Publication

Novel drug-independent sedation level estimation based on machine learning of quantitative frontal electroencephalogram features in healthy volunteers

Ramaswamy, S. M., Kuizenga, M. H., Weerink, M. A. S., Vereecke, H. E. M., Struys, M. M. R. F. & Nagaraj, S. B., Oct-2019, In : British Journal of Anaesthesia. 123, 4, p. 479-487 9 p.

Research output: Contribution to journalArticleAcademicpeer-review

APA

Ramaswamy, S. M., Kuizenga, M. H., Weerink, M. A. S., Vereecke, H. E. M., Struys, M. M. R. F., & Nagaraj, S. B. (2019). Novel drug-independent sedation level estimation based on machine learning of quantitative frontal electroencephalogram features in healthy volunteers. British Journal of Anaesthesia, 123(4), 479-487. https://doi.org/10.1016/j.bja.2019.06.004

Author

Ramaswamy, Sowmya M ; Kuizenga, Merel H ; Weerink, Maud A S ; Vereecke, Hugo E M ; Struys, Michel M R F ; Nagaraj, Sunil B. / Novel drug-independent sedation level estimation based on machine learning of quantitative frontal electroencephalogram features in healthy volunteers. In: British Journal of Anaesthesia. 2019 ; Vol. 123, No. 4. pp. 479-487.

Harvard

Ramaswamy, SM, Kuizenga, MH, Weerink, MAS, Vereecke, HEM, Struys, MMRF & Nagaraj, SB 2019, 'Novel drug-independent sedation level estimation based on machine learning of quantitative frontal electroencephalogram features in healthy volunteers', British Journal of Anaesthesia, vol. 123, no. 4, pp. 479-487. https://doi.org/10.1016/j.bja.2019.06.004

Standard

Novel drug-independent sedation level estimation based on machine learning of quantitative frontal electroencephalogram features in healthy volunteers. / Ramaswamy, Sowmya M; Kuizenga, Merel H; Weerink, Maud A S; Vereecke, Hugo E M; Struys, Michel M R F; Nagaraj, Sunil B.

In: British Journal of Anaesthesia, Vol. 123, No. 4, 10.2019, p. 479-487.

Research output: Contribution to journalArticleAcademicpeer-review

Vancouver

Ramaswamy SM, Kuizenga MH, Weerink MAS, Vereecke HEM, Struys MMRF, Nagaraj SB. Novel drug-independent sedation level estimation based on machine learning of quantitative frontal electroencephalogram features in healthy volunteers. British Journal of Anaesthesia. 2019 Oct;123(4):479-487. https://doi.org/10.1016/j.bja.2019.06.004


BibTeX

@article{05f68264c76b43dfa281b5407d2b147e,
title = "Novel drug-independent sedation level estimation based on machine learning of quantitative frontal electroencephalogram features in healthy volunteers",
abstract = "BACKGROUND: Sedation indicators based on a single quantitative EEG (QEEG) feature have been criticised for their limited performance. We hypothesised that integration of multiple QEEG features into a single sedation-level estimator using a machine learning algorithm could reliably predict levels of sedation, independent of the sedative drug used.METHODS: In total, 102 subjects receiving propofol (N=36; 16 male/20 female), sevoflurane (N=36; 16 male/20 female), or dexmedetomidine (N=30; 15 male/15 female) were included in this study of healthy volunteers. Sedation level was assessed using the Modified Observer's Assessment of Alertness/Sedation (MOAA/S) score. We used 44 QEEG features estimated from the EEG data in a logistic regression algorithm, and an elastic-net regularisation method was used for feature selection. The area under the receiver operator characteristic curve (AUC) was used to assess the performance of the logistic regression model.RESULTS: The performances obtained when the system was trained and tested as drug-dependent mode to distinguish between awake and sedated states (mean AUC [standard deviation]) were propofol=0.97 (0.03), sevoflurane=0.74 (0.25), and dexmedetomidine=0.77 (0.10). The drug-independent system resulted in mean AUC=0.83 (0.17) to discriminate between the awake and sedated states.CONCLUSIONS: The incorporation of large numbers of QEEG features and machine learning algorithms is feasible for next-generation monitors of sedation level. Different QEEG features were selected for propofol, sevoflurane, and dexmedetomidine groups, but the sedation-level estimator maintained a high performance for predicting MOAA/S independent of the drug used.CLINICAL TRIAL REGISTRATION: NCT02043938; NCT03143972.",
author = "Ramaswamy, {Sowmya M} and Kuizenga, {Merel H} and Weerink, {Maud A S} and Vereecke, {Hugo E M} and Struys, {Michel M R F} and Nagaraj, {Sunil B}",
note = "Copyright {\circledC} 2019 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved.",
year = "2019",
month = "10",
doi = "10.1016/j.bja.2019.06.004",
language = "English",
volume = "123",
pages = "479--487",
journal = "British Journal of Anaesthesia",
issn = "0007-0912",
publisher = "ELSEVIER SCI LTD",
number = "4",

