Comparison of Three Nonlinear Filters for Fault Detection in Continuous Glucose Monitors

Zeinab Mahmoudi; Sabrina Lyngbye Wendt; Dimitri Boiroux; Morten Hagdrup; Kirsten Nørgaard; Niels Kjølstad Poulsen; Henrik Madsen; John Bagterp Jørgensen

Comparison of Three Nonlinear Filters for Fault Detection in Continuous Glucose Monitors

Zeinab Mahmoudi, Sabrina Lyngbye Wendt, Dimitri Boiroux, Morten Hagdrup, Kirsten Nørgaard, Niels Kjølstad Poulsen, Henrik Madsen, John Bagterp Jørgensen

Copenhagen University Hospital Herlev and Gentofte

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Abstract

The purpose of this study is to compare the performance of three nonlinear filters in online drift detection of continuous glucose monitors. The nonlinear filters are the extended Kalman filter (EKF), the unscented Kalman filter (UKF), and the particle filter (PF). They are all based on a nonlinear model of the glucose-insulin dynamics in people with type 1 diabetes. Drift is modelled by a Gaussian random walk and is detected based on the statistical tests of the 90-min prediction residuals of the filters. The unscented Kalman filter had the highest average F score of 85.9%, and the smallest average detection delay of 84.1%, with the average detection sensitivity of 82.6%, and average specificity of 91.0%.

Original language	English
Title of host publication	Proceedings of the 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC’2016)
Publisher	IEEE
Publication date	2016
Pages	3507-3510
ISBN (Print)	978-1-4577-0220-4
Publication status	Published - 2016
Event	38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society - Orlando, United States Duration: 16 Aug 2016 → 20 Aug 2016 Conference number: 38 http://embc.embs.org/2016/

Conference

Conference	38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society
Number	38
Country/Territory	United States
City	Orlando
Period	16/08/2016 → 20/08/2016
Internet address	http://embc.embs.org/2016/

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

OpenUrl availability

Full text

Cite this

@inproceedings{61b8b595dcdf4054b1e24e284a34e58f,

title = "Comparison of Three Nonlinear Filters for Fault Detection in Continuous Glucose Monitors",

abstract = "The purpose of this study is to compare the performance of three nonlinear filters in online drift detection of continuous glucose monitors. The nonlinear filters are the extended Kalman filter (EKF), the unscented Kalman filter (UKF), and the particle filter (PF). They are all based on a nonlinear model of the glucose-insulin dynamics in people with type 1 diabetes. Drift is modelled by a Gaussian random walk and is detected based on the statistical tests of the 90-min prediction residuals of the filters. The unscented Kalman filter had the highest average F score of 85.9%, and the smallest average detection delay of 84.1%, with the average detection sensitivity of 82.6%, and average specificity of 91.0%.",

author = "Zeinab Mahmoudi and Wendt, {Sabrina Lyngbye} and Dimitri Boiroux and Morten Hagdrup and Kirsten N{\o}rgaard and Poulsen, {Niels Kj{\o}lstad} and Henrik Madsen and J{\o}rgensen, {John Bagterp}",

year = "2016",

language = "English",

isbn = "978-1-4577-0220-4",

pages = "3507--3510",

booktitle = "Proceedings of the 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC{\textquoteright}2016)",

publisher = "IEEE",

address = "United States",

note = "38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016 ; Conference date: 16-08-2016 Through 20-08-2016",

url = "http://embc.embs.org/2016/",

}

Mahmoudi, Z, Wendt, SL, Boiroux, D, Hagdrup, M, Nørgaard, K, Poulsen, NK, Madsen, H & Jørgensen, JB 2016, Comparison of Three Nonlinear Filters for Fault Detection in Continuous Glucose Monitors. in Proceedings of the 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC’2016). IEEE, pp. 3507-3510, 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Orlando, Florida, United States, 16/08/2016.

Comparison of Three Nonlinear Filters for Fault Detection in Continuous Glucose Monitors. / Mahmoudi, Zeinab; Wendt, Sabrina Lyngbye; Boiroux, Dimitri et al.
Proceedings of the 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC’2016). IEEE, 2016. p. 3507-3510.

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

TY - GEN

T1 - Comparison of Three Nonlinear Filters for Fault Detection in Continuous Glucose Monitors

AU - Mahmoudi, Zeinab

AU - Wendt, Sabrina Lyngbye

AU - Boiroux, Dimitri

AU - Hagdrup, Morten

AU - Nørgaard, Kirsten

AU - Poulsen, Niels Kjølstad

AU - Madsen, Henrik

AU - Jørgensen, John Bagterp

N1 - Conference code: 38

PY - 2016

Y1 - 2016

N2 - The purpose of this study is to compare the performance of three nonlinear filters in online drift detection of continuous glucose monitors. The nonlinear filters are the extended Kalman filter (EKF), the unscented Kalman filter (UKF), and the particle filter (PF). They are all based on a nonlinear model of the glucose-insulin dynamics in people with type 1 diabetes. Drift is modelled by a Gaussian random walk and is detected based on the statistical tests of the 90-min prediction residuals of the filters. The unscented Kalman filter had the highest average F score of 85.9%, and the smallest average detection delay of 84.1%, with the average detection sensitivity of 82.6%, and average specificity of 91.0%.

AB - The purpose of this study is to compare the performance of three nonlinear filters in online drift detection of continuous glucose monitors. The nonlinear filters are the extended Kalman filter (EKF), the unscented Kalman filter (UKF), and the particle filter (PF). They are all based on a nonlinear model of the glucose-insulin dynamics in people with type 1 diabetes. Drift is modelled by a Gaussian random walk and is detected based on the statistical tests of the 90-min prediction residuals of the filters. The unscented Kalman filter had the highest average F score of 85.9%, and the smallest average detection delay of 84.1%, with the average detection sensitivity of 82.6%, and average specificity of 91.0%.

M3 - Article in proceedings

SN - 978-1-4577-0220-4

SP - 3507

EP - 3510

BT - Proceedings of the 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC’2016)

PB - IEEE

T2 - 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society

Y2 - 16 August 2016 through 20 August 2016

ER -

Comparison of Three Nonlinear Filters for Fault Detection in Continuous Glucose Monitors

Abstract

Conference

UN SDGs

OpenUrl availability

Fingerprint

Cite this