Bayesian Network Approach Research Articles

Assessing the Impact of Risk Factors on Vaccination Uptake Policy Decisions Using a Bayesian Network (BN) Approach

This study evaluates the propagation impact of three risk categories (hazard and exposure, socio-economic vulnerability, and lack of coping capacity) and their associated factors on vaccination uptake policy decisions in Pakistan. This study proposed Bayesian influence diagrams using expert elicitation and data-driven approaches. The Bayesian network (BN) approach uses the best policy algorithm to determine the expected utility of decisions. The study found that the government’s firm vaccine uptake decisions had a positive effect in Pakistan. The findings on hazard and exposure-related factors show that people living in rural areas were more susceptible to COVID-19 than people living in urban areas. Among socio-economic vulnerability factors, household characteristics were affected due to household economic situations, fear of using health facilities due to the spread of COVID-19, lack of public transportation services, food insecurity, a temporary halt in education, and weak governance, which affected the vaccination uptake decision. The factors linked with coping capacity show that the government’s financial assistance and development of digital platforms raised digital health literacy and increased vaccine uptake decision utility. The proposed methodology and results of this study can be used to develop contingency planning for any future potential pandemic situations.

Uncertainty quantification in multivariable regression for material property prediction with Bayesian neural networks

With the increased use of data-driven approaches and machine learning-based methods in material science, the importance of reliable uncertainty quantification (UQ) of the predicted variables for informed decision-making cannot be overstated. UQ in material property prediction poses unique challenges, including multi-scale and multi-physics nature of materials, intricate interactions between numerous factors, limited availability of large curated datasets, etc. In this work, we introduce a physics-informed Bayesian Neural Networks (BNNs) approach for UQ, which integrates knowledge from governing laws in materials to guide the models toward physically consistent predictions. To evaluate the approach, we present case studies for predicting the creep rupture life of steel alloys. Experimental validation with three datasets of creep tests demonstrates that this method produces point predictions and uncertainty estimations that are competitive or exceed the performance of conventional UQ methods such as Gaussian Process Regression. Additionally, we evaluate the suitability of employing UQ in an active learning scenario and report competitive performance. The most promising framework for creep life prediction is BNNs based on Markov Chain Monte Carlo approximation of the posterior distribution of network parameters, as it provided more reliable results in comparison to BNNs based on variational inference approximation or related NNs with probabilistic outputs.

Pipeline corrosion prediction and uncertainty analysis with an ensemble Bayesian neural network approach

This study developed an ensemble Bayesian Neural Network (BNN) model for pipeline corrosion prediction incorporating uncertainty analysis. The performance of the proposed BNN model was compared with different empirical and data-driven models. Shapley Additive Explanation (SHAP) values were used to quantify the importance of input factors. Results indicate that the ensemble BNN model demonstrates superior prediction accuracy, achieving the highest R2 value of 0.949, with the lowest Mean Absolute Error (MAE) of 0.33 mm and Root Mean Square Error (RMSE) of 0.449 mm on corrosion depth. It outperforms other models in predicting maximum corrosion depth and uniquely captures uncertainty information, which deterministic models cannot provide. Key factors influencing corrosion include chloride content, pH, pipe-to-soil potential, and pipeline age. Specifically, conditions such as a pipe-to-soil potential exceeding − 0.85 V, a pipeline age over 20 years, and pH values below 7 are found to significantly accelerate corrosion. The model also illustrates the effects of different influencing factors on corrosion growth uncertainty. Higher chloride content, lower pH values, and higher pipe-to-soil potentials may expand the uncertainty bounds. The proposed model presents the dynamic changes in prediction uncertainty over time, which is usually ignored in previous models. It can improve the precision and reliability of long-term predictions concerning pipeline corrosion depth. This knowledge of corrosion uncertainty is crucial for making informed decisions about pipeline inspection and maintenance strategies.

