ObjectiveTo combine specific examples and R Studio language code, to apply the Bayesian quantile regression method in the analysis of clinical medicine data, and show the advantages of Bayesian quantile regression method, so as to provide references for improving the accuracy of medical research. Methods The clinical data of 250 patients with knee osteoarthritis from the capital special research on the application of clinical characteristics project were used. A Bayesian quantile regression model based on data set was constructed to explore the relationship between the level of serum IgG and the age of the patients. Results The Monte Carlo algorithm converge can judge the efficiency of parameter estimation based on Gibbs sampling which was used to draw samples from the posterior distribution of parameters in Bayesian quantile regression. By generating the parameter into the regression formula, we can obtain the regression under different quantiles: Y1=−6.022 063 47+2.026 913 73X−0.015 077 69X2……Y5=24.610 542 414−0.395 059 497X+0.004 205 064X2. It can be found that the serum level of IgG was obviously increased with age. Conclusion Bayesian quantile regression parameter estimation results are accurate and highly credible, and reliable parameter information can be obtained even under small sample conditions. It has great advantages in the research of clinical medicine data and has certain promotional value.
BUGSnet is a powerful R project package for Bayesian network meta-analysis. The package is based on JAGS and enables high-quality Bayesian network meta-analysis according to recognized reporting guidelines (PRISMA, ISPOR-AMPC-NCA and NICE-DSU). In this paper, we introduced the procedure of the BUGSnet package for Bayesian network meta-analysis through an example of network meta-analysis of steroid adjuvant treatment of pemphigus with continuous or dichotomous data.
The recurrent neural network architecture improves the processing ability of time-series data. However, issues such as exploding gradients and poor feature extraction limit its application in the automatic diagnosis of mild cognitive impairment (MCI). This paper proposed a research approach for building an MCI diagnostic model using a Bayesian-optimized bidirectional long short-term memory network (BO-BiLSTM) to address this problem. The diagnostic model was based on a Bayesian algorithm and combined prior distribution and posterior probability results to optimize the BO-BiLSTM network hyperparameters. It also used multiple feature quantities that fully reflected the cognitive state of the MCI brain, such as power spectral density, fuzzy entropy, and multifractal spectrum, as the input of the diagnostic model to achieve automatic MCI diagnosis. The results showed that the feature-fused Bayesian-optimized BiLSTM network model achieved an MCI diagnostic accuracy of 98.64% and effectively completed the diagnostic assessment of MCI. In conclusion, based on this optimization, the long short-term neural network model has achieved automatic diagnostic assessment of MCI, providing a new diagnostic model for intelligent diagnosis of MCI.
ObjectiveA simulation study was used to generate the multivariate normal distribution data with a residual effect based on series of N-of-1 trials. The statistical performance of paired t-test, mixed effect model and Bayesian mixed effect model were compared.MethodsThree-cycles N-of-1 trials were set, and the participants were randomly assigned to 2 different treatments in each cycle. The simulation study included the following procedures: producing six-dimensional normal distribution data, randomly allocating intervention methods and patients, adding residual effects, constructing and evaluating 3 models, and setting the parameters. The sample sizes were set as 3, 5, 8 and 10, and the correlation coefficients among different times were set as 0.0, 0.5 and 0.8. Different proportions of residual effects for the 2 groups were set. Type I error, power, mean error (ME), and mean square error (MSE) were used to compare the 3 models.ResultsWhen there was no residual effect in the 2 groups, type I errors of 3 models were approximately 0.05, and their MEs were approximately 0. Paired t-test had the highest power and the lowest MSE. When the residual effect existed in the 2 groups, the type I error of paired t-test increased, and its estimated value deviated from the true value (ME≠0). Type I errors of the mixed effect model and Bayesian mixed-effect model were approximately 0.05, and they had the same power. The estimated values of the two models were close to the true value (ME was approximately 0).ConclusionsWhen there is no residual effect (0% vs. 0%), paired t-test is suitable for data analysis of N-of-1 trials. When there is a residual effect, the mixed effect model and Bayesian mixed-effect model are suitable for data analysis of N-of-1 trials.
