面向云计算的大规模电子病历知识协同约简是当前医疗大数据背景下临床决策支持系统研究中一个具有挑战性的课题。本项目以大数据知识协同约简和海量电子病历智能处理为研究对象，探索基于粗糙集知识约简理论和云计算技术的大规模电子病历知识协同约简统一框架、模型和算法，剖析云计算环境下参与大规模电子病历知识协同约简的协同进化种群局部解与全局优势解一致性机理，以及开展知识协同约简实际效能评估和自适应优化研究。本项目预期提出一套完整的面向云计算大规模电子病历知识协同约简理论和方法，形成用于电子病历临床决策支持的云端服务系统，弥补以往解决大规模电子病历知识约简方法的不足，为实现基于大规模电子病历的临床智能决策支持系统提供较好理论模型和求解算法支持。本项目研究成果对探索云计算环境下电子病历知识约简及其相关疾病智能辅助诊断具有重要意义与价值，同时也为大数据知识协同约简问题求解提供重要的科学依据。

In the context of big medical data, the knowledge collaborative reduction of large-scale electronic medical records for cloud computing is a challenging topic in clinical decision support systems. The knowledge collaborative reduction of big data and the intelligent processing of massive electronic medical records for cloud computing will be focused to research in this project. We will explore the unified knowledge collaborative reduction framework, model and algorithm of large-scale electronic medical records based on the knowledge reduction theory of Rough Set and cloud computing technology, and analyze the consistency mechanism between the local solutions and global dominance solution achieved by the co-evolutionary populations which participate in the knowledge collaborative reduction process of large-scale electronic medical records. Furthermore, we will carry out the effectiveness evaluation of knowledge collaborative reduction and its adaptive optimization in such environment. Based on above research, we will propose a new complete set of knowledge collaborative reduction theories and approaches, and meanwhile we will construct the cloud service system of clinical decision support analysis for electronic medical records, which will make up for the shortcoming of traditional methods. This project is potential to provide the better theoretical models and solution algorithms for the realization of clinical intelligent decision support system based on large-scale electronic medical records. The significance of this project is beneficial for the knowledge reduction of electronic medical records and the intelligent assistant diagnosis of related diseases in the cloud computing environment, and offers an essential scientific basis to solve the knowledge collaborative reduction problem of big data.