Probabilistic safety and reliability studies toward licensing and deploying HTGR technology in the Polish chemical industry

The pivotal aspects of the Probabilistic Safety Assessment (PSA) and reliability studies for the foreseen HTGR-based nuclear-chemical (Nuc-Chem) facilities are discussed. The traditional PSA methodology has been extended to include the mutual dependence of nuclear and chemical systems of the hybrid Nuc-Chem installations, where HTGR serves as a heat source for various chemical processes. The Dynamic Bayesian Network (DBN) approach has been utilized for modeling time sequences of failures, including interactions between nuclear and non-nuclear systems, as well as external hazards. The Global Variance Method (GVM), based on the Sobol indices, has been proposed for uncertainty and sensitivity studies concerning numerous sources of variability in a consistent manner. The paper presents the outcomes and achievements from research oriented toward licensing and deploying High Temperature Gas-cooled Reactors (HTGR) within the Polish chemical industry. The general framework developed in this work for PSA of the hybrid Nuc-Chem facilities can be applied to support the licensing process and safe operation of the Polish chemical plants to be equipped with HTGRs.

How Do Developers Influence the Transaction Costs of China’s Prefabricated Housing Development Process? An Investigation through the Bayesian Belief Network Approach

The implementation of prefabricated housing (PH) has become prevalent in China recently due to its advantages in enhancing production and energy-saving efficiency within the construction system. However, stakeholders may not always fully realize the benefits of adopting PH due to the emergence of transaction costs (TCs) in the development process of such projects. This study investigated the strategies for developers to make rational choices for minimizing the TCs of the PH project considering their own attributes and external constraints. A Bayesian Belief Network model was applied as the analytical method, based on surveys conducted in China. A single sensitivity analysis indicated that developers influence the TCs of PH through the following three most impactful factors: prefabrication rate, PH experience, and contract payment method. Integrated strategies are recommended for developers in various situations based on a multiple sensitivity analysis. Developers facing challenges due to high prefabrication rates are advised to reduce the risks by procuring highly qualified general contractors and adopting unit-price contracts. For developers with limited PH experience, adopting the Engineering–Procurement–Construction procurement method is the most efficient way to reduce their TCs in the context of China’s PH market. This study contributes to the current body of knowledge concerning the effect of traders’ attributes and choices on TCs, expanding the application of TC theory and fulfilling the study on the determinants of TCs in construction management.

Robust identification of interactions between heat-stress responsive genes in the chicken brain using Bayesian networks and augmented expression data

Bayesian networks represent a useful tool to explore interactions within biological systems. The aims of this study were to identify a reduced number of genes associated with a stress condition in chickens (Gallus gallus) and to unravel their interactions by implementing a Bayesian network approach. Initially, one publicly available dataset (3 control vs. 3 heat-stressed chickens) was used to identify the stress signal, represented by 25 differentially expressed genes (DEGs). The dataset was augmented by looking for the 25 DEGs in other four publicly available databases. Bayesian network algorithms were used to discover the informative relationships between the DEGs. Only ten out of the 25 DEGs displayed interactions. Four of them were Heat Shock Proteins that could be playing a key role, especially under stress conditions, where maintaining the correct functioning of the cell machinery might be crucial. One of the DEGs is an open reading frame whose function is yet unknown, highlighting the power of Bayesian networks in knowledge discovery. Identifying an initial stress signal, augmenting it by combining other databases, and finally learning the structure of Bayesian networks allowed us to find genes closely related to stress, with the possibility of further exploring the system in future studies.

Abstract LB014: The role of genetic alterations in metastatic development insights from a large-scale genomic database

