A Bayesian network for modelling blood glucose concentration and exercise in type 1 diabetes
- Submitting institution
-
University of Southampton
- Unit of assessment
- 12 - Engineering
- Output identifier
- 20670492
- Type
- D - Journal article
- DOI
-
10.1177/0962280214520732
- Title of journal
- Statistical Methods in Medical Research
- Article number
- -
- First page
- 342
- Volume
- 24
- Issue
- 3
- ISSN
- 0962-2802
- Open access status
- Out of scope for open access requirements
- Month of publication
- February
- Year of publication
- 2014
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
4
- Research group(s)
-
-
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Glycaemic excursions in people with diabetes are difficult to predict and if unmanaged can lead to serious complications. This article provides a new statistical approach to blood glucose prediction, developed from real-time human volunteer data, that can account for physical activity. Existing parametric models, e.g. ARMAX, are unable to predict blood glucose levels reliably beyond a short, 1-2 hour window, and are thus of little value in therapies such as the artificial pancreas. The ideas in this collaborative work between engineering, mathematic and medicine have been adopted by researchers, particularly in addressing the influence of confounders in predictive diabetic control.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -