Degree Type

Thesis

Date of Award

2011

Degree Name

Master of Science

Department

Chemical and Biological Engineering

First Advisor

Derrick K. Rollins

Abstract

Type 1 diabetics often experience extreme variations in glucose concentrations which can have adverse long– and short–term effects such as severe hypoglycemia, hyperglycemia and organ degeneration. Studies have established that there is a need to maintain the glucose levels within a normal range (e.g. 80—150 mg/dL) to avoid complications caused by diabetes. However, initial attempts to regulate blood glucose levels using insulin infusion pumps, multiple injections or a combination of the two have had limited success as they lack the ability to decide the appropriate rate and/or dose of insulin based on the current metabolic state of the body. Consequently, what is needed is an automatic insulin delivery system (i.e., artificial pancreas) with the ability to determine continuously the rate of insulin delivery required to provide optimum closed-loop glucose control (i.e., to minimize the variability around a desired glucose level) and to eliminate the individual from the insulin dosage decision making in this control loop. Due to recent advances in biomedical technology, such as automatic insulin delivery systems using glucose sensors and insulin pumps, blood glucose modeling and control has received considerable attention in the process control community and models of various degrees of complexity have been developed. Glucose levels are affected by many variables, such as stress, physical activity, hormonal changes, periods of growth, medications, illness/infection, fatigue, as well as food intake and insulin tolerance. Furthermore, not only does glucose change from several sources of disturbances but their impact on blood glucose level is highly correlated, dynamic and nonlinear making it difficult to distinguish the effect each input has on blood glucose. Thus, the objective of this research is to introduce a modeling methodology that is able to take into account the simultaneous and multiple effects of food, activity, stress and their interactions.

The research presented in this thesis is carried out on 15 Type 1 diabetic subjects where thirteen variables (i.e., three food variables, seven activity variables, basal insulin, bolus insulin, and time of day (TOD)) are collected for two weeks and modeled using the Wiener block–oriented model. Three types of models are compared: input–only (Model 1), input–output (Model 2), and output–only (Model 3). Results are given for k –steps ahead prediction (k –SAP) from 5 minutes to 3 hours in the future and show the importance of taking into account the interactions between input variables.

DOI

https://doi.org/10.31274/etd-180810-134

Copyright Owner

Kaylee Renee Kotz

Language

en

Date Available

2012-04-30

File Format

application/pdf

File Size

91 pages

Share

COinS