最小模型参数的估计对胰岛素敏感性指数的影响  

The Effect of the Estimated Parameters of Bergman's Minimal Model on Insulin Sensitivity Index

作　　者：李萌[1] 李林[1] 闫岩[1] 

机构地区：[1]首都医科大学生物医学工程学院,北京100069

出　　处：《北京生物医学工程》2010年第1期23-29,共7页Beijing Biomedical Engineering

基　　金：北京市教育委员会科技发展计划项目(KM200810025020)资助

摘　　要：Bergman最小模型是研究葡萄糖和胰岛素代谢最具代表性的模型。通过随机变动Bergman模型中部分参数值,数值计算得到各时间点胰岛素和葡萄糖的浓度值。利用这些数据拟合模型参数并估计出SG值和ISI值,使用统计学方法分析Bergman最小模型法估计值的SG值和ISI值随模型参数变化的情况。分析结果表明,P1在临床指标内取值,最小模型计算得到的估计值数据分布为正态;胰岛素动力学方程的参数与SG和ISI的估计值存在着强线性相关关系;而P1和P3的取值对SG和ISI的估计值有着直接的影响;SG和ISI的估计值存在着低估现象,但误差的范围小于2.5%。Bergman＇ s minimal model is the most representative model to examine the glucose and insulin metabolism. In this paper, the model is used to calculate the concentration values of insulin and glucose at several time points through changing some parameters randomly. Then the values are used to simulate the model parameters and to estimate the values of SG and ISI. After that the values of the SG and the ISI estimated by Bergman＇s minimal model are analyzed statistically with changes of model parameters. The results show that, when the value of P1 is selected in the clinical indicators, the estimated values distribute in normality. There is a strong linear correlation between the parameters in insulin kinetic equation and the estimated values of SG and ISI. The values of P1 and P3 impact the estimation of SG and ISI directly. The estimated values of SG and ISI are underestimated, but the scope of error is less than 2.5%.

关 键 词：Bergman最小模型 胰岛素敏感性指数 葡萄糖利用效能 统计分析 

分 类 号：R318.04[医药卫生—生物医学工程]