Abstract Metastasis remains a major challenge in cancer treatment, with somatic genetic alterations being a crucial feature in progression. This study harnesses the whole genome sequencing data in the Genomics England biobank (GEL), cataloging of somatic alterations in the context of cancer metastasis. Our research aims to uncover the genetic underpinnings that facilitate the spread of cancer from its primary site to secondary locations by analyzing clinicogenomic data from 3047 cancer patients with metastatic progression. In this study, we analyzed the complex relationship between somatic alterations and cancer progression to uncover significant interactions among these alterations in a broad distribution of common tumor types among well-established oncogenic and tumor suppressor genes. In our study, we employed a Bayesian network approach, enabling us to not only pinpoint crucial genetic contributors to metastasis but also to consider other clinical variables, including patient treatment, sex, age, tumor grade and cancer types. In our analysis, LILRB2 was found to directly increase metastatic risk in hematologic cancer, with prior evidence suggesting its potential as a therapeutic biomarker in colorectal cancer and adversely related to patient prognosis in lung cancer. Similarly, MYL1, initially linked to tumor metastasis and immune infiltration in head and neck squamous cell carcinoma, was also found to increase the risk of metastasis in renal cancer, while interacting with the transcription factor E2F4, marker of poor prognosis in hepatocellular carcinoma. Our study not only confirmed the numerous key somatic gene alterations significantly linked to metastatic behavior across various cancers, but also established their significant roles.Our work highlights the potential of leveraging extensive genomic databases such as GEL to enhance our understanding of cancer metastasis. GEL’s paired germline and somatic data allows us to not only identify mutations unique to cancer cells but also to uncover germline variants that are associated with an elevated risk of cancer and the presence of more aggressive tumor characteristics. By characterizing metastatic risk features, we can explore the opportunities to stratify high-risk patients in clinical trials, with similar effort has been applied to brain metastasis in lung adenocarcinoma. Although tumor heterogeneity poses a treatment challenge due to cell diversity, previous work has demonstrated value in targeting the metastatic potential. For instance, trastuzumab target HER2-positive breast cancers prone to metastasis, markedly improving patient outcomes. Similarly, vemurafenib effectively controls metastatic Melanoma progression by targeting specific mutations like BRAFV600E. Understanding the genetic basis of metastasis could help identify high risk patients and lead to personalized combination therapies addressing various tumor subpopulations. Future research based on these findings could lead to the development of personalized combination treatment strategies that improve patient outcomes. Citation Format: Songjun Xu, Danyang Yu, Shuwei Li, Christopher Moy, Julio Molineros. The role of genetic alterations in metastatic development insights from a large-scale genomic database [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr LB014.

A two-step Bayesian network approach to identify key SNPs associated to multiple phenotypic traits in four purebred laying hen lines.

When purebred laying hen chicks hatch, they remain at a rearing farm until approximately 17 weeks of age, after which they are transferred to a laying farm. Chicks or pullets are removed from the flocks during these 17 weeks if they display any rearing abnormality. The aim of this study was to investigate associations between single nucleotide polymorphisms (SNPs) and rearing success of 4 purebred White Leghorns layer lines by implementing a Bayesian network approach. Phenotypic traits and SNPs of four purebred genetic White Leghorn layer lines were available for 23,000 rearing batches obtained between 2010 and 2020. Associations between incubation traits (clutch size, embryo mortality), rearing traits (genetic line, first week mortality, rearing abnormalities, natural death, rearing success, pullet flock age, and season) and SNPs were analyzed, using a two-step Bayesian Network (BN) approach. Furthermore, the SNPs were connected to their corresponding genes, which were further explored in bioinformatics databases. BN analysis revealed a total of 28 SNPs associated with some of the traits: ten SNPs were associated with clutch size, another 10 with rearing abnormalities, a single SNP with natural death, and seven SNPs with first week mortality. Exploration via bioinformatics databases showed that one of the SNPs (ENAH) had a protein predicted network composed of 11 other proteins. The major hub of this SNP was CDC42 protein, which has a role in egg production and reproduction. The results highlight the power of BNs in knowledge discovery and how their application in complex biological systems can help getting a deeper understanding of functionality underlying genetic variation of rearing success in laying hens. Improved welfare and production might result from the identified SNPs. Selecting for these SNPs through breeding could reduce stress and increase livability during rearing.

Operationalizing a fisheries social-ecological system through a Bayesian belief network reveals hotspots for its adaptive capacity in the southern North sea

Fisheries social-ecological systems (SES) in the North Sea region confront multifaceted challenges stemming from environmental changes, offshore wind farm expansion, and marine protected area establishment. In this paper, we demonstrate the utility of a Bayesian Belief Network (BN) approach in comprehensively capturing and assessing the intricate spatial dynamics within the German plaice-related fisheries SES. The BN integrates ecological, economic, and socio-cultural factors to generate high-resolution maps of profitability and adaptive capacity potential (ACP) as prospective management targets. Our analysis of future scenarios, delineating changes in spatial constraints, economics, and socio-cultural aspects, identifies factors that will exert significant influence on this fisheries SES in the near future. These include the loss of fishing grounds due to the installation of offshore wind farms and marine protected areas, as well as reduced plaice landings due to climate change. The identified ACP hotspots hold the potential to guide the development of localized management strategies and sustainable planning efforts by highlighting the consequences of management decisions. Our findings emphasize the need to consider detailed spatial dynamics of fisheries SES within marine spatial planning (MSP) and illustrate how this information may assist decision-makers and practitioners in area prioritization. We, therefore, propose adopting the concept of fisheries SES within broader integrated management approaches to foster sustainable development of inherently dynamic SES in a rapidly evolving marine environment.

