Research Study Abstract

ARX model for interstitial glucose prediction during and after physical activities

  • Published on July 23, 2019

Abstract
This paper presents the first autoregressive with exogenous input (ARX) model using energy expenditure, carbohydrates on board, and insulin on board as input to predict interstitial glucose (IG). The proposed model may be used for predicting IG even during physical activity (PA). A population-based model, obtained from a first database composed of 14 type 1 diabetes (T1D) patients, achieved a root-mean-square error (RMSE) of mg/dL, on IG prediction (30-min ahead) at the end of a PA, on a second database (15 T1D patients). Patient-specific ARX models, obtained on the second database, improved prediction accuracy (RMSE = mg/dL), outperforming the results found in the literature.

Author(s)

  • Hector M. Romero-Ugalde 1,2,6
  • M. Garnotel 3
  • M. Doron 1,2
  • P. Jallon 1,2
  • G. Charpentier 4,5
  • S. Franc 4,5
  • E. Huneker 6
  • C. Simon 3
  • S. Bonnet 1,2

Institution(s)

  • 1

    Univ. Grenoble Alpes, F-38000 Grenoble, France
  • 2

    CEA, LETI, MINATEC Campus, F-38054 Grenoble, France
  • 3

    CARMEN INSERM U1060/Université de Lyon 1/INRA U1235, CRNH-Rhône-Alpes, Lyon, France
  • 4

    Centre Hospitalier Sud-Francilien, Department of Diabetes and Endocrinology, Corbeil-Essones, France
  • 5

    Centre d’Etudes et de Recherche pour l’Intensification du Traitement du Diabète (CERITD), Corbeil-Essonnes, France
  • 6

    Diabeloop SA, 155 Cours Berriat, F-38000 Grenoble, France

Journal

Control Engineering Process

Categories