Bayesian Network Method Research Articles

Signal detection of drugs associated with obstructive and central sleep apnoea

We aim to discover new safety signals of drug-induced sleep apnoea (SA), a global health problem affecting approximately 1 billion people worldwide.We first conducted a series of sequence symmetry analyses (SSA) in a cohort composed from all patients who received a first SA diagnosis or treatment between 2006 and 2018 in the Echantillon Généraliste des Bénéficaires (EGB), a random sample of the French healthcare database. We used two primary outcomes to estimate the sequence ratio (SR) for all drug classes available in France: a sensitive one (diagnosis or treatment of SA) and a specific one (Positive Airway Pressure (PAP) therapy). We then performed disproportionality analyses using the “Bayesian neural network method” on all cases of sleep apnoea (MedDRA high level term) reported up to November 2023 in the World Health Organisation (WHO) pharmacovigilance database.Among the 728,167 individuals, 46,193 had an incident diagnosis or treatment for SA and 17,080 had started an incident treatment by PAP therapy. Fifty-eight drug classes had a significant SR, with 7 considered highly plausible: opium alkaloids and derivatives, benzodiazepine derivatives, other centrally acting agents, other anxiolytics, carbamic acid esters, quinine and derivatives and antivertigo preparations; with consistent signals found for the first 3 drug classes in the disproportionality analysis.In this signal detection study, we found that opioids, benzodiazepines (but not Z-drugs) and myorelaxing agents are associated with the onset or aggravation of SA. Moreover, a new safety signal for antivertigo preparations such as betahistine emerged and needs to be further explored.

A human reliability analysis method based on STPA-IDAC and BN-SLIM for driver take-over in Level 3 automated driving

Human factors play an important role in the take-over process of Level 3 (L3) automated driving. This paper combines Systems Theoretic Process Analysis (STPA) and Information, Decision and Action in Crew context (IDAC) for qualitative analysis and Bayesian Network (BN) and Success Likelihood Index Method (SLIM) for quantitative calculation to obtain the main performance shaping factors (PSFs) and evaluation indicators that cause human errors. Firstly, the STPA-IDAC method is used to analyze unsafe control actions (UCAs) for take-over process and form the mapping relationship of UCAs-IDA-PSFs. Secondly, the BN of human reliability analysis for take-over process is constructed based on the BN-SLIM method. Uncertainty in rates of PSFs and evaluation indicators is addressed in a probabilistic manner using expert opinions and empirical data. After diagnostic reasoning of BN, mean variation is used to identify the main PSFs and evaluation indicators. This method can effectively identify the main PSFs and evaluation indicators that cause human errors, facilitate risk assessment and management, and reduce the human error probability (HEP).

Data-Driven Bayesian Risk Assessment of Factors Influencing the Severity of Marine Accidents in Port Areas

The most prevalent types of ship accidents in port areas are allisions, collisions, and groundings. A comprehensive risk assessment is needed to prevent and mitigate these accidents and their consequences. This study evaluates the risk of such accidents in port areas by presenting a model that elucidates the relationship between risk-identifying factors (RIFs) and accident severity. In this context, the RIFs are determined by analyszing the reports of 528 accidents that occurred in port areas between 1995 and 2023. Subsequently, the model is created by analysing the data derived from these reports using the Tree Augmented Naive Bayes (TAN) algorithm, through an analysis of the data derived from these reports using the Tree Augmented Naive Bayes (TAN) algorithm, which is an approach of the data-driven Bayesian network method. The findings of the study indicate that accident type, wind, ship age, and vessel type are the most influential factors in predicting the severity of accidents in port areas. It is thought that the model will assist port authorities in identifying operational risks contributing to accidents and that contribute to accidents and in formulating preventive regulations.

The time lag in local field potential signals for the development of its Bayesian belief network

PurposeThe objective is to suggest time as an important variable to consider in the network model, specifically when discussing causality.MethodsThere is a consideration of the context of functional connectivity because of the time importance of observing the feature inside the neuroscience context. A network model was constructed using the Bayesian network method, utilizing a dataset consisting of three rats’ local field potentials. The model took into consideration the time delay of communication among brain areas, as recorded in this study. In pursuit of this objective, the delayed mutual information method was employed to ascertain the temporal delay between local field potentials and K2 score for the purpose of model comparison.ResultsBayesian network depicted the probabilistic relationship among rat’s brain areas. Delayed mutual information captured the lag among brain areas, and after its appliance on the Bayesian network model, posed better results.ConclusionThe primary novelty of this research lies in its integration of minor delays within the Bayesian network approach, accomplished through the utilization of the delayed mutual information technique prior to its implementation. The suggested methodology incorporates an essential feature that supports the analysis of functional connectivity among brain areas, thereby providing support for the dynamics of neurophysiology.

