External Artificial Pancreas for Type 1 Diabetes
Author | : Ahmad Haidar |
Publisher | : |
Total Pages | : |
Release | : 2013 |
ISBN-10 | : OCLC:922008077 |
ISBN-13 | : |
Rating | : 4/5 ( Downloads) |
Download or read book External Artificial Pancreas for Type 1 Diabetes written by Ahmad Haidar and published by . This book was released on 2013 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: "Most patients with type 1 diabetes do not achieve glycemic targets. Glucose-responsive closed-loop hormonal delivery that automatically regulates glucose levels (the "artificial pancreas") is a long-awaited goal for diabetes and its development was recently triggered by advances in glucose sensors and infusion pumps. The objective of this thesis is to accel-erate the development of closed-loop delivery systems. Closed-loop regulation of postprandial glucose remains a major challenge, but could be overcome with better understanding of postprandial glucose fluxes. However, ill-conditioning of the computational problem accompanying dual- and triple-tracer experimental techniques hinder their assessment. In this thesis, a novel computational method based on Bayesian inference to estimate postprandial glucose fluxes is presented. The method is then used to validate the dual- and triple-tracer techniques against an independent standard and shows that both methods provide accurate estimates of glucose appearance. Pre-clinical testing of closed-loop systems in a computer-simulation environment facilitates optimization of dosing algorithms. In this thesis, a novel fully probabilistic method to generate stochastic virtual patients for the assessment of closed-loop systems is presented. The method performs one to one mapping of an individual's experimental data into a stochastic virtual patient in a process we termed "stochastic e-cloning". Finally, a novel bi-hormonal (insulin and glucagon) closed-loop delivery system is presented. The dosing algorithm is based on fuzzy-supervised model predictive controller combined with extended Kalman filtering and a set of heuristic rules. The system is compared with conventional pump therapy in 15 patients for a 15-hour experimental period (4pm-7am) utilizing a randomized crossover study design. The closed-loop system significantly improves glucose control and reduces the risk of hypoglycemia." --