Bayesian Network Research Articles

Gaussian mixture model and Bayesian convolutional neural network for abnormal musculoskeletal radiographs classification

Gaussian mixture model and Bayesian convolutional neural network for abnormal musculoskeletal radiographs classification

DB-Net and DVR-Net: Optimized New Deep Learning Models for Efficient Cardiovascular Disease Prediction

Cardiovascular Disease (CVD) is one of the main causes of death in recent years. To overcome the challenges faced during diagnosing CVD at an early stage, deep learning has been used. With advancements in technology, the clinical practice in the health care industry is likely to transform significantly. To predict CVD, we constructed two models: Dense Belief Network (DB-Net) and Deep Vanilla Recurrent Network (DVR-Net). Proximity Weighted Random Affine Shadow sampling balancing technique is used for balancing the highly imbalanced Heart Disease Health Indicator dataset. SHapley Additive exPlanations exhibits each feature’s contribution. It is used to visualize features contribution to the output of DB-Net and DVR-Net in CVD prediction. Furthermore, 10-Fold Cross Validation is performed for evaluating the proposed models performance. Cross-dataset evaluation is also conducted on proposed models to see how well our proposed models generalize on unseen data. Various evaluation measures are used for assessment of models. The proposed DB-Net outperforms all the base models by achieving an accuracy of 91%, F1-score of 91%, precision of 93%, recall of 89%, and execution time of 1883 s on 30 epochs with batch size 32. The DVR-Net beats the state-of-art models with an accuracy of 90%, F1-score of 90%, precision of 90%, recall of 90%, and execution time of 2853 s on 30 epochs with batch size 32.

Bayesian-Neural-Network-Based Approach for Probabilistic Prediction of Building-Energy Demands

Reliable prediction of building-level energy demand is crucial for the building managers to optimize and regulate energy consumption. Conventional prediction models omit the uncertainties associated with demand over time; hence, they are mostly inaccurate and unreliable. In this study, a Bayesian neural network (BNN)-based probabilistic prediction model is proposed to tackle this challenge. By quantifying the uncertainty, BNNs provide probabilistic predictions that capture the variations in the energy demand. The proposed model is trained and evaluated on a subset of the building operations dataset of Lawrence Berkeley National Laboratory (LBNL), Berkeley, California, which includes diverse attributes related to climate and key building-performance indicators. We have performed thorough hyperparameter tuning and used fixed-horizon validation to evaluate trained models on various test data to assess generalization ability. To validate the results, quantile random forest (QRF) was used as a benchmark. This study compared BNN with LSTM, showing that BNN outperformed LSTM in uncertainty quantification.

Comparative Safety of Ultrasound Enhancing Agents: A Systematic Review and Bayesian Network Meta-Analysis: Comparative Safety Evaluation of Optison.

Ultrasound enhancing agents (UEAs), including Optison, Definity, and Lumason, enhance the diagnostic performance of echocardiography, particularly in patients with suboptimal acoustic windows. However, there remains a paucity of comparative safety assessments among the Food and Drug Administration (FDA)-approved UEAs. A PRISMA-guided systematic literature review performed in August 2023 searched Medline, Embase, Cochrane Library, and grey literature. Randomized and non-randomized comparative evidence on safety of contrast enhanced procedures were included. A feasibility assessment ensured homogeneity across studies by comparing patient characteristics and outcomes. Bayesian hierarchical network meta-regression was employed to indirectly compare published safety outcomes across different UEAs. In addition, adverse events (AEs) between 2019-2023 for each UEA were retrieved from the FDA Adverse Events Reporting System (FAERS), and comparative safety outcomes were derived using annual UEA administration in the US. Screening of 4146 records rendered 19 studies for inclusion in the indirect comparison analysis after feasibility assessment. Optison demonstrated favorable safety in direct comparisons with Definity and Lumason across the majority of investigated AEs. Network meta-regression results on serious adverse events (SAEs) within 30 minutes revealed odds ratios (ORs) of 0.59 (95% confidence interval: 0.27-1.23) and 0.63 (0.29-1.33) for Optison vs. Lumason and Definity, respectively, indicating a decreased risk of SAEs for Optison vs. Lumason and Definity, further supported by statistically significant ORs within FAERS: 0.01 (0-0.02), P<0.001 and 0.15 (0.05-0.48), P=0.001, for Optison vs. Lumason and Definity, respectively. In conclusion, our results restate the favorable comparative safety profile of Optison, providing new evidence to inform clinical decision-making.

