Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Predicting Glycemia in Type 1 Diabetes Mellitus with Subspace-Based Linear Multistep Predictors

Cescon, Marzia LU ; Johansson, Rolf LU orcid and Renard, Eric (2016) p.107-132
Abstract
A major challenge for a person with diabetes is to adapt insulin dosage regimens and food intake to keep blood glucose within tolerable limits during daily life activities. The accurate prediction of blood glucose levels in response to inputs would support the patients with invaluable information for appropriate on-the-spot decision making concerning the management of the disease. Against this background, in this paper we propose multistep data-driven predictors to the purpose of predicting blood glucose multiple steps ahead in the future. We formulate the predictors based on the state-space construction step in subspace identification methods for linear systems. The clinical data of 14 type 1 diabetic patients collected during a 3-days... (More)
A major challenge for a person with diabetes is to adapt insulin dosage regimens and food intake to keep blood glucose within tolerable limits during daily life activities. The accurate prediction of blood glucose levels in response to inputs would support the patients with invaluable information for appropriate on-the-spot decision making concerning the management of the disease. Against this background, in this paper we propose multistep data-driven predictors to the purpose of predicting blood glucose multiple steps ahead in the future. We formulate the predictors based on the state-space construction step in subspace identification methods for linear systems. The clinical data of 14 type 1 diabetic patients collected during a 3-days long hospital visit were used. We exploited physiological models from the literature to filter the raw information on carbohydrate and insulin intakes in order to retrieve the inputs signals to the predictors. Predictions were based on the collected CGMS measurements, recalibrated against finger stick samples and smoothed through a regularization step. Performances were assessed with respect to YSI blood glucose samples and compared to those achieved with a Kalman filter identified from data. Results proved the competitiveness of the proposed approach. (Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
host publication
Prediction Methods for Blood Glucose Concentration
editor
Kirchsteiger, Harald ; Bagterp Jørgensen, John ; Renard, Eric and del Re, Luigi
pages
26 pages
publisher
Springer International Publishing
ISBN
978-3-319-25913-0
DOI
10.1007/978-3-319-25913-0_7
project
DIAdvisor
language
English
LU publication?
yes
id
7bb360e2-cec0-4d0c-83d3-d7a5cf5f1369
date added to LUP
2021-11-18 22:46:18
date last changed
2021-11-25 14:47:12
@inbook{7bb360e2-cec0-4d0c-83d3-d7a5cf5f1369,
  abstract     = {{A major challenge for a person with diabetes is to adapt insulin dosage regimens and food intake to keep blood glucose within tolerable limits during daily life activities. The accurate prediction of blood glucose levels in response to inputs would support the patients with invaluable information for appropriate on-the-spot decision making concerning the management of the disease. Against this background, in this paper we propose multistep data-driven predictors to the purpose of predicting blood glucose multiple steps ahead in the future. We formulate the predictors based on the state-space construction step in subspace identification methods for linear systems. The clinical data of 14 type 1 diabetic patients collected during a 3-days long hospital visit were used. We exploited physiological models from the literature to filter the raw information on carbohydrate and insulin intakes in order to retrieve the inputs signals to the predictors. Predictions were based on the collected CGMS measurements, recalibrated against finger stick samples and smoothed through a regularization step. Performances were assessed with respect to YSI blood glucose samples and compared to those achieved with a Kalman filter identified from data. Results proved the competitiveness of the proposed approach.}},
  author       = {{Cescon, Marzia and Johansson, Rolf and Renard, Eric}},
  booktitle    = {{Prediction Methods for Blood Glucose Concentration}},
  editor       = {{Kirchsteiger, Harald and Bagterp Jørgensen, John and Renard, Eric and del Re, Luigi}},
  isbn         = {{978-3-319-25913-0}},
  language     = {{eng}},
  pages        = {{107--132}},
  publisher    = {{Springer International Publishing}},
  title        = {{Predicting Glycemia in Type 1 Diabetes Mellitus with Subspace-Based Linear Multistep Predictors}},
  url          = {{http://dx.doi.org/10.1007/978-3-319-25913-0_7}},
  doi          = {{10.1007/978-3-319-25913-0_7}},
  year         = {{2016}},
}