The Library
Mathematical modelling of blood glucose dynamics in normal and impaired glucose tolerance
Tools
Eichenlaub, Manuel Max Werner (2020) Mathematical modelling of blood glucose dynamics in normal and impaired glucose tolerance. PhD thesis, University of Warwick.
|
PDF
WRAP_Theses_Eichenlaub_2020.pdf - Submitted Version - Requires a PDF viewer. Download (21Mb) | Preview |
Official URL: http://webcat.warwick.ac.uk/record=b3518289~S15
Abstract
Type 2 diabetes mellitus and its preliminary stages are characterised by chronically elevated blood glucose levels, particularly after food intake. Assessing the postprandial glucose metabolism is, therefore, crucial to facilitate appropriate treatment strategies such as dietary interventions. This thesis develops mathematical models for the description of glucose profiles in response to food intake using glucose data alone. These glucose-only models thereby overcome the necessity of measuring insulin which is laborious and unreliable, thus enabling their widespread use in clinical practice. The main purpose of the developed models is the extraction of information on insulin sensitivity and meal-related glucose appearance, both of which have a significant influence on the postprandial glucose response. The extracted information is validated against the results from the established oral minimal model requiring both glucose and insulin data for identification. For both oral minimal and glucose-only models, this work proposes a novel input function for the description of the meal-related glucose appearance. This new function is fully differentiable and more suitable for modelling consecutive meal responses on the same day in comparison to the conventional but highly impractical piecewise-linear function. The models are identified from both a literature dataset and a dataset collected during an experimental study designed and conducted in the context of this work. The latter includes subjects with normal glucose tolerance, prediabetes and type 2 diabetes mellitus and features the use of continuous glucose monitoring. The model identification procedure is carried out using a variational Bayesian technique, which offers an efficient method for the probabilistic treatment of the parameter estimation task. The results demonstrate that the developed glucose-only models can be used to infer information on insulin sensitivity as they contain a parameter highly correlated to the insulin sensitivity inferred from the established oral minimal model. Furthermore, it is shown that the glucose appearance profiles inferred from the glucose only models allow the same interpretation of trends in glucose appearance with respect meal composition as the oral minimal model. Using the information on insulin sensitivity and glucose appearance, the developed models could thus support healthcare professionals in designing effective treatment strategies such as dietary interventions and monitor the disease progression from prediabetes to type 2 diabetes.
Item Type: | Thesis (PhD) | ||||
---|---|---|---|---|---|
Subjects: | Q Science > QD Chemistry R Medicine > RC Internal medicine T Technology > TP Chemical technology |
||||
Library of Congress Subject Headings (LCSH): | Glucose -- Mathematical models, Blood sugar -- Analysis, Blood sugar monitoring, Diabetes | ||||
Official Date: | October 2020 | ||||
Dates: |
|
||||
Institution: | University of Warwick | ||||
Theses Department: | School of Engineering | ||||
Thesis Type: | PhD | ||||
Publication Status: | Unpublished | ||||
Supervisor(s)/Advisor: | Khovanova, Natasha ; Hattersley, John | ||||
Sponsors: | University Hospitals Coventry and Warwickshire NHS Trust | ||||
Extent: | xvii, 218 leaves : illustrations, charts, photographs | ||||
Language: | eng |
Request changes or add full text files to a record
Repository staff actions (login required)
View Item |
Downloads
Downloads per month over past year