Artificial intelligence modeling of biomarker-based physiological age: Impact on phase 1 drug-metabolizing enzyme phenotypes.

Age and aging are important predictors of health status, disease progression, drug kinetics, and effects. The purpose was to develop ensemble learning-based physiological age (PA) models for evaluating drug metabolism. National Health and Nutrition Examination Survey (NHANES) data were modeled with ensemble learning to obtain two PA models, PA-M1 and PA-M2. PA-M1 included body composition, blood and urine biomarkers, and disease variables as predictors. PA-M2 had blood and urine-derived variables as predictors. Activity phenotypes for cytochrome-P450 (CYP) CYP2E1, CYP1A2, CYP2A6, xanthine oxidase (XO), and N-acetyltransferase-2 (NAT-2) and telomere attrition were assessed. Bayesian networks were used to obtain mechanistic systems pharmacology model structures for PA. The study included n = 22,307 NHANES participants (51.5% female, mean age 46.0 years, range: 18-79 years). The PA-M1 and PA-M2 distributions had greater dispersion across age strata with a right skew for younger age strata and a left skew for older age strata. There was no evidence of algorithmic bias based on sex or race/ethnicity. Klotho, lean body mass, glycohemoglobin, and systolic blood pressure were the top four predictors for PA-M1. Glycohemoglobin, serum creatinine, total cholesterol, and urine creatinine were the top four predictors for PA-M2. The models also performed satisfactorily in independent validation. Model-predicted PA was associated with CYP2E1, CYP1A2, CYP2A6, XO, and NAT-2 activity. Telomere attrition was associated with greater PA-M1 and PA-M2. Ensemble learning models provide robust assessments of PA from easily obtained blood and urine biomarkers. PA is associated with Phase I drug-metabolizing enzyme phenotypes.

Spinal Cord Stimulation vs Medical Management for Chronic Back and Leg Pain: A Systematic Review and Network Meta-Analysis.

Chronic back and lower extremity pain is one of the leading causes of disability worldwide. Spinal cord stimulation (SCS) aims to improve symptoms and quality of life. To evaluate the efficacy of SCS therapies compared with conventional medical management (CMM). MEDLINE, Embase, and Cochrane Library were systematically searched from inception to September 2, 2022. Selected studies were randomized clinical trials comparing SCS therapies with sham (placebo) and/or CMM or standard treatments for adults with chronic back or leg pain who had not previously used SCS. Evidence synthesis estimated odds ratios (ORs) and mean differences (MDs) and their associated credible intervals (CrI) through bayesian network meta-analysis. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline for network meta-analyses was followed. The primary outcomes were pain-related end points, including pain intensity (measured by visual analog scale) and proportion of patients achieving at least 50% pain relief (responder rate) in the back or leg. Quality of life (measured by EQ-5D index score) and functional disability (measured by the Oswestry Disability Index score) were also considered. A total of 13 studies of 1561 patients were included in the network meta-analysis comparing conventional and novel SCS therapies with CMM across the 6 outcomes of interest at the 6-month follow-up. Both conventional and novel SCS therapies were associated with superior efficacy compared with CMM in responder rates in back (conventional SCS: OR, 3.00; 95% CrI, 1.49 to 6.72; novel SCS: OR, 8.76; 95% CrI, 3.84 to 22.31), pain intensity in back (conventional SCS: MD, -1.17; 95% CrI, -1.64 to -0.70; novel SCS: MD, -2.34; 95% CrI, -2.96 to -1.73), pain intensity in leg (conventional SCS: MD, -2.89; 95% CrI, -4.03 to -1.81; novel SCS: MD, -4.01; 95% CrI, -5.31 to -2.75), and EQ-5D index score (conventional SCS: MD, 0.15; 95% CrI, 0.09 to 0.21; novel SCS: MD, 0.17; 95% CrI, 0.13 to 0.21). For functional disability, conventional SCS was superior to CMM (MD, -7.10; 95% CrI, -10.91 to -3.36). No statistically significant differences were observed for other comparisons. This systematic review and network meta-analysis found that SCS therapies for treatment of chronic pain in back and/or lower extremities were associated with greater improvements in pain compared with CMM. These findings highlight the potential of SCS therapies as an effective and valuable option in chronic pain management.

Network analysis: An overview for mental health research.

Network approaches to psychopathology have become increasingly common in mental health research, with many theoretical and methodological developments quickly gaining traction. This article illustrates contemporary practices in applying network analytical tools, bridging the gap between network concepts and their empirical applications. We explain how we can use graphs to construct networks representing complex associations among observable psychological variables. We then discuss key network models, including dynamic networks, time-varying networks, network models derived from panel data, network intervention analysis, latent networks, and moderated models. In addition, we discuss Bayesian networks and their role in causal inference with a focus on cross-sectional data. After presenting the different methods, we discuss how network models and psychopathology theories can meaningfully inform each other. We conclude with a discussion that summarizes the insights each technique can provide in mental health research.

Sensitivity of Bayesian Networks to Errors in Their Structure

Sensitivity of Bayesian Networks to Errors in Their Structure

Cyclin-dependent kinase 4/6 inhibitors and cardiotoxic events in breast cancer: A pharmacovigilance study based on the FAERS database.

Recent trials have highlighted the cardiotoxicity of ribociclib, a CDK4/6 inhibitor, particularly its association with QT prolongation. However, studies on the link between CDK4/6 inhibitors and cardiotoxic events show inconsistent results, and the factors influencing these events and related drug interactions remain underexplored. To address these uncertainties, our study utilizes the FDA adverse event reporting system database (Q1 2015 to Q1 2024) to examine the cardiotoxic events of CDK4/6 inhibitors in breast cancer patients. We employed a comprehensive analytical framework, applying disproportionality methods including Bayesian confidence propagation neural network, proportional reporting ratio, and reporting odds ratio. Our findings highlight significant variability in cardiotoxic events among different CDK4/6 inhibitors, with ribociclib (IC: 0.26, 95%CI: 0.18-0.34) exhibiting pronounced cardiotoxicity. Notably, ribociclib was associated with serious cardiotoxic events such as torsade de pointes/QT prolongation (IC: 2.11, 95% CI: 1.90-2.29) and conduction defects (IC: 2.07, 95% CI: 1.87-2.23). For the first time, palbociclib has been identified with positive signals for cardiotoxic events at the preferred terms level, including pulmonary oedema, increased blood pressure, myocardial infarction, and cardiac flutter. Moreover, multivariable logistic regression and Bayesian network analyses reveal that age, geographic location, and the number of concomitant medications significantly influence cardiotoxic events. Our study also highlights significant drug interactions that increase the probability of specific cardiotoxic outcomes, notably with drugs like sertraline, lansoprazole, capecitabine, and torasemide. These findings highlight the need for personalized treatment plans to mitigate cardiotoxic events and improve patient safety.

