基于大数据技术的“老年衰弱征”辅助诊断功能开发  

Explore and Assessment of Frailty for Chinese Elderly Based on Big Data Technology

作　　者：刘鹏[1] 龚瑞[2] LIU Peng;GONG Rui(Information Center,The People' Hospital of Xinjiang Uygur Autonomous Region,Urumqi 830001,Xinjiang Uygur Autonomous Region,P.R.C.)

机构地区：[1]新疆维吾尔自治区人民医院信息中心 [2]新疆维吾尔自治区人民医院老年医学中心

出　　处：《中国数字医学》2018年第12期108-110,共3页China Digital Medicine

摘　　要：目的:探索利用大数据技术提取和整合医疗信息资源,实现对"老年衰弱征"诊断提供辅助的可行性。方法:选择118例老年医学中心入院患者为研究对象。利用大数据技术(Hadoop)建立评价指标体系模拟计算FI值。用统计学方法确定最终指标集合并利用该指标集合编写程序,进行入院患者的衰弱筛查,将评估结果推送至医师。结果:两种评估结果呈正相关(r=0.607,P=0.000);ROC曲线下面积(AUC)为0.912(95%CI:0.861～0.963,P=0.000),其灵敏度、特异度及Youden指数分别为99%,75%,74%。一致性Kappa值为0.169(P=0.000)。结论:应用大数据技术自动对入院老年人进行衰弱筛查,辅助医务工作者进行诊断是可行的。Objective:To explore the feasibility by big data technology to extract and integrate medical information resources to provide assistance for the diagnosis of "senile weakness".Methods:In the geriatric center,118 patients were selected as the study subjects.The use of big data technology (Hadoop)to establish an evaluation index system simulates and calculates the FI value. Statistical methods are used to determine the final set of indicators and use this set of indicators to write a program to perform debilitating screening of patients admitted to hospital and transmit the results of the assessment to physicians.Results:Two evaluation results were positively correlated (r=0.607,P=0.000),The area under ROC curve (AUC)is 0.912(95% CI:0.86-0.963,?=0.000), The FI threshold was 0.28,and its sensitivity,specificity and Youden index were 99%,75% and 0.74% respectively.The consistent Kappa value of the large data technology coarse sieve (FI=0.28)and the phenotype definition stage is 0.169(P=0.000).Conclusion: The application of big data technology to automatic screening of elderly patients admitted to hospital,and assisting medical workers in the diagnosis is feasible.

关 键 词：大数据 衰弱 评估 表型 

分 类 号：R319[医药卫生—基础医学]