Fault diagnosis method of Bayesian network based on association rules

Fault diagnosis method of Bayesian network based on association rules

Rhabdomyolysis associated with concomitant use of colchicine and statins in the real world: identifying the likelihood of drug-drug interactions through the FDA adverse event reporting system.

Currently, there remains substantial controversy in research regarding whether the concomitant use of colchicine and statins increases the occurrence of rhabdomyolysis, warranting further substantiation. This study aimed to identify the likelihood drug-drug interactions (DDIs) for the co-administration of colchicine and statins resulting in rhabdomyolysis. A disproportionality analysis was conducted by using data sourced from the US Food and Drug Administration Adverse Event Reporting System (FAERS) to detect rhabdomyolysis signals associated with the combined use of colchicine and statins. The association between (colchicine/statins/colchicine and statins) and rhabdomyolysis were evaluated using information component (IC). DDI signals were calculated based on the Ω shrinkage measure and Bayesian confidence propagation neural network (BCPNN) method. Furthermore, stratification was performed based on colchicine and individual statins agents. In total, 11,119 reports of rhabdomyolysis were identified in the FAERS database, 255 (2.29%) involved both colchicine and statins. Our analysis showed potential DDI signals of rhabdomyolysis (Ω025 = 1.17) among individuals concurrent use of colchicine and statins. Moreover, further drug-specific analysis suggests DDI signals in the colchicine-atorvastatin pair (Ω025 = 1.12), and colchicine-rosuvastatin pair (Ω025 = 1.05), along with a higher proportion of rhabdomyolysis (IC025 = 5.20) and (IC025 = 4.26), respectively. The findings suggest that concomitant use of colchicine and statins may increase the risk of rhabdomyolysis, particularly when combined with atorvastatin or rosuvastatin. Therefore, healthcare professionals should pay special attention to life-threatening AE such as rhabdomyolysis, when co-prescribing colchicine statins.

A real-world pharmacovigilance study of esketamine nasal spray.

Mining and updating the post-marketing safety signals of esketamine nasal spray for better identification of adverse drug event (ADE) signals and medication monitoring during clinical use to ensure patient medication safety. Downloading data from the US Food and Drug Administration Adverse Event Reporting System from Q1 2019 to Q2 2023, the reporting odds ratio, proportional reporting ratio, Multi-item Gamma Poisson Shrinker, and Bayesian Confidence Propagation Neural Network methods of the disproportionality method were used to mine and analyze ADEs, and finally to screen for signals of ADEs with esketamine nasal spray as the primary suspected drug. The Preferred Terminology of the Medical Dictionary of Regulatory Activities (version 26.0) was used to standardize the description of ADEs and to attribute ADEs to the System Organ Classification. A total of 5132 ADEs reports of esketamine nasal spray as the primary suspected drug were obtained from the Food and Drug Administration Adverse Event Reporting System. The most frequently observed ADEs are dissociation, sedation, and hypertension, while some new rare signals have been detected, such as interstitial cystitis, substance abuse, and drug diversion. The present study identified significant new ADEs signals for esketamine nasal spray, which may provide a source for healthcare professionals to assess patients' symptoms and risk identification.

Signal mining of adverse events of proteasome inhibitors in multiple myeloma based on FAERS.

To mine and analyze adverse events (AEs) related to proteasome inhibitors in multiple myeloma based on the FDA Adverse Event Reporting System (FAERS), providing references for rational clinical medication. AE data related to multiple myeloma proteasome inhibitors were collected from the FAERS from the first quarter of 2010 to the first quarter of 2024. Signal mining of AEs was conducted using the reporting odds ratio method and Bayesian confidence propagation neural network method. A total of 8,805 reports for bortezomib, 5,264 for carfilzomib, and 8,771 for ixazomib were collected, with corresponding AE signals of 474, 279, and 287, respectively, involving 23, 21, and 22 System Organ Classes (SOCs). The report information for the three drugs tended to be consistent: more cases were reported in males than in females; the majority of patients were 65 years and over; AEs mostly occurred within 6months of medication; the outcomes primarily consisted of hospitalization, prolonged hospital stay, and other serious adverse events; the primary reporting country was the United States. The most affected SOCs were infections and infestations, general disorders and administration site conditions, and blood and lymphatic system disorders. The overall distribution of AEs for the three multiple myeloma proteasome inhibitors was consistent, but there were certain differences in specific AE signal characteristics, which should be noted in clinical applications.