Association between different GLP-1 receptor agonists and acute pancreatitis: case series and real-world pharmacovigilance analysis

ObjectiveGlucagon-like peptide-1 receptor agonists (GLP-1 RAs) have shown notable advancements in managing blood sugar control. Nevertheless, there remains a gap in real-world data regarding the variation in acute pancreatitis (AP) risk among different GLP-1 RAs. Our study aimed to characterize and evaluate AP associated with different GLP-1 RAs (exenatide, lixisenatide, liraglutide, albiglutide, semaglutide, dulaglutide and tirzepatide) in a public adverse events database and to review the relevant case reports.MethodsWe described a case series of patients experiencing AP while on GLP-1 RAs. Additionally, we utilized various algorithms including reporting odds ratio (ROR), proportional reporting ratio (PRR), Bayesian confidence propagation neural network (BCPNN), and multi-item gamma Poisson shrinker (MGPS) to analyze data from the Food and Drug Administration’s Adverse Event Reporting System (FAERS) regarding suspected adverse events of AP linked to GLP-1 RAs from January 2005 to September 2023.ResultsOur case series comprised thirty-nine patients who experienced AP events while on GLP-1 RAs. Within the FAERS database, we retrieved a total of 6,751 individual case safety reports (ICSRs) involving various GLP-1 RAs. The median age of the patients included in our study was 57 years (range: 14–99), with 98.3% of cases classified as serious. Signals indicating AP were observed across all GLP-1 RAs, with particular emphasis on exenatide and liraglutide.ConclusionThere is a notable reporting signal of AP associated with all GLP-1 RAs. Healthcare providers must remain vigilant and closely monitor this potentially life-threatening adverse event.

Comparing the effectiveness of five traditional Chinese exercises in improving balance function in older adults: a systematic review and Bayesian network meta-analysis

Background Despite numerous studies affirming the potential of traditional Chinese exercises (TCEs) in enhancing balance in older adults, systematic integration is lacking. This study evaluated the effectiveness of five TCEs—Baduanjin, Liuzijue, Tai Chi, Wuqinxi, and Yijinjing—in improving balance among older adults using network meta-analysis. Methods This meta-analysis was registered in PROSPERO with the registration number CRD42023481450. Related articles indexed by Web of Science, Cochrane, PubMed, Embase, China National Knowledge Infrastructure (CNKI), Wanfang, and VIP databases before October 2023 were searched. Randomized controlled trials (RCTs) involving TCEs interventions to improve balance function conducted in older adults who aged ≥60 years were included. Two researchers used Review Manager to assess the quality of the studies, and analyzed the data using Stata and R. Results In total, 46 RCTs and 3,333 older adults were included. The aforementioned TCEs had positive effects on improving balance in older adults. Tai Chi revealed significant intervention effects in performing the Single-Leg Stand with eyes Closed (SLSC), 6-Min Walk Test (6MWT), and Short Form 36-Item Physical Component Summary (SF-36PCS). Liuzijue significantly improved performance in the Timed Up and Go Test (TUGT), 6MWT, SF-36PCS, and Berg Balance Scale (BBS). Baduanjin, Wuqinxi, and Yijinjing showed noteworthy intervention effects on the BBS. Tai Chi ranked highest in the SLSC, 6MWT, and SF-36PCS in the surface under the cumulative ranking, while Liuzijue and Yijinjing ranked highest in the TUGT and BBS, respectively. Conclusion Tai Chi, Liuzijue, and Yijinjing improved the static, dynamic, and overall balance outcomes, respectively. Older adults can make a reasonable choice among these TCEs based on their needs.

Novel statistically equivalent signature-based hybrid feature selection and ensemble deep learning LSTM and GRU for chronic kidney disease classification

