Comparison of three nonlinear filters for fault detection in continuous glucose monitors.

Zeinab Mahmoudi; Sabrina Lyngbye Wendt; Dimitri Boiroux; Morten Hagdrup; Kirsten Norgaard; Niels Kjolstad Poulsen; Henrik Madsen; John Bagterp Jorgensen

doi:10.1109/EMBC.2016.7591484

Comparison of three nonlinear filters for fault detection in continuous glucose monitors.

Zeinab Mahmoudi, Sabrina Lyngbye Wendt, Dimitri Boiroux, Morten Hagdrup, Kirsten Norgaard, Niels Kjolstad Poulsen, Henrik Madsen, John Bagterp Jorgensen

Endokrinologisk Afdeling

8 Citationer (Scopus)

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%.

Originalsprog	Engelsk
Tidsskrift	Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference
Vol/bind	2016
Sider (fra-til)	3507-3510
Antal sider	4
ISSN	2375-7477
DOI	https://doi.org/10.1109/EMBC.2016.7591484
Status	Udgivet - 2016

Adgang til dokumentet

10.1109/EMBC.2016.7591484

Citationsformater

Mahmoudi, Z., Wendt, S. L., Boiroux, D., Hagdrup, M., Norgaard, K., Poulsen, N. K., Madsen, H., & Jorgensen, J. B. (2016). Comparison of three nonlinear filters for fault detection in continuous glucose monitors. Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference, 2016, 3507-3510. https://doi.org/10.1109/EMBC.2016.7591484

Comparison of three nonlinear filters for fault detection in continuous glucose monitors. / Mahmoudi, Zeinab; Wendt, Sabrina Lyngbye; Boiroux, Dimitri et al.
I: Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference, Bind 2016, 2016, s. 3507-3510.

Publikation: Bidrag til tidsskrift › Konferenceartikel › Forskning › peer review

Mahmoudi, Z, Wendt, SL, Boiroux, D, Hagdrup, M, Norgaard, K, Poulsen, NK, Madsen, H & Jorgensen, JB 2016, 'Comparison of three nonlinear filters for fault detection in continuous glucose monitors.', Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference, bind 2016, s. 3507-3510. https://doi.org/10.1109/EMBC.2016.7591484

Mahmoudi Z, Wendt SL, Boiroux D, Hagdrup M, Norgaard K, Poulsen NK et al. Comparison of three nonlinear filters for fault detection in continuous glucose monitors. Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference. 2016;2016:3507-3510. doi: 10.1109/EMBC.2016.7591484

Mahmoudi, Zeinab ; Wendt, Sabrina Lyngbye ; Boiroux, Dimitri et al. / Comparison of three nonlinear filters for fault detection in continuous glucose monitors. I: Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference. 2016 ; Bind 2016. s. 3507-3510.

@inproceedings{98ac489e18fc418ba8aae52e8a616d9f,

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%.",

keywords = "Blood Chemical Analysis, Blood Glucose, Humans, Models, Biological, Nonlinear Dynamics, Normal Distribution, Signal Processing, Computer-Assisted, Comparative Study, Journal Article, Research Support, Non-U.S. Gov't",

author = "Zeinab Mahmoudi and Wendt, {Sabrina Lyngbye} and Dimitri Boiroux and Morten Hagdrup and Kirsten Norgaard and Poulsen, {Niels Kjolstad} and Henrik Madsen and Jorgensen, {John Bagterp}",

year = "2016",

doi = "10.1109/EMBC.2016.7591484",

language = "English",

volume = "2016",

pages = "3507--3510",

journal = "Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference",

issn = "2375-7477",

publisher = "IEEE",

}

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 - Norgaard, Kirsten

AU - Poulsen, Niels Kjolstad

AU - Madsen, Henrik

AU - Jorgensen, John Bagterp

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%.

KW - Blood Chemical Analysis

KW - Blood Glucose

KW - Humans

KW - Models, Biological

KW - Nonlinear Dynamics

KW - Normal Distribution

KW - Signal Processing, Computer-Assisted

KW - Comparative Study

KW - Journal Article

KW - Research Support, Non-U.S. Gov't

U2 - 10.1109/EMBC.2016.7591484

DO - 10.1109/EMBC.2016.7591484

M3 - Conference article

C2 - 28269054

SN - 2375-7477

VL - 2016

SP - 3507

EP - 3510

JO - Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference

JF - Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference

ER -

Comparison of three nonlinear filters for fault detection in continuous glucose monitors.

Abstract

Adgang til dokumentet

Fingeraftryk

Citationsformater