Post-earthquake functional state assessment of communication base station using Bayesian network

The reliability and resilience of communication base stations are critical to the post-earthquake performance of the communication system, and consequently influence the communication, rescue, and emergence management after an earthquake. There is a lack of models that can fully evaluate the post-earthquake functional states of base stations with the consideration of the dependencies between different components. This paper proposes a Bayesian network method to evaluate the post-earthquake functionality of communication base stations. The method considers the dependence between the equipment and its hosting building structure, and the impact of power outages. This model produces seismic functional fragility curves for typical base stations that enable both qualitative and quantitative evaluations of base station functionality. The model is validated using seismic damage data from the Ludian Earthquake. It was found that the proposed model can reasonably predict the post-earthquake functional failure of base stations, in good agreement with the observed seismic damage data. The application of this model provides support for the operation and maintenance of communication base stations, and insights for managing seismic risk and resilience of the communication system.

Comprehensive analysis of ship sinking accidents using Bayesian network

ABSTRACT Sinking accidents can cause catastrophic losses in human, economic resources and environmental damage. Sinking accidents are one of the most common maritime accidents that occur in Indonesia. Therefore, this study aims to analyze the cause-and-effect probability of ship sinkings in Indonesian waters. Bayesian Network method and accident factor level are used to analyze the causal factors of ship sinking accidents. The results found that the Loss of Stability of the vessel is the most sensitive node. An additional take from the investigated sinking is that half of the accidents occurred on passenger vessels, and ships over 25 years old are prone to sinking due to corrosion and lack of proper maintenance. While 70.5% of accident caused by human error that is majorly triggered by unsafe acts. The real-case analysis also performed in this study reveals that eliminating Human Error and ship’s associated factors can reduce the probability of sinking by 17.2%.

Predicting 28Si projectile fragmentation cross sections with Bayesian neural network method

Abstract This study utilizes the Bayesian neural network (BNN) method in machine learning to learn and predict the cross-section data of 28Si projectile fragmentation for different targets at different energies, and to quantify the uncertainty. The detailed modeling process of BNN is presented, and its prediction results are compared with Cummings, Nilsen, EPAX2, EPAX3, and FRACS models, as well as experimental measurement values. The results reveal that the BNN method achieves the smallest root-mean-square error (RMSE) and the highest agreement with the experimental values compared to other models. Only the BNN method and FRACS model show a significant odd-even staggering effect, but BNN method is more closer to the experimental value. Furthermore, the BNN method is the only model capable of reproducing data features with low cross-section values at Z = 9, and the average ratio of the BNN's predicted to experimental values is close to 1.0. These results demonstrate that the BNN method can accurately reproduce and predict the fragment production cross sections of the 28Si projectile fragmentation, as well as its ability to capture key data characteristics.

A disproportionality analysis of CDK4/6 inhibitors in the FDA Adverse Event Reporting System (FAERS)

ABSTRACT Objective The FDA Adverse Event Reporting System (FAERS) was used to mine and evaluate adverse events (AEs) associated with cyclin-dependent kinase (CDK) 4/6 inhibitors, thereby providing a reference for clinical rational drug use. Methods AE data related to CDK4/6 inhibitors from the first quarter of 2015 to the first quarter of 2023 were acquired from FAERS, while the signal mining was processed using the reporting odds ratio (ROR) method and Bayesian confidence propagation neural network (BCPNN) method. Results The number of AE reports for CDK4/6 inhibitors was, respectively, 132,494 for palbociclib, 56,151 for ribociclib, and 7,014 for abemaciclib. The corresponding numbers of AE signals were 319, 517, and 59, with the number of involved System Organ Class (SOC) being 23, 23, and 15, mainly involving blood and lymphatic system disorders, respiratory, thoracic and mediastinal disorders, hepatobiliary disorders, skin and subcutaneous tissue disorders, etc. Conclusion CDK4/6 inhibitors could lead to pulmonary toxicity, myelosuppression, skin reactions, etc. Special attention should be paid to abemaciclib for interstitial lung disease (ILD), erythema multiforme, and thrombosis risk; ribociclib for cardiac toxicity, hepatotoxicity, and musculoskeletal toxicity; palbociclib for neurocognitive impairment and osteonecrosis of the jaw.

Phylogeographic Analysis of Mycobacterium kansasii Isolates from Patients with M. kansasii Lung Disease in Industrialized City, Taiwan.