This study introduced the construction of individualized risk assessment model based on Bayesian networks, comparing with traditional regression-based logistic models using practical examples. It evaluates the model's performance and demonstrates its implementation in the R software, serving as a valuable reference for researchers seeking to understand and utilize Bayesian network models.
The choice of genetic models was main difficulty in the meta-analysis of gene-disease association studies. In this study, we made a further discussion about the genetic model-free approach that proposed by Minelli et al. The program that coded by JAGS and R was carried out to perform the Bayesian procedure. In a real example, several kinds of prior distribution were used, including non-informative prior distribution and external clinical prior information. Especially, compared to Minelli’s study, we introduced clinical prior information. The results indicated that the pooled results were rather robust no matters the prior distribution were non-informative or informative, especially when the number of included studies were large.
Statistical analysis of clinical trials has traditionally relied on frequentist methods, but Bayesian statistics has attracted considerable attention from regulators and researchers in recent years due to its unique advantages, and its use in clinical trials is increasing. Despite the obvious advantages of Bayesian statistics, the complexity of its design, implementation and analysis poses a number of challenges to its practical application, which may lead to an increased risk of unregulated use. This study aims to comprehensively sort out the application scenarios, common methods, special considerations and key elements of reporting of Bayesian statistical methods in clinical trials, with the aim of providing researchers with references for conducting Bayesian clinical trials, and promoting the scientific and rational application of Bayesian statistical methods in clinical trials.
ObjectiveTo introduce a Bayesian meta-analysis method for quantitatively integrating evidence from both randomized controlled trials (RCTs) and non-randomized studies of interventions (NRSIs), using concrete examples and R code, thereby supporting the combined utilization of both study types in empirical research. MethodsUsing a meta-analysis on the association between low-dose methotrexate exposure and melanoma as an example, we employed the jarbes package in R to conduct both a traditional Bayesian meta-analysis and a Bayesian nonparametric bias-correction meta-analysis model for quantitative integration. The differences between the two pooled results were then compared. ResultsThe traditional Bayesian meta-analysis indicated a posterior probability of 99% that low-dose methotrexate exposure increases melanoma risk. The Bayesian nonparametric bias-correction meta-analysis model showed a posterior probability of 92% that low-dose methotrexate exposure increases melanoma risk. ConclusionCompared to the traditional Bayesian meta-analysis model, the nonparametric bias-correction meta-analysis model is more suitable for quantitatively integrating evidence from RCTs and NRSIs, demonstrating potential for broader application. However, the comparability between the two evidence bodies should be carefully assessed prior to quantitative integration.
ObjectiveTo introduce Bayesian meta-analysis of dichotomous data using PROC MCMC in SAS software.MethodsA previous published systematic review was used as an example, Bayesian meta-analysis of dichotomous data was implemented by PROC MCMC in SAS software, and programming code was provided.ResultsThe log-transformed value of odds ratio (OR) was used as the efficacy. The results of the Bayesian meta-analysis were very similar to those obtained by the frequency method.ConclusionsBased on the powerful programming capabilities of SAS, PROC MCMC can easily implement Bayesian meta-analysis of dichotomous data. With the rapid development of Bayesian statistical theory, Bayesian meta-analysis will play an important role in the field of meta-analysis.
The "bnma" package is a Bayesian network meta-analysis software package developed based on the R programming language. The network meta-analysis was performed utilizing JAGS software, which yielded relevant results and visual graphs. Moreover, this software package provides support for various data structures and types, while also providing the advantages of flexible utilization, user-friendly operation, and deliver of rich and accurate outcomes. In this paper, using a network meta-analysis example of different therapies for androgenetic alopecia, the operational process of conducting network meta-analysis using the "bnma" package is briefly introduced.