Causal Model Framework for Nuclear Power Plant Licensing Process

Interests in clean energy revived the nuclear power industry. For the first time in decades, innovative technologies and plant designs are being considered by regulatory agencies. This paper explores a Bayesian Network and AHP approach to causal modeling of the Combined License review process for new nuclear power plants (NPP). Historically lengthy and expensive, NPP licensing is critical to ensuring safe operation of the plants. With this comes a high standard for applicants to reach that can result in multiple revision cycles and long review times.The U.S. Nuclear Regulatory Commission's combined license process is examined in this study as an opportunity to predict delays based on project criteria and aid informed decision making. A Bayesian Network combined with Analytic Hierarchy Process inputs is presented to show causality relationships between licensing process steps, plant parameters, and time delays.

Numerical investigation of a fractional order Wolbachia invasive model using stochastic Bayesian neural network

The major goal of this research study is to solve the fractional order Wolbachia invasive model (FWIM) by developing a computational framework based on the Bayesian regularization backpropagation neural network (BRB-NN) approach. The population of mosquitoes is categorized into two classes, Wolbachia-infected mosquitoes and Wolbachia-uninfected mosquitoes. We also incorporate incomplete cytoplasmic incompatibility and imperfect maternal transmission. We investigate the effects of the fractional order derivative (α) and reproduction rate of Wolbachia-infected mosquitoes ϕw on the dynamics of mosquitoes. The proposed Bayesian regularization backpropagation scheme is applied to three distinct cases using 80% and 20% of the created dataset for training and testing, respectively, with 15 hidden neurons. Comparisons of the results are presented to verify the validity of the proposed technique for solving the model. The Bayesian regularization approach is used to lower the mean square error (MSE) for the fractional order Wolbachia invasive model. The achieved results are based on MSE, correlation, state transitions, error histograms, and regression analysis to confirm the effectiveness of the suggested approach. Additionally, the absolute error value modifies the designed approach’s accuracy.

Analyzing human reliability for the operation of cargo oil pump using fuzzy CREAM extended Bayesian Network (BN)

Crude oil cargo discharging operations can be performed by cargo oil pumps in tanker ships. Considering the harmful effects of crude oil, any failure during the cargo oil pump operation may pose acute hazards such as temperature and pressure increase in pipelines, leakage, fire, and following undesired events that lead to marine environmental pollution and toxic effects to humans. The process is managed by the ship's crew, and it requires following the operational steps in order. This paper aims at performing a detailed human reliability analysis (HRA) for the operational process of cargo oil pumps on a tanker ship. In the paper, the Cognitive Reliability and Error Analysis Method (CREAM) is used to calculate human error probability under a fuzzy set which deals with the uncertainty and ambiguity of the CPC (Common Performance Condition). The Bayesian Network (BN) approach is adopted to determine the probability distribution of the control modes in reliability assessment. In view of the findings, the human reliability for the entire process of cargo oil pump operation on tanker ship is found 8.82E-01. The outcomes of the paper provide valuable insight into the improvement of tanker safety in shipboard operation and performance reliability assessment.

How do navy escorts influence piracy risk in East Africa? A Bayesian network approach.

Navy escorts are considered crucial in countering illegal piracy attacks. In this paper, a novel approach is developed to investigate the effect of navy escorts on piracy incidents by models based on two enhanced Tree-Augmented Naïve (TAN) Bayesian networks. This approach offers a systematic investigation into the various factors that influence pirate activities, and helps to identify changes in piracy attack behaviors when confronted by navy escorts and assess the effectiveness of anti-piracy measures. An empirical study is conducted utilizing a unique data set compiled from multiple sources from 2000 to 2019. The empirical evidence shows that there was a gradual reduction in the incidence of piracy attacks in East Africa following the implementation of navy escorts in 2009, but with a surge in 2010 and 2011. The data set is, thus, divided into two time periods at the point of 2009 to facilitate a robust and comprehensive analysis, resulting in the development of two TAN models. Meanwhile, the geographical distribution of pirate attacks has shifted from international waters to port areas and territorial waters. We argue that the surge and geographical shift could be attributed to the calculating behavior of pirates when they encounter external pressures. Finally, a Shapely approach is introduced to evaluate the potential effectiveness of the implemented risk management strategies from a Game Theory perspective. This study offers new insights into the promotion of navy escorts and contributes to the development of a framework for assessing piracy risks in uncertain and dynamic anti-piracy environments.

