%–3%. Maternal age and a previous abortion were the main confounders. A significant high risk for short IBI was found in mothers ≤19 years while mothers ≥30 were at low risk, mediated by a previous abortion. The risk of short IBI, adjusted by confounders, was significant for LBW but not for PTB. An unadjusted risk of short IBI was observed for gastroschisis, which disappeared after adjusting for confounders. Maternal age ≤19 and previous abortion were directly related with gastroschisis; the relationship between gastroschisis and short IBI occurred through any of these two variables. A direct relationship between gastroschisis and maternal age ≥30 was observed. ConclusionsOnly young mothers were directly related with short IBI. In older mothers, a short IBI mainly occurred after a previous abortion. Short IBI was a risk factor only for LBW. The PTB and gastroschisis relationship with short IBI was indirect, mediated by young maternal age and/or a previous abortion.

Scalable Hybrid Beamforming for Multi-User MISO Systems: A Graph Neural Network Approach

Scalable Hybrid Beamforming for Multi-User MISO Systems: A Graph Neural Network Approach

Jointly Optimizing Terahertz Based Sensing and Communications in Vehicular Networks: A Dynamic Graph Neural Network Approach

Jointly Optimizing Terahertz Based Sensing and Communications in Vehicular Networks: A Dynamic Graph Neural Network Approach

Convolutional neural network approach for fault detection and characterization in medium voltage distribution networks

Power outages significantly impact the power industry by disrupting social welfare and economic stability. Still, existing methods for fault detection face challenges due to load and network topology, conditions, and installed equipment. However, recent advances in artificial intelligence (AI) are enabling researchers to create alternative approaches for fault detection and location strategies. Therefore, this paper introduces a novel method for detecting, classifying, and locating faults in power systems through voltage waveform analysis using a convolutional neural network (CNN) integrated with the Piecewise Function Put Together (PFPT) algorithm for fault detection and fault zone localization in a power distribution network. Utilizing Park's transformation, noise reduction PFPT sine fitting, and CNNs, the proposed method distinguishes between 'healthy' and 'faulty' conditions. Simulation results reveal that while the voltage Park's vector time behavior of a healthy system remains stable, it exhibits circular or mixed patterns under faulty conditions. These patterns enable the identification of four types of short circuit faults—single-line-to-ground (LG), line-to-line (LL), line-to-line-to-ground (LLG), and three-line (3L) faults—by analyzing 3D voltage Park's waveforms at network buses. The study validates fault type identification through the observation of rotating Park vectors from sine fitting of time-based voltage waveforms. By converting 3D voltage waveforms into high-resolution images, the method utilizes a CNN for fault recognition, achieving an accuracy of 93.1%. This innovative approach underscores the robustness and precision of combining traditional electrical engineering techniques with modern AI.

The effect of ESG divergence on the financial performance of Hong Kong-listed firms: an artificial neural network approach

This paper applies an advanced machine learning algorithm, the Artificial Neural Network (ANN), to examine both linear and nonlinear effects between firm-level characteristics and ESG performance of all firms listed on the Hong Kong Stock Exchange (HKEX) with ESG scores during 2019-2021. To mitigate the problem of data-specific findings due to rating bias from a single rating agency, we employ novel iScore (divergence-adjusted ESG measure). The documented findings indicate the unsuitability of traditional linear regression models to capture the nonlinear effects and to detect some linear relationships. Furthermore, the results show the superiority of the self-organising map (SOM) ANN framework in explaining the impact of firm-level factors on ESG performance.

Assessment of hybrid composite drilling and prediction of cutting parameters by ANFIS and deep neural network approach

Assessment of hybrid composite drilling and prediction of cutting parameters by ANFIS and deep neural network approach

Predictive residual stress analysis in gear units: an artificial intelligence approach based on finite element data

Determining residual stresses is essential for ensuring the safety and longevity of gear components. Traditional methods, such as analytical, numerical, and experimental approaches, are often costly, time-consuming, and sometimes destructive. This study proposes a new method to predict residual stresses in gear units using artificial intelligence based on data from finite element analysis. To achieve this, a commercial finite element tool models the heat treatment and carburization processes, grounded in thermomechanical metallurgical physics. The residual stresses obtained from finite element analysis are then compared to the ones from experiment using deep hole drilling approach. This is followed by a sensitivity analysis. Next, a specific class of gearbox component geometries and the computed stress fields are utilized to train artificial neural networks. The performance of the artificial neural networks approach is compared to that of two different machine learning regression methods. It has been concluded that this technique successfully anticipates residual stresses in components with scaled geometries and similar heat treatment steps, outperforming the two regression models in accuracy and efficiency. Importantly, it achieves this with incomparable, significantly less computation time compared to standard finite element analysis.

A network approach to PhD students’ well-being: insights into risk and protective factors

ABSTRACT Due to the high demands of doctoral education, concerns have been raised about PhD students’ well-being. In addition to the adverse individual impact, poor well-being of PhD students is also problematic for universities, which might lose talented researchers via absenteeism and drop-out. Despite substantial knowledge about factors influencing well-being in academia, we know little about how these factors jointly promote or impede PhD students’ well-being. We applied network analysis, alongside regression analysis, to cross-sectional data from 665 PhD students in the Netherlands to explore how PhD students’ job demands, job resources, and personal resources are associated with subjective and archival well-being indices. Findings suggest that job demands, job resources, and personal resources influence PhD students’ well-being directly and indirectly via associations with other factors. Among job demands, both publication pressure and work-home conflict were associated with higher burnout. Additionally, work-home conflict was associated with higher sickness absenteeism. Psychological capital stood out as the most important resource within the network and was related to lower burnout, higher engagement, and lower sickness absenteeism. By highlighting the importance of personal resources, particularly psychological capital, for PhD students’ well-being, this study can guide the development of interventions to promote PhD students’ well-being.