An essential part of this work is to provide a data-driven model for predicting blood glucose levels that will help to warn the person with type 1 diabetes about a potential hypo- or hyperglycemic event in an easy-to-manage and discreet way. In this work, we apply a convolutional recurrent neural network on a real dataset of 6 contributors, provided by the University of Ohio [5]. Our model is capable of predicting glucose levels with high precision with a 30- minute horizon (RMSE = 17.45 [mg/dL] and MAE = 11.22 [mg/dL]), and RMSE = 33.67 [mg/dL] and MAE = 23.25 [mg/dL] for the 60- minute horizon. We believe this precision can greatly impact the long-term health condition as well as the daily management of people with type 1 diabetes.
Détails
Titre
A deep learning approach for blood glucose prediction of type 1 diabetes
Auteur(s)/ trice(s)
Freiburghaus, Jonas (School of Engineering – HE-Arc Ingénierie, HES-SO University of Applied Sciences Western Switzerland) Rizzotti, Aïcha (School of Engineering – HE-Arc Ingénierie, HES-SO University of Applied Sciences Western Switzerland) Albertetti, Fabrizio (School of Engineering – HE-Arc Ingénierie, HES-SO University of Applied Sciences Western Switzerland)
Date
2020-08
Publié dans
Proceedings of the Proceedings of the 5th International Workshop on Knowledge Discovery in Healthcare Data co-located with 24th European Conference on Artificial Intelligence (ECAI 2020), 29-30 August 2020, Santiago de Compostela, Spain
Volume
2020, vo. 2675
Editeur
Santiago de Compostela, Spain, 29-30 August 2020
Pagination
131-135
Présenté à
Knowledge Discovery in Healthcare Data (KDH), Santiago de Compostela, Spain, 2020-08-29, 2020-08-30
Type de papier
published full paper
Domaine
Ingénierie et Architecture
Ecole
HE-Arc Ingénierie
Institut
Aucun institut
Note
Due to the COVID-19 outbreak, The KDH 2020 conference venue in Santiago de Compostela, Spain was cancelled. The proceedings of the online conference are however published according to the original schedule.