Chronic kidney disease (CKD) involves numerous variables, but only a few significantly impact the classification task. The statistically equivalent signature (SES) method, inspired by constraint-based learning of Bayesian networks, is employed to identify essential features in CKD. Unlike conventional feature selection methods, which typically focus on a single set of features with the highest predictive potential, the SES method can identify multiple predictive feature subsets with similar performance. However, most feature selection (FS) classifiers perform suboptimally with strongly correlated data. The FS approach faces challenges in identifying crucial features and selecting the most effective classifier, particularly in high-dimensional data. This study proposes using the Least Absolute Shrinkage and Selection Operator (LASSO) in conjunction with the SES method for feature selection in CKD identification. Following this, an ensemble deep-learning model combining long short-term memory (LSTM) and gated recurrent unit (GRU) networks is proposed for CKD classification. The features selected by the hybrid feature selection method are fed into the ensemble deep-learning model. The model’s performance is evaluated using accuracy, precision, recall, and F1 score metrics. The experimental results are compared with individual classifiers, including decision tree (DT), Random Forest (RF), logistic regression (LR), and support vector machine (SVM). The findings indicate a 2% improvement in classification accuracy when using the proposed hybrid feature selection method combined with the LSTM and GRU ensemble deep-learning model. Further analysis reveals that certain features, such as HEMO, POT, bacteria, and coronary artery disease, contribute minimally to the classification task. Future research could explore additional feature selection methods, including dynamic feature selection that adapts to evolving datasets and incorporates clinical knowledge to enhance CKD classification accuracy further.

Comparison of different surgical strategies for cervical dystonia: Evidence from Bayesian network analysis.

Several surgical techniques have been used to treat cervical dystonia (CD), however, to date, the optimal surgical technique for CD remains controversial. We therefore conducted the first network meta-analysis to compare different surgical strategies for CD to inform clinical practice. Electronic databases were searched for surgical strategies for treating CD. The primary outcome was improvement in total Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) score. Subgroup analysis was performed to compare short-term (< 1 year) and long-term (≥ 1 year) outcomes. Safety outcomes included surgery-related adverse events (AEs). A total of 55 trials with 2032 patients employing five surgical strategies were identified, including globus pallidus internus (GPi)/subthalamic nucleus (STN)-deep brain stimulation (DBS), selective peripheral denervation (SPD), microvascular decompression (MVD) and pallidotomy. All strategies led to significant improvement in total TWSTRS score (mean improvement range 18.65-28.22). GPi-DBS showed significantly greater enhancement than SPD for the whole dataset (mean difference [MD] 7.03, 95% credible interval [Crl] 1.53-12.56), while both GPi-DBS (MD 8.05, 95% Crl 2.35-13.80) and STN-DBS (MD 10.71, 95% Crl 2.22-19.20) exhibited more long-term improvement than SPD. Regarding safety outcomes, GPi/STN-DBS and MVD were associated with fewer surgery-related AEs than SPD (ln odds ratio range -1.68 to -1.41). We conclude that DBS should be the preferred surgical option for CD, and the STN is a promising alternative target choice due to its comparable efficacy with the GPi. However, more direct evidence is still required.

Estimation of Cation Exchange Capacity for Low-Activity Clay Soil Fractions Using Experimental Data from South China

The cation exchange capacity (CEC) of the clay fraction (&lt;2 μm), denoted as CECclay, serves as a crucial indicator for identifying low-activity clay (LAC) soils and is an essential criterion in soil classification. Traditional methods of estimating CECclay, such as dividing the whole-soil CEC (CECsoil) by the clay content, can be problematic due to biases introduced by soil organic matter and different types of clay minerals. To address this issue, we introduced a soil pedotransfer functions (PTFs) approach to predict CECclay from CECsoil using experimental soil data. We conducted a study on 122 pedons in South China, focusing on highly weathered and strongly leached soils. Samples from the B horizon were used, and eight models and PTFs (four machine learning methods, multiple linear regression (MLR) and three PTFs from publication) were evaluated for their predictive performance. Four covariate datasets were combined based on available soil data and environmental variables and various parameters for machine learning techniques including an artificial neural network, a deep belief network, support vector regression and random forest were optimized. The results, based on 10-fold cross-validation, showed that the simple division of CECsoil by clay content led to significant overestimation of CECclay, with a mean error of 14.42 cmol(+) kg−1. MLR produced the most accurate predictions, with an R2 of 0.63–0.71 and root mean squared errors (RMSE) of 3.21–3.64 cmol(+) kg−1. The incorporation of environmental variables improved the accuracy by 2–10%. A linear model was fitted to enhance the current calculation method, resulting in the equation: CECclay = 15.31 + 15.90 × (CECsoil/Clay), with an R2 of 0.41 and RMSE of 4.48 cmol(+) kg−1. Therefore, given limited soil data, the MLR PTFs with explicit equations were recommended for predicting the CECclay of B horizons in humid subtropical regions.

