样本量估计是医学研究设计的一个重要环节，它直接关系到研究项目的效率乃至成败。然而，目前应用较广的某些统计方法尚缺少相应的样本量估计方法，如诊断试验评价中的Youden指数、卫生经济学评价中较复杂设计的增量成本效果比（ICER）、群随机试验中的多组比较等。此外，现有国际权威样本量估计软件尚存在功能不完善和操作不简便等缺陷。为此，本项目将针对样本量估计领域中所涉及的尚未建立和尚待完善的方法展开研究，重点解决：建立基于灵敏度和特异度任意赋权并考虑两者关联的Youden指数样本量估计方法；基于Bayesian思想提出配对设计和交叉设计的ICER样本量估计方法；建立群随机设计中多组均值/率比较的样本量估计方法；研发适用范围更广、操作更简便、拥有自主知识产权的样本量估计软件。本项目的预期成果不仅可丰富样本量估计的理论和方法，还能为样本量估计提供功能完善、操作简便的工具，满足医学研究领域迫切的应用需求。

Sample size estimation plays a key role in medical research, and it can affect the efficiency or even determine the success of the studies. However, some most widely used statistical methods still lack the corresponding sample size estimation method at present, such as Youden index for diagnostic test evaluation, incremental cost effectiveness ratio (ICER) for health economics evaluation in complex study design, multiple group comparison in cluster randomized trials, and so on. In addition, the function and operation of the existing international authoritative sample size estimation software can still be improved. Therefore, this project will focus on establishing the sample size estimation method for Youden index which is based on the association between sensitivity and specificity and the different weighting of sensitivity and specificity, proposing methods to estimate the sample size for the ICER using Bayesian method for paired design and cross-over design; establishing the sample size estimation methods for multiple groups comparison in means or proportions in cluster randomized trials, and developing a sample size estimation software with wider application, more convenient operation, and independent intellectual property rights. The expected results of this project can not only enrich the theory and methodology in the field of estimation of the sample size, but also provide the perfect function and simple operation tool for the application, which will meet the urgent need of medical research.