A Simulation-Based Bayesian Network Approach to the Joint Decision of the International Transportation Mode and Safety Inventory Policy

Uncertainty in international transportation can have a significant impact on inventory costs and customer service levels in a global supply chain network. A very limited number of studies in the literature have addressed the international transportation mode and safety inventory policy (ITM-SIP) issues simultaneously. This study formulates the problem and proposes a simulation-based Bayesian network (SBN) approach for the joint decisions on ITM-SIP, where Bayesian networks are used to dynamically estimate international replenishment lead times for railway and maritime transportation modes. We implement our approaches in an international automotive spare parts supply chain experiencing fluctuations in its shipping supply market and with dynamic demand. For this joint decision problem, we build an international supply chain network simulation model in which the time-delay uncertainties of key bottleneck nodes in maritime and railway transportation routes are considered. The simulation experiment results show the performance improvement of the SBN approach in substantially reducing total costs and achieving service levels. We also provide insights into the value of decision tactics for responding to different fluctuation modes in the shipping supply market in the global supply chain network.

A dynamic system‐based model for analyzing human factors: Enhanced AcciMap with spherical fuzzy dynamic Bayesian network approach

AbstractIn today's interconnected global economy, maritime trade is a pillar of prosperity, yet maritime accidents loom as a formidable challenge. The intricate nature of these accidents, coupled with rapid technological advancements, necessitates the evolution of systematic analysis methods. Conventional systemic approaches, while valuable, struggle to encapsulate the intricate web of mutual and dynamic dependencies inherent in these incidents. Furthermore, the call for more quantitative support in decision‐making and the ability to account for emergent factors has become increasingly imperative. This study aims to analyze maritime accidents by introducing a quantitative and dynamic model. The endeavour begins with establishing an extended Accident Map‐based model, a robust framework that unveils a sophisticated accident causation model. This preliminary action establishes the groundwork for integrating an innovative Spherical Fuzzy Set, navigating the complex landscape of knowledge acquisition. The subsequent phase charts a transformative course by mapping the model onto a dynamic Bayesian Network to conduct a forward and backward analysis. The essence of the model lies in its dynamic nature, allowing for real‐time updates that reflect the evolving maritime accidents risk factors. The approach is validated through a partial benchmark exercise, a reality check, an independent peer review, and a sensitivity analysis. The model can explore emerging contributing factors, reduce uncertainty, and consider relationships between factors that yield designing more effective safety measures.

Bayesian network approach for reliability analysis of mining trucks

Having a safe and efficient system for mineral transportation is a top priority for all mining operations. Trucks are the most widely used material transportation systems that are applied in both surface and underground mines. Any truck failure disrupts the mineral transportation process and consequently decreases the overall output. Therefore, the reliability analysis of such equipment plays a critical role in increasing the efficiency and productivity of a mining operation. This paper proposes a novel method for analyzing the reliability of a fleet of mining trucks based on the Bayesian Network modeling. Considering the reliability block diagram, the fault tree of trucks was developed according to the logical relationship between the units. Then, a dynamic Bayesian network was constructed according to the conditional probability analysis. Moreover, the relative contributions of each truck’s component to the occurrence of the fleet failure were studied by using critical analysis. The results of this paper show that the successful operation of the fleet of trucks is most sensitive to truck no. 5, which has the highest reliability level in all time intervals. The reliability of the fleet of trucks reaches 0.881 at 20 h, and the fuel injection system of the truck’s engine is the main leading cause of the trucks failure. A proper preventive maintenance strategy should be paid more attention to improve the reliability and availability of the engine system.

