基于MATLAB的CGMS数据整合分析软件的开发和在糖尿病合并低血糖时血糖波动特点研究上的使用  

CGMS Data-integration Software Based on MATLAB Analysis and Its Use in Diabetic Patients with Hypoglycemia

作　　者：沙雯君[1,2] 谭维锴[3] 沈远东[1,2] 朱婷[4] 张伟妃[1,2] 黄文瑾[1,2] 江海丽 牛晶晶[1,2] 李俊燕[1,2] 周里钢[1,2] 

机构地区：[1]上海中医药大学附属曙光医院内分泌科,上海201203 [2]上海中医药大学糖尿病研究中心,上海201203 [3]上海同济大学电子与信息工程学院,上海201804 [4]上海中医药大学附属曙光医院老干部科,上海201203

出　　处：《中国医学物理学杂志》2014年第3期4936-4940,共5页Chinese Journal of Medical Physics

基　　金：国家自然科学基金面上项目(81370932);美国MERCK IISP基金(#40313 and #40309)

摘　　要：目的:观察低血糖血糖波动趋势,探究低血糖发生的特点。方法:通过矩阵实验室(Matrix Laboratory,MATLAB)编程设计了与MiniMed解决方案(MiniMed Solutions CGMS Sensor,MSCS)的接口,利用MATLAB分析能力,实现大样本量的CGMS数据整合分析。结果:对829例CGMS数据进行筛选,符合研究条件405例,其中出现低血糖患者204例占50.62%。CGMS血糖监测系统与指测血糖在不同血糖范围均有较高一致性(P<0.01)。低血糖患者整体血糖低于非低血糖患者,低血糖发生时间尤以半夜及午餐前居多。低血糖聚集发生时间平均在1点09分,标准差时间为5小时31分。不同年龄段低血糖患者比较,低血糖发生时间随年龄增长而提前(P<0.01)。结论:MATLAB在大量CGMS监测数据处理具有速度快、设计灵活、使用方便等优点。低血糖的发生具有规律性,我们今后的研究方向是和低血糖发生可能相关的方向调节激素,神经内分泌的调节和脑生物钟的变化。Objectlve： To observe the fluctuating blood glucose level in diabetes with hypoglycemia patients, we investigated the characteristic of hypoglycemia including the time of occurrence and the glucose variability. Methods： Using the super analyze power of matrix laboratory （MATLAB） software, we designed a new software compatible with the commercial CGMS （Continuing Glucose Monitoring System） software （MiniMed Solutions Sensor Link）, and achieved mass CGMS in formation of patient into integration and analysis. Results： With our software, 829 cases of CGMS data was filtered, 405 cases were brought into statistics, among which 204 of patients （50.62%） were found to be suffered from hypoglycemia. The correlation and coincidence between CGMS glucose values and BGMS （ Finger Meter Glucose Monitoring System） glucose values were favorable in the whole blood glucose range （P〈0.01）. The all-days glucose level diabetic patient with hypoglycemia is lower than that in ones without hypoglycemia, the hypoglycemia episode apt to occur before lunch and during the mid-night. We found that the most common time for happening of hypoglycemia time is 1：09AM e5h31min, which is consistent to the previous studies. The study in the different aging groups indicated that hypoglycemia occurred earlier in the aging patients than that of younger one （P〈0.01）. Conclusions： Our recent soitware designed for MATLAB has the advantages of speediness, design flexible, convenient for processing a huge number of CGMS data duel. Hypoglycemia happens in law and our future study will focus on the mechanism of time of occurrence hypoglycemia on counterregulation hormone response, neuroendocrine and brain bioclock.

关 键 词：MATLAB CGMS数据整合 低血糖 

分 类 号：R443.8[医药卫生—诊断学]