Abstract 4123373: Estimating obesity-attributable cardiovascular disease burden in the United States: a Bayesian network analysis

Background: Obesity is a known contributor to the development of cardiovascular disease (CVD) and increases the risk of subsequent mortality. To fully understand the burden of obesity, accurate information regarding its effects on CVD is necessary. Current estimates are limited by a focus on direct effects, use of non-representative data sources, and a reliance on inaccurate diagnosis codes. Aim: To determine the size of CVD burden attributable to obesity in the US and identify the mechanisms through which obesity affects CVD. Methods: This study used data collected from US adults (aged ≥35 years) participating in the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2020 and US Census projections. NHANES data were longitudinally adapted, and participants were excluded if they had missing data for weight history, risk factors, or a given outcome of interest. Bayesian networks (BNs) mapping out the relationship between risk factors and each outcome were constructed to estimate current CVD risk and expected CVD risk in the absence of obesity. Obesity-attributable CVD burden was then calculated for each outcome, and the direct and indirect pathways through which changes in CVD risk occur were characterized. A subgroup analysis was conducted in non-smoking participants aged ≤65 years for whom obesity is likely the primary CVD risk factor. Results: The final sample included data from approximately 29,000 participants. Of these, 1286 (4.3%) had a history of myocardial infarction (MI), 1283 (4.4%) had a history of stroke, 1092 (3.7%) had a history of heart failure (HF), and 1992 (6.7%) had a history of CVD (composite of MI, stroke, angina, coronary heart disease, and HF). Results from the fitted BN indicate that 21% of MI, 16% of stroke, 38% of HF, and 19% of CVD cases were attributable to obesity, with ≥80% of this burden due to the effect of obesity on intermediate risk factors (hypertension, type 2 diabetes, and prior cardiovascular events). In the subgroup analysis, the obesity-attributable CVD burden was approximately twice as high for non-smokers aged ≤65 years, with 39% of MI, 32% of stroke, 65% of HF, and 38% of CVD cases attributable to obesity. Conclusion: Through a sophisticated network modeling approach, we generated estimates of obesity-attributable CVD burden in the US. Results highlight the importance of obesity as a CVD risk factor, particularly in younger non-smokers, and provide insight into the nature of this relationship.

Bayesian mesh optimization for graph neural networks to enhance engineering performance prediction

Abstract In engineering design, surrogate models are widely employed to reduce the computational costs of simulations by approximating design variables and geometric parameters from computer-aided design (CAD) models. However, traditional surrogate models often lose critical information when simplified to lower dimensions, and face challenges in handling the complexity of 3D shapes, especially in industrial datasets. To address these limitations, we propose a Bayesian graph neural network (GNN) framework that directly learns geometric features from CAD mesh representations for accurate engineering performance prediction. Our framework leverages Bayesian optimization (BO) to dynamically determine the optimal mesh element size, significantly improving model accuracy while balancing computational efficiency. This approach addresses the challenge of mesh quality by optimizing mesh resolution, ensuring that critical geometric features are preserved in 3D deep-learning-based surrogate models. Unlike existing models that rely on static or predefined mesh resolutions, our method adapts mesh size based on the task, making it highly flexible for a wide range of engineering applications. Experimental results demonstrate that mesh quality directly impacts prediction accuracy, and the proposed BO-EI GNN model outperforms state-of-the-art models such as 3D CNN, SubdivNet, GCN, and GNN in predicting mass, rim stiffness, and disk stiffness. Our method also significantly reduces the computational costs compared to traditional optimization techniques. The proposed framework shows promising potential for application in finite element analysis (FEA) and other mesh-based simulations, enhancing the utility of surrogate models across various engineering fields.

