随着中医院各项服务的信息化建设的深入，日常医疗工作中积累下来的业务数据，是临床回顾性研究的重要数据资源。然而业务数据分布在不同的系统，难以进行高效的查询以及分析。利用Python+Django+MySQL技术构建数据仓库，从各个业务系统提取数据，整合到统一的平台，以便实现中医、西医临床业务数据的多维度，高效率的筛选以及提取，为临床回顾性研究提供高效准确的数据来源。极大地提高我们探寻中医药宝库的效率。我们对数据仓库的结构，构造的过程以及实际的应用进行介绍。

As the hospital information service construction development, daily medical work has accumulated huge value of data, which is an important data resource for the clinical retrospective study. However, clinical data is distributed in different sub system. It is difficult to carry out efficient query and analysis. In this study, Python + Django + MySQL technology was used to build a data warehouse, in order to extract data from each sub system and to integrate them into a unified platform. Therefore, the multi-dimensional efficient screening and extraction of clinical data from both traditional Chinese medicine (TCM) and modern medicine are achieved. It provided efficient and accurate data sources for clinical retrospective research. It has also greatly improved the efficiency of our search in TCM. This paper introduced the structure of the database, the process of construction and the actual application of this data warehouse.

广东省科技厅重大项目(2014B010118005)：大数据开放型的个性化健康医疗公共服务平台，负责人：陈达灿；2014年国家中医药行业科研专项(No.201407005)：中医综合调养方案对慢性肾脏病进展高位因素干预的前瞻性登记研究，负责人：刘旭生；“十二五”国家科技支撑计划课题(No. 2013BAI02B04): 中医药治疗慢性肾脏病5期（非透析）的效果比较研究，负责人：刘旭生；广东省中医院院内专项课题(2014KT1305)：补脾益肾法防治慢性肾脏病4-5期的理论及机制研究，负责人：刘旭生。