Little is known about environmental transmission of Mycobacterium kansasii. We retrospectively investigated potential environmental acquisition, primarily water sources, of M. kansasii among 216 patients with pulmonary disease from an industrial city in Taiwan during 2015-2017. We analyzed sputum mycobacterial cultures using whole-genome sequencing and used hierarchical Bayesian spatial network methods to evaluate risk factors for genetic relatedness of M. kansasii strains. The mean age of participants was 67 years; 24.1% had previously had tuberculosis. We found that persons from districts served by 2 water purification plants were at higher risk of being infected with genetically related M. kansasii isolates. The adjusted odds ratios were 1.81 (1.25-2.60) for the Weng Park plant and 1.39 (1.12-1.71) for the Fongshan plant. Those findings unveiled the association between water purification plants and M. kansasii pulmonary disease, highlighting the need for further environmental investigations to evaluate the risk for M. kansasii transmission.

Adverse events related to neuromuscular blocking agents: a disproportionality analysis of the FDA adverse event reporting system.

Background: Neuromuscular blocking agents (NMBAs) are primarily used during surgical procedures to facilitate endotracheal intubation and optimize surgical conditions. This study aimed to explore the adverse event signals of NMBAs, providing reference for clinical safety. Methods: This study collected reports of atracurium, cisatracurium, rocuronium, and vecuronium as primary suspect drugs in The US Food and Drug Administration Adverse Event Reporting System (FAERS) from the first quarter of 2004 to the third quarter of 2023. The adverse events (AEs) reported in the study were retrieved based on the Preferred Terms (PTs) of the Medical Dictionary for Regulatory Activities. In addition, we conducted disproportionality analysis on relevant reports using the reporting odds ratio (ROR) method and Bayesian confidence propagation neural network (BCPNN) method. A positive signal was generated when both algorithms show an association between the target drug and the AE. Results: A total of 11,518 NMBA-related AEs were reported in the FAERS database. The most AEs of rocuronium were collected. NMBA-related AEs involved 27 different system organs (SOCs), all of the four NMBAs had positive signals in "cardiac disorders," "immune system disorders," "respiratory, thoracic and mediastinal disorders" and "vascular disorders." At the PTs level, a total of 523 effective AEs signals were obtained for the four NMBAs. AEs labled in the instructions such as anaphylaxis (include anaphylactic reaction and anaphylactic shock), bronchospasm, respiratory arrest and hypotension were detected positive signals among all NMBAs. In addition, we also found some new AEs, such as ventricular fibrillation for the four NMBAs, hyperglycaemia for atracurium, kounis syndrome and stress cardiomyopathy for rocuronium, hepatocellular injury for cisatracurium, hyperkalaemia for vecuronium. To further investigated the AEs associated with serious clinical outcomes, we found that cardiac arrest and anaphylaxis were the important risk factors for death due to NMBAs. Conclusion: NMBA-related AEs have a significant potential to cause clinically severe consequences. Our study provides valuable references for the safety profile of NMBAs, and considering the limitations of the FAERS database, further clinical data are needed to validate the findings of this study.

A novel bayesian network-based ensemble classifier chains for multi-label classification

In this paper, we address the challenges of random label ordering and limited interpretability associated with Ensemble Classifier Chains (ECC) by introducing a novel ECC method, ECC-MOO&BN, which integrates Bayesian Networks (BN) and Multi-Objective Optimization (MOO). This approach is designed to concurrently overcome these ECC limitations. The ECC-MOO&BN method focuses on extracting diverse and interpretable label orderings for the ECC classifier. We initiated this process by employing mutual information to investigate label relationships and establish the initial structures of the BN. Subsequently, an enhanced NSGA-II algorithm was applied to develop a series of Directed Acyclic Graphs (DAGs) that effectively balance the likelihood and complexity of the BN structure. The rationale behind using the MOO method lies in its ability to optimize both complexity and likelihood simultaneously, which not only diversifies DAG generation but also helps avoid overfitting during the production of label orderings. The DAGs, once sorted topologically, yielded a series of label orderings, which were then seamlessly integrated into the ECC framework for addressing multi-label classification (MLC) problems. Experimental results show that when benchmarked against eleven leading-edge MLC algorithms, our proposed method achieves the highest average ranking across seven evaluation criteria on nine out of thirteen MLC datasets. The results of the Friedman test and Nemenyi test also indicate that the performance of the proposed method has a significant advantage compared to other algorithms.

