UC San Diego
Human blood glucose dynamics
- Author(s): Rahaghi, Farbod N.
- et al.
The control of blood glucose concentration has become central to the prevention of morbidity in diabetes. Currently sensors are becoming available to make available near continuous measurements of tissue glucose concentrations. Frequently measured values provide an opportunity to analyze the dynamics of these measurements in addition to statistical analysis. The dynamics can be used to verify sensor validity, to provide a physiologic control target, and serve as a tool to diagnose and monitor disease progression as well as therapeutic interventions. In this document, analysis methods from a diverse set of physical and engineering sciences are applied to blood glucose data that has been published in the literature, and measured in clinical studies, in humans. The objective is to evaluate the utility of different techniques for time-series analysis, as well as to pave the way towards more data intensive studies to further the applications mentioned above. Data from nondiabetics, type I diabetics, type II diabetics as well as other disease is presented and analyzed using methods which appeared most promising during the author's research. Based on the limited available data, observations are made regarding the characteristics of dynamics in each population groups, and the potential utility of dynamic measurements in diagnosis and assessment of patient metabolic state are demonstrated. In particular, the utility and challenges associated with various methods of time-series analysis as applied to the human blood glucose signal are explored. It is hoped that this can provide a beginning to a very promising future of human glucose time-series analysis and that it will help in sensor and controller design