Objective To compare the ability of three propensity score weighting methods to balance the covariates and the advantages and disadvantages to estimate the treatment effects when dealing with multiple treatment data under different sample sizes. Methods Monte Carlo simulation was used to generate data sets and the advantages and disadvantages of balancing covariates and estimating the treatment effects of three propensity score weighting methods, Logistic-IPTW, Logistic-OW and GBM-OW were compared. The evaluation index of covariate equilibrium level was the absolute standard mean difference. The evaluation indexes of effect estimation included the point estimate of treatment effect, root mean square error and confidence interval coverage. Results Compared with Logistic-IPTW and Logistic-OW, GBM-OW was better in effect estimation and had a smaller root mean square error in five scenarios where covariates were related to treatment factors and outcome variables with different varying degrees of complexity. In terms of covariate equilibrium, all three methods had good effects. GBM-OW method performed better when the overlap of propensity score distribution of multiple treatment data was relatively low and covariables had increasingly complex nonlinear relationships with treatment factors and outcome variables. Conclusion When dealing with multiple treatment data, GBM-OW method has advantages over the other two methods when there is nonlinearity and/or interaction between covariates and treatment factors and outcome variables. Using this method, the effect estimation is closer to the real value, which is a better choice.
Randomized controlled trials (RCTs) are often limited because of ethical or operational reasons. Quasi-experimental studies could be an alternative to RCTs to make causal inferences without randomization by controlling the confounding effects of the study. This paper introduced the general statistical analysis methods of quasi-experimental design through basic ideas, characteristics, limitations and applications in medicine, including difference-in-difference models, instrumental variables, regression discontinuity design, interrupted time series, and so on, and to provide references for future research.
This study introduced the inverse probability weight and overlap weight by propensity score and how to test the balance and estimate the effect after weighting. Four R packages that can be used for propensity score weight analysis were introduced and compared.
Objective To elaborate on the statistical analysis methods for evaluating the accuracy of imaging diagnostic tests in a multiple-reader multiple-case (MRMC) design through formula derivation and real cases. Methods This study consisted of two parts: theoretical derivation and a real case study. The theoretical part discussed in detail the principles and procedures of MRMC statistical analysis methods, particularly the Obuchowski-Rockette (OR) and Dorfman-Berbaum-Metz (DBM) methods. The real case included 100 subjects, of whom 67 had disease. Four readers interpreted all the cases based on both traditional film imaging methods and digital imaging methods. OR and DBM methods were employed for data analysis. Results The real case showed that the OR and DBM methods had a high degree of consistency, with only slight differences in the confidence intervals. Conclusion It is recommended to use the OR and DBM methods for the statistical analysis of imaging diagnostic test accuracy, ensuring that the impact of reader factors on the evaluation results is fully considered. The results from the OR and DBM methods are relatively similar; when applying these methods in practice, one should consider the specific characteristics of the data and the research design to choose the appropriate analysis method. Besides, there are still challenges when applying the OR and DBM methods, such as software implementation and missing data handling, which require further exploration.
ObjectiveTo explore two methods of sample size estimation in multi-reader multi-case study of radiological diagnostic test and realize them by software. MethodsDemonstration programs were conducted in R software using the Van Dyke dataset, calculating combinations of readers and cases using the OR and DBM methods. These serve as pilot test results for multi-reader multi-case studies, providing a reference for parameter settings in subsequent formal experiments. ResultsWhen the effect size was 0.044, 6 readers and 247 cases could yield 0.80 power, while with an effect size of 0.088, only 6 readers and 44 cases were needed to reach 80.5% power. The sample sizes calculated using the OR method and the DBM method were consistent, and the same sample size calculation results could be obtained through conversion between the two methods. ConclusionFor the estimation of sample size in multi-reader multi-case studies, R software provides a convenient and mature software package for sample size estimation using multi-reader multi-case designs in radiological diagnostic tests, thereby offering a reference for selecting appropriate sample size estimation and statistical analysis methods in radiological diagnostic tests.
With the continuous improvement of the accuracy of predictive models, the problems of algorithmic bias and the lack of fairness have become increasingly prominent. This not only may lead to significant differences in the accessibility of medical services and the effectiveness of diagnosis and treatment among different groups, but also may trigger medical risks such as misdiagnosis and missed diagnosis, treatment delays, as well as ethical disputes and legal issues under algorithmic discrimination, becoming the key reasons restricting the implementation of technology. At present, research on fairness has shifted from ethical and conceptual discussions to quantitative calculations, exploration of bias layer mechanisms, and exploration of technical mitigation strategies. This article, through a comprehensive analysis and discussion of the fairness issue of clinical prediction models, systematically sorts out the sources of model unfairness, fairness assessment and mitigation methods, as well as commonly used toolkits, and at the same time explores the future development trends. This study aims to sort out the current status of fairness studies on predictive models and promote the responsible application of artificial intelligence in the medical field.
ObjectiveTo introduce and validate the censoring unbiased deep learning (CUDL) algorithm suitable for survival data and to explore its utility in survival analysis. MethodsThe principles of the CUDL algorithm were detailed, followed by an introduction to its software implementation steps and parameter settings. Using the construction of a long-term survival prognosis model for patients with myocardial infarction (MI) as a case study, the algorithms were compared with the conventional Cox proportional hazards model and random survival forest model. ResultsIn the task of predicting the 5-year survival probability of patients, the mean Brier score (BS) of both the Buckley-James censoring unbiased deep learning (BJDL) model and the regularized BJDL were superior to that of the Cox proportional hazards model (mean BS: 0.1722 and 0.1715 vs. 0.1765, P<0.001). The regularized BJDL also outperformed the random survival forest model (RSF500) (mean BS: 0.1715 vs. 0.1750, P<0.001). ConclusionThe BJDL can effectively leverage censored information within survival data to achieve accurate predictions without the need to satisfy the proportional hazards assumption. It provides a novel methodological alternative for the analysis of medical survival data.