A real−world pharmacovigilance study of FDA Adverse Event Reporting System events for pralsetinib

BackgroundPralsetinib, a selective oral inhibitor of rearranged during transfection (RET) fusion proteins and oncogenic RET mutants, has shown significant efficacy in treating RET fusion-positive non-small cell lung cancer and thyroid cancer. However, since pralsetinib was approved in the United States in September 2020, there have been limited reports of post-marketing adverse events (AEs). In this study, we aimed to analyze the AE signals with pralsetinib on the basis of the United States Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) to provide instructions in clinical practice.MethodsAll AE reports were obtained from the FAERS database from the first quarter (Q3) of 2020 to the second quarter (Q2) of 2024. Various signal quantification techniques were used for analysis, including reporting odds ratios, proportional reporting ratios, Bayesian confidence propagation neural network, and multi-item gamma Poisson shrinker (MGPS)-based empirical Bayesian geometric mean.ResultsOut of 8,341,673 case reports in the FAERS database, 1,064 reports of pralsetinib as the “primary suspected (PS)” AEs were recorded, covering 26 system organ classes and 256 preferred terms. Of the reports, 62.5% were from consumers rather than healthcare professionals. The most common systems were general disorders and administration site conditions (n = 704), investigations (n = 516), and gastrointestinal disorders (n = 405). A total of 95 significant disproportionality preferred terms (PTs) conformed to the four algorithms simultaneously. AEs that ranked the top three at the PT level were hypertension (n = 80), asthenia (n = 79), and anemia (n = 65). Of the 95 PTs with significant disproportionation, unexpected significant AEs such as increased blood calcitonin, increased myocardial necrosis marker, and bacterial cystitis were observed, which were not mentioned in the drug’s instructions. The median onset time of pralsetinib-associated AEs was 41 days [interquartile range (IQR) 14–86 days]. The majority of the AEs occurred in 30 days (42.86%).ConclusionOur pharmacovigilance analysis of real-world data from the FEARS database revealed the safety signals and potential risks of pralsetinib usage. These results can provide valuable evidence for further clinical application of pralsetinib and are important in enhancing clinical medication safety.

Efficacy and safety of complement inhibitors in patients with geographic atrophy associated with age-related macular degeneration: a network meta-analysis of randomized controlled trials

ImportanceClinical trials in recent years have shown significant effectiveness of complement inhibitors for geographic atrophy (GA) treatment. Two complement inhibitor drugs have been approved by the Food and Drug Administration (FDA).Objectiveto compare and rank the different complement inhibitors in the treatment of GA secondary to age-related macular degeneration (AMD).Data sourcesA systematic literature search was conducted in the Cochrane Central, Web of Science Core Collection, PubMed, LWW Medical Journals, ClinicalTrials.gov, and WHO ICTRP from inception to October 2023.Study selectionAll randomized clinical trials evaluating the effectiveness of complement inhibitors in patients diagnosed with secondary GA in AMD were identified.Data extraction and synthesisThis study followed Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) network meta-analysis Checklist of Items and the Cochrane Risk of Bias Assessment Tool for assessing the study quality. Multiple authors independently coded all titles and abstracts, reviewed full-text articles against the inclusion and exclusion criteria, and resolved all discrepancies by consensus. Random-effects network meta-analyses were applied. Bayesian network meta-analysis was performed using the BUGSnet package in R (4.2.0).Main outcomes and measuresThe primary efficacy outcome was the change in GA lesion size (mm2) from baseline to month 12. The secondary efficacy outcome was the mean change in best-corrected visual acuity (BCVA) from baseline to month 12. Safety outcome measures included the number of subjects with serious adverse events (SAEs) and macular neovascularization (MNV).ResultsTen randomized controlled trials including 4,405 participants and five complement inhibitors were identified. Comparison with sham and SUCRA analysis showed that avacincaptad pegol 2 mg (MD: −0.58, 95% CrI: −0.97 to −0.18, SUCRA: 93.55), pegcetacoplan monthly (MD: −0.38, 95% CrI: −0.57 to −0.20, SUCRA: 81.37), and pegcetacoplan every other month (MD: −0.30, 95% CrI: −0.49 to −0.11, SUCRA: 70.16) have significant changes in GA lesion reduction. No treatments showed significant changes in BCVA and SAE compared with sham. Pegcetacoplan monthly (OR: 4.30, 95% CrI: 1.48–16.72) increased the risk of MNV. Avacincaptad pegol 2 mg demonstrated favorable outcomes in terms of SAE and MNV.Conclusion and relevanceAvacincaptad pegol 2 mg is the most effective complement inhibitor with better safety for the treatment of GA secondary to AMD.Systematic Review Registrationhttps://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42022351515, Identifier PROSPERO CRD42022351515.

