Detection of a Meal Using Continuous Glucose Monitoring (CGM): Implications for an Artificial β-cell
- Eyal Dassau, PhD*,
- B.Wayne Bequette, PhD†,
- Bruce A. Buckingham, MD‡ and
- Francis J. Doyle III, PhD (frank.doyle{at}icb.ucsb.edu)*
- *Department of Chemical Engineering, University of California Santa Barbara, Santa Barbara, CA 93106-5080
- †Department of Chemical and Biological Engineering, Rensselaer Polytechnic Institute, Troy, NY 12180-3590
- ‡Stanford Medical Center, Stanford, CA 94305-5208
Abstract
Objective: The purpose of this study was to introduce a novel meal detection algorithm (MDA) to be used as part of an artificial β-cell that utilizes continuous glucose monitoring (CGM).
Research Design And Methods: We developed our MDA on a data set of 26 meal events using records from nineteen children 1 to 6 year old who used the Medtronic Minimed CGMS Gold®. We then applied this algorithm to CGM records from a DirecNet pilot study of the FreeStyle Navigator® (Abbott Diabetes Care). During a research center admission, breakfast insulin was withheld for 1 hour and discrete glucose levels were obtained every 10 minutes following the meal.
Results: Based on the Navigator readings, the MDA detected a meal at a mean time of 30 minutes from the onset of eating, at which time the mean serum glucose was 21 mg/dL above baseline (range 2 to 36 mg/dL), and more than 90% of meals were detected before the glucose had risen 40 mg/dL from baseline.
Conclusion: The meal detection algorithm will enable automated insulin dosing in response to meals, facilitating the development of an artificial pancreas.
Footnotes
-
- Received July 6, 2007.
- Accepted October 21, 2007.
- Copyright © American Diabetes Association