Development of an improved Bayesian network method for maritime accident safety assessment based on multiscale scenario analysis theory

The strait is infrastructure for connecting ports, ensuring the national economy and human life. Assessing maritime safety is crucial for strait operation and maintenance. Current methods over-rely on statistical analysis of historical data, which lacks scientific algorithms and cannot accurately assess safety timeliness. Therefore, this study proposes an improved Bayesian network (BN) model based on the multiscale scenario analysis theory to process to the maritime safety assessment process. Firstly, the maritime safety assessment model is constructed based on the knowledge element theory (KET) and scenario analysis theory (SAT). Secondly, a heuristic learning algorithm is used to construct the BN model of maritime accidents to define scenario elements and key parameters. Thirdly, a scenario simulation is conducted with the Bohai Strait as a case to assessment its emergency capability in catastrophic accidents. This paper successfully developed a maritime accident safety assessment framework using an improved BN algorithm within a scenario analysis perspective, significantly enhancing the accuracy and reliability of safety assessments. This methodology used in this study can be used a tool suitable for the safety assessment of major infrastructure accidents at the national level and the results obtained can serve as a reference for dynamically revising emergency plans.

Recommendations for the selection of nucleoside analogues as antihuman herpesvirus drugs: a real-world analysis of reported cases from the FDA adverse event reporting system

ABSTRACT Objective The aim of this study is to provide guidance for refining medication protocols, developing alternative strategies, and enhancing protection against herpesvirus infections in personalized clinical settings. Methods Adverse drug events (ADEs) data for anti-herpesvirus from the first quarter of 2004 to the fourth quarter of 2022 were collected from the FDA Adverse Event Reporting System (FAERS). Disproportionality analysis was performed using Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), and Bayesian Confidence Propagation Neural Network (BCPNN) methods for data mining. Results A total of 18,591, 24,206, 6,150, and 419 reports of ADEs associated with acyclovir (ACV), valacyclovir (VACV), ganciclovir (GCV), and famciclovir (FCV) were screened and extracted from the FAERS. In this study, the report summarized the high frequency and strong correlation of ADEs for the four drugs at the Preferred Term (PT) level. Additionally, the analysis also identified the relationship between ADEs and factors such as age, gender, and severity of outcome at the System Organ Class (SOC) level. Conclusion The safety reports for the four-nucleoside analogue anti-herpesvirus drugs are diverse and interconnected. Dosing for patients with herpesvirus infections should be tailored to their specific conditions and the potential risk of disease.

Robustness of Structural Health Monitoring Systems for Offshore Energy Structures based on Damage Characteristics

Offshore energy structures such as fixed/floating wind turbines, tidal turbines and wave energy converters are exposed to loads coming from different sources. Recent advances in SHM offer many unique opportunities to assess the structural integrity of offshore energy infrastructure. The process of SHM generally involves the use of a system of sensors mounted on a structure; the processing of signals received from the sensors to attain damage sensitive features; and the detection of damage in the structure. Over the past decades, most of the research in SHM has been focused on improving and developing new sensor technology, signal processing, and damage detection algorithms. However, there are very few studies assessing the performance of SHM systems with respect to damage detection, localization, and quantification. Moreover, most of the performance analysis models for SHM systems rely only on 'probability of detection (POD)', that is an assessment index used in NDT to evaluate the probability of detecting a damage as a function of its size. For SHM systems, the replicability of POD is influenced by the environment, measurement noise, and the deterioration of sensors and instruments. Therefore, the distribution of POD is often determined by performing an extensive number of test sets with existing damage of varying magnitude. This paper aims to reduce the volume of experimental data required, as well as the uncertainties associated with POD prediction by using a model-assisted probability of detection (MAPOD) approach. We develop a Bayesian network (BN) method to evaluate the performance of SHM systems in terms of damage detection, localization, and assessment. In addition to POD, the probability of accurate localization (POAL) and probability of accurate assessment (POAA) are also incorporated into the analysis. The BN and limit state functions are applied to a fixed offshore energy structure equipped with a vibration based SHM system.

A Probabilistic Data Recovery Framework Against Load Redistribution Attacks Based on Bayesian Network and Bias Correction Method

A Probabilistic Data Recovery Framework Against Load Redistribution Attacks Based on Bayesian Network and Bias Correction Method

Time-varying dynamic Bayesian network learning for an fMRI study of emotion processing.

This article presents a novel method for learning time-varying dynamic Bayesian networks. The proposed method breaks down the dynamic Bayesian network learning problem into a sequence of regression inference problems and tackles each problem using the Markov neighborhood regression technique. Notably, the method demonstrates scalability concerning data dimensionality, accommodates time-varying network structure, and naturally handles multi-subject data. The proposed method exhibits consistency and offers superior performance compared to existing methods in terms of estimation accuracy and computational efficiency, as supported by extensive numerical experiments. To showcase its effectiveness, we apply the proposed method to an fMRI study investigating the effective connectivity among various regions of interest (ROIs) during an emotion-processing task. Our findings reveal the pivotal role of the subcortical-cerebellum in emotion processing.