医疗大数据背景下的疼痛发展机遇.ppt

科研数据库系统的应用 Step5：数据录入 资料收集整理，录入科研数据库系统 在科研系统中填写CRF表 科研数据库系统的建设情况 Step6：研究分析 提取信息进行研究分析 查询病例导出数据 查看数据自动抓取情况 科研数据库系统的建设情况 Step7：数据提取结果 PART7: 设置数据导出条件 数据导出结果 科研数据库系统的应用 Step8：数据导出电子表格 导出电子表格 THANK YOU! * * $770 DuoFertility vs. $13,775 in vitro fertilization 20,000 measures A day, 1/1000 degree * 大数据 是 海量 数据 + 分析方法 易变性（Variability），准确性 (Veracity) 和复杂性 (Complexity) 四个特性也是大数据的也是为什么传统数据库工具无法进行处理原因。 不管怎样，任何形式、内容的数据都具有价值。 * * * * * * * * * * * * 美国医疗大数据研究的数据 现阶段国内研究方向 主要是统计学方法 * For K-M plots: This figure provides Kaplan-Meier plots of the overall cancer survival for Vanderbilt University and Mayo clinic. For Vanderbilt University, the overall cancer survival decreases from the groups of diabetic patients taking metformin, non-diabetic patients, diabetic patients taking other drugs, to diabetic patients taking insulin. For Mayo clinic, the overall cancer survival decreases from the groups of diabetic patients taking metformin, non-diabetic patients, diabetic patients taking insulin, to diabetic patients taking other drugs. For HR diagram: We used the stratified Cox proportional hazards model to investigate the adjusted treatment effects by stratifying on tumor stage and adjusting for age, gender, race, smoking status, insulin usage, BMI, and tumor type. The hazard ratio estimates and their associated 95% confidence intervals are presented in this figure. The vertical gray line represents the null effect. If the 95% confidence interval crosses this vertical line, the p-value of the corresponding comparison is non-significant, otherwise, it is significant. For Vanderbilt patients, the metformin group has significant lower hazards of cancer related death comparing to other drugs group and insulin group when we study the overall cancer survival (I cannot remember exactly the other conclusions we have, but we should have another one or two significant result from subgroup analysis. Similar statement for Mayo clinic). * * * * * 国家卫生信息化“十二五”规划“35212工程”。规划