Human errors analysis for remotely controlled ships during collision avoidance

To address human errors in collision avoidance tasks of remotely controlled ships, this study aims to develop a comprehensive framework for human error analysis within the context of autonomous ships. Firstly, the Hierarchical Task Analysis method is utilized to identify crew collision avoidance tasks associated with the traditional ship, and these tasks are then dissected into different operational stages using the Information Decision Action in a Crew cognitive model. Secondly, a combination of the fault hypothesis method and expert opinions are used to identify potential human error that may occur during collision avoidance operations of remotely controlled ships. Thirdly, an integrated approach is proposed to build a quantitative risk assessment model, which combines Failure Mode and Effects Analysis, Evidential Reasoning, and Belief rules-based Bayesian Network. Then, axiomatic analysis is used to verify the robustness and applicability of the risk assessment model. Finally, based on the results of quantitative risk assessment, specific measures are proposed for enhancing the safety of collision avoidance process of remotely controlled ships. The findings show that uncoordinated interactions of human-computer systems during the decision-making stage are a pivotal factor in the collision avoidance process. Therefore, future design efforts for remote-control centre should prioritize improving the clarity of human-computer interaction interfaces.

Efficacy and safety of first-line immune checkpoint inhibitor combination therapies in patients with advanced esophageal squamous cell carcinoma: a network meta-analysis

BackgroundWe performed a network meta-analysis of phase III trials to compare the efficacy and safety of first-line regimens for patients with advanced esophageal squamous cell carcinoma (ESCC).MethodsA systematic review and Bayesian network meta-analysis were conducted by retrieving relevant literature from PubMed, Embase, the Cochrane Library, and the Web of Science. We included published sources of randomized clinical trials comparing immunotherapy combinations for treating advanced ESCC.ResultsWe analyzed seven studies involving eight immunotherapy combinations and 4688 patients. For patients without programmed death-ligand 1 (PD-L1) selection, it was found that the combination of toripalimab and chemotherapy provided better overall survival than chemotherapy alone (hazard ratio = 0.58, 95% confidence interval (CI) 0.43-0.78). Compared with chemotherapy alone, Sintilimab or camrelizumab plus chemotherapy seemed to achieve the best progression-free survival (hazard ratio = 0.56, 95% CI 0.46-0.68). Nivolumab plus chemotherapy appeared to provide the best objective response rate, with significant differences versus chemotherapy alone (odds ratio = 0.49, 95% CI 0.38-0.64). Nivolumab plus ipilimumab resulted in a relatively lower incidence of adverse events of grade ≥3 than other regimens.ConclusionsThe combination of immune checkpoint inhibitors (ICIs) with chemotherapy provided a high probability of more effective treatment in comparison with chemotherapy alone for patients with advanced ESCC. Toripalimab and sintilimab plus chemotherapy were ranked as providing the highest OS and PFS benefit in the first-line setting, respectively.