}

RIS

TY - JOUR

T1 - Novel drug-independent sedation level estimation based on machine learning of quantitative frontal electroencephalogram features in healthy volunteers

AU - Ramaswamy, Sowmya M

AU - Kuizenga, Merel H

AU - Weerink, Maud A S

AU - Vereecke, Hugo E M

AU - Struys, Michel M R F

AU - Nagaraj, Sunil B

N1 - Copyright © 2019 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved.

PY - 2019/10

Y1 - 2019/10

N2 - BACKGROUND: Sedation indicators based on a single quantitative EEG (QEEG) feature have been criticised for their limited performance. We hypothesised that integration of multiple QEEG features into a single sedation-level estimator using a machine learning algorithm could reliably predict levels of sedation, independent of the sedative drug used.METHODS: In total, 102 subjects receiving propofol (N=36; 16 male/20 female), sevoflurane (N=36; 16 male/20 female), or dexmedetomidine (N=30; 15 male/15 female) were included in this study of healthy volunteers. Sedation level was assessed using the Modified Observer's Assessment of Alertness/Sedation (MOAA/S) score. We used 44 QEEG features estimated from the EEG data in a logistic regression algorithm, and an elastic-net regularisation method was used for feature selection. The area under the receiver operator characteristic curve (AUC) was used to assess the performance of the logistic regression model.RESULTS: The performances obtained when the system was trained and tested as drug-dependent mode to distinguish between awake and sedated states (mean AUC [standard deviation]) were propofol=0.97 (0.03), sevoflurane=0.74 (0.25), and dexmedetomidine=0.77 (0.10). The drug-independent system resulted in mean AUC=0.83 (0.17) to discriminate between the awake and sedated states.CONCLUSIONS: The incorporation of large numbers of QEEG features and machine learning algorithms is feasible for next-generation monitors of sedation level. Different QEEG features were selected for propofol, sevoflurane, and dexmedetomidine groups, but the sedation-level estimator maintained a high performance for predicting MOAA/S independent of the drug used.CLINICAL TRIAL REGISTRATION: NCT02043938; NCT03143972.

AB - BACKGROUND: Sedation indicators based on a single quantitative EEG (QEEG) feature have been criticised for their limited performance. We hypothesised that integration of multiple QEEG features into a single sedation-level estimator using a machine learning algorithm could reliably predict levels of sedation, independent of the sedative drug used.METHODS: In total, 102 subjects receiving propofol (N=36; 16 male/20 female), sevoflurane (N=36; 16 male/20 female), or dexmedetomidine (N=30; 15 male/15 female) were included in this study of healthy volunteers. Sedation level was assessed using the Modified Observer's Assessment of Alertness/Sedation (MOAA/S) score. We used 44 QEEG features estimated from the EEG data in a logistic regression algorithm, and an elastic-net regularisation method was used for feature selection. The area under the receiver operator characteristic curve (AUC) was used to assess the performance of the logistic regression model.RESULTS: The performances obtained when the system was trained and tested as drug-dependent mode to distinguish between awake and sedated states (mean AUC [standard deviation]) were propofol=0.97 (0.03), sevoflurane=0.74 (0.25), and dexmedetomidine=0.77 (0.10). The drug-independent system resulted in mean AUC=0.83 (0.17) to discriminate between the awake and sedated states.CONCLUSIONS: The incorporation of large numbers of QEEG features and machine learning algorithms is feasible for next-generation monitors of sedation level. Different QEEG features were selected for propofol, sevoflurane, and dexmedetomidine groups, but the sedation-level estimator maintained a high performance for predicting MOAA/S independent of the drug used.CLINICAL TRIAL REGISTRATION: NCT02043938; NCT03143972.

U2 - 10.1016/j.bja.2019.06.004

DO - 10.1016/j.bja.2019.06.004

M3 - Article

C2 - 31326088

VL - 123

SP - 479

EP - 487

JO - British Journal of Anaesthesia

JF - British Journal of Anaesthesia

SN - 0007-0912

IS - 4

ER -

ID: 92360631