Item


A CBR-based bolus recommender system for type 1 diabetes

Comunicació presentada a: Workshop on Artificial Intelligence for Diabetes (2nd: 2017: Viena, Àustria). Aquest workshop ha rebut finançament del programa d’investigació i innovació EU Horizon 2020 sota el núm. d’ajut 689810

People with type 1 diabetes mellitus usually need to administer bolus insulin before each meal to keep the blood glucose level in the target glycaemic range. However, the factors involved in the calculation of the appropriate dose can change due to multiple factors and with an unknown relation. This may increase the error in the bolus calculation, and therefore, increase the chances of hypoglycaemia and hyperglycaemia. This paper proposes a bolus recommender system based on case based reasoning developed under project PEPPER, with the objective of recommending personalised and adaptive bolus doses. The system has been tested with in silico adults with UVA/PADOVA T1DM simulator. Results show that the use of the proposed bolus recommender system increases the percentage of time in the target glycaemic range

This project has received funding from the grant of the University of Girona 2016-2018 (MPCUdG2016) and the European Union Horizon 2020 research and innovation programme under grant agreement No. 689810, www.pepper.eu.com/, PEPPER. The work has been developed with the support of the research group SITES awarded with distinction by the Generalitat de Catalunya (SGR 2014-2016).

Artificial Intelligence for Diabetes (AID), Artificial Intelligence in Medicine (AIME), PEPPER

Author: Torrent-Fontbona, Ferran
López Ibáñez, Beatriz
Pozo-Alonso, Alejandro
Date: 2018 June 5
Abstract: Comunicació presentada a: Workshop on Artificial Intelligence for Diabetes (2nd: 2017: Viena, Àustria). Aquest workshop ha rebut finançament del programa d’investigació i innovació EU Horizon 2020 sota el núm. d’ajut 689810
People with type 1 diabetes mellitus usually need to administer bolus insulin before each meal to keep the blood glucose level in the target glycaemic range. However, the factors involved in the calculation of the appropriate dose can change due to multiple factors and with an unknown relation. This may increase the error in the bolus calculation, and therefore, increase the chances of hypoglycaemia and hyperglycaemia. This paper proposes a bolus recommender system based on case based reasoning developed under project PEPPER, with the objective of recommending personalised and adaptive bolus doses. The system has been tested with in silico adults with UVA/PADOVA T1DM simulator. Results show that the use of the proposed bolus recommender system increases the percentage of time in the target glycaemic range
This project has received funding from the grant of the University of Girona 2016-2018 (MPCUdG2016) and the European Union Horizon 2020 research and innovation programme under grant agreement No. 689810, www.pepper.eu.com/, PEPPER. The work has been developed with the support of the research group SITES awarded with distinction by the Generalitat de Catalunya (SGR 2014-2016).
Document access: http://hdl.handle.net/2072/319716
Language: eng
Publisher: Artificial Intelligence for Diabetes (AID), Artificial Intelligence in Medicine (AIME), PEPPER
Rights: Tots els drets reservats
Subject: Diabetis
Diabetes
Title: A CBR-based bolus recommender system for type 1 diabetes
Type: info:eu-repo/semantics/article
Repository: Recercat

Subjects

Authors