Integrating joint latent class mixed models and Bayesian network for uncovering clinical subgroups of COVID-19 patients

When modelling the dynamics of biomarkers in biomedical studies, it is essential to identify homogeneous clusters of patients and analyse them from a precision medicine perspective. This need has emerged as crucial and urgent during the COVID-19 pandemic: early understanding of symptoms and patient heterogeneity has significant implications for prevention, early diagnosis, effective management, and treatment. Additionally, biomarker progression may be associated with clinically relevant time-toevent data. Therefore, statistical models are necessary to gain insight into complex disease mechanisms by properly accounting for unobservable heterogeneity in patients while jointly modelling longitudinal and time-to-event data. In this study, we leverage the key features of Latent Class modelling and Bayesian Network approaches and propose a unified framework to (a) uncover homogeneous subgroups of patients concerning their longitudinal and survival data and (b) describe patient subgroups within a multivariate framework.

Can Data and Machine Learning Change the Future of Basic Income Models? A Bayesian Belief Networks Approach

Appeals to governments for implementing basic income are contemporary. The theoretical backgrounds of the basic income notion only prescribe transferring equal amounts to individuals irrespective of their specific attributes. However, the most recent basic income initiatives all around the world are attached to certain rules with regard to the attributes of the households. This approach is facing significant challenges to appropriately recognize vulnerable groups. A possible alternative for setting rules with regard to the welfare attributes of the households is to employ artificial intelligence algorithms that can process unprecedented amounts of data. Can integrating machine learning change the future of basic income by predicting households vulnerable to future poverty? In this paper, we utilize multidimensional and longitudinal welfare data comprising one and a half million individuals’ data and a Bayesian beliefs network approach to examine the feasibility of predicting households’ vulnerability to future poverty based on the existing households’ welfare attributes.

Bayesian Network Analysis for Shoreline Dynamics, Coastal Water Quality, and Their Related Risks in the Venice Littoral Zone, Italy

The coastal environment is vulnerable to natural hazards and human-induced stressors. The assessment and management of coastal risks have become a challenging task, due to many environmental and socio-economic risk factors together with the complex interactions that might arise through natural and human-induced pressures. This work evaluates the combined effect of climate-related stressors on low-lying coastal areas by applying a multi-risk scenario analysis through a Bayesian Network (BN) approach for the Venice coast. Based on the available open-source and remote sensing data for detecting shoreline changes, the developed BN model was trained and validated with oceanographic variables for the 2015–2019 timeframe, allowing us to understand the dynamics of local-scale shoreline erosion and related water quality parameters. Three “what-if” scenarios were carried out to analyze the relationships between oceanographic boundary conditions, shoreline evolution, and water quality parameters. The results demonstrate that changes in sea surface height and significant wave height may significantly increase the probability of high-erosion and high-accretion states. Moreover, by altering the wave direction, the water quality variables show significant changes in the higher-risk class. The outcome of this study allowed us to identify current and future coastal risk scenarios, supporting local authorities in developing adaptation plans.

A Bayesian network-based approach for identifying risk factors and predicting ischemic stroke in infective endocarditis patients.

This study aimed to seek the risk factors and develop a predictive model for ischemic stroke (IS) in patients with infective endocarditis (IE) utilizing a Bayesian network (BN) approach. Data were obtained from the electronic medical records of all adult patients at three hospitals between 1 January 2018, and 31 December 2022. Two predictive models, logistic regression and BN, were used. Patients were randomly assigned to the training and test sets in a 7:3 ratio. We established a BN model with the training dataset and validated it with the testing dataset. The Bayesian network model was built by using the Tabu search algorithm. The areas under the receiver operating characteristic curve (AUCs), calibration curve, and decision curve were used to evaluate the prediction performance between the BN and logistic models. A total of 542 patients [mean (SD) age, 49.6 (15.3) years; 137 (25.3%) female] were enrolled, including 151 (27.9%) with IS and 391 (72.1%) without IS. Hyperlipidemia, hypertension, age, vegetation size (>10 mm), S. aureus infection, and early prosthetic valve IE were closely correlated with IS. The BN models outperformed the logistic regression in training and testing sets, with accuracies of 76.06% and 74.1%, AUC of 0.744 and 0.703, sensitivities of 25.93% and 20.93%, and specificities of 96.27% and 90.24%, respectively. The BN model is more efficient than the logistic regression model. Therefore, BN models may be suitable for the early diagnosis and prevention of IS in IE patients.