Peak Blood Glucose Prediction Algorithm Following a Meal Intake

TitlePeak Blood Glucose Prediction Algorithm Following a Meal Intake
Publication TypeConference Paper
Year of Publication2007
AuthorsIslam, M. S., J. Leech, C. C. Y. Lin, and L. Chrostowski
Conference NameElectrical and Computer Engineering, 2007. CCECE 2007. Canadian Conference on
Pagination579 -582
Date Publishedapr.
Keywordsblood glucose excursions, diabetes patients, diseases, electrical engineering computing, glucose monitors, glucose pattern, glucose reading, health care, health problems, hyperglycemia, insulin dose, insulin infusions, patient care, peak blood glucose prediction algorithm
Abstract

Motivated by the increasing number of diabetes patients around the globe, we have created an algorithm that predicts blood glucose trends and levels for the period of time after a meal intake. This information will help both type-I and type-II diabetic patients properly predict their bodies' effect to food intake and avoid dangerous hypoglycemic and hyperglycemic states. Most of the long-term health problems (e.g. nephropathy, retinopathy, and neuropathy) result from sustained hyperglycemia and frequent blood glucose excursions in time due to inaccurate insulin infusions; this is a direct result of a misjudgement of dosage or from poor patient compliance. Some devices already developed, such as one time glucose meters and continuous glucose monitors, lack information diabetics need to fully understand glucose level trends, such as an increasing or decreasing glucose pattern and the rate of change. Patients must therefore estimate their insulin dose based on their present glucose reading and estimated food intake, which cumbersome and inaccurate. This is where our algorithm will help diabetic patients. Our algorithm differs from other developed devices and models (Lehman and Deutsch, 1992) in that it predicts insulin dosage based on the predicted peak blood glucose level just after the meal-intake. Since we are dealing with human body, it is quite difficult to model how each organ will behave, thus a number of parameters need to be collected to calibrate the model for each individual. In the future, this predictive system could be integrated with an insulin pump that will automatically inject insulin into the body as prescribed by our algorithm.

URLhttp://dx.doi.org/10.1109/CCECE.2007.149
DOI10.1109/CCECE.2007.149

a place of mind, The University of British Columbia

Electrical and Computer Engineering
2332 Main Mall
Vancouver, BC Canada V6T 1Z4
Tel +1.604.822.2872
Fax +1.604.822.5949
Email:

Emergency Procedures | Accessibility | Contact UBC | © Copyright 2020 The University of British Columbia