Speaker
Description
Background:
Monitoring anesthesia levels via EEG is essential to ensure adequate unconsciousness while minimizing complications such as intraoperative awareness or postoperative delirium. Established metrics like SEF95 and SpEn are robust but computationally demanding due to frequency domain transformation. This study evaluates whether low-complexity time domain EEG parameters can provide comparable performance to spectral measures in distinguishing awake and anaesthetized states \cite{1}.
Methods:
EEG data from 76 patients under general anesthesia were retrospectively analyzed using frontal recordings \cite{2}. Power spectral density was estimated via Welch’s method to derive SpEn and SEF95. The spectral exponent (SpEx), was obtained using the FOOOF algorithm, separating oscillatory from aperiodic (1/f) activity \cite{3}. Spectral parameters were compared to four low-complexity time-domain metrics (Benford-1-Index, Coastline, Entropy of Difference (EoD), and Zero-Crossing Rate (ZCR)) using Friedman tests with post-hoc comparisons.
Conclusion:
All parameters significantly distinguished ECBaseline from BeforeCut. SpEx increased with deeper anesthesia \cite{4}, while SpEn and SEF95 decreased. SpEx achieved the best performance. Benford-1-Index and EoD performed very well, while Coastline was weakest. ZCR performed similarly to SEF and SpEx, highlighting its potential as a reliable and efficient metric for real-time anesthesia monitoring, especially in resource-limited or time-critical settings.
Bibliography
@article{1,
title = {Clinical {Electroencephalography} for {Anesthesiologists}: {Part} {I}: {Background} and {Basic} {Signatures}},
volume = {123},
issn = {0003-3022},
shorttitle = {Clinical {Electroencephalography} for {Anesthesiologists}},
url = {https://journals.lww.com/10.1097/ALN.0000000000000841},
doi = {10.1097/ALN.0000000000000841},
language = {en},
number = {4},
urldate = {2026-05-11},
journal = {Anesthesiology},
author = {Purdon, Patrick L. and Sampson, Aaron and Pavone, Kara J. and Brown, Emery N.},
month = oct,
year = {2015},
pages = {937--960},
}
@article{2,
title = {State entropy and burst suppression ratio can show contradictory information: {A} retrospective study},
volume = {37},
issn = {0265-0215, 1365-2346},
shorttitle = {State entropy and burst suppression ratio can show contradictory information},
url = {https://journals.lww.com/10.1097/EJA.0000000000001312},
doi = {10.1097/EJA.0000000000001312},
language = {en},
number = {12},
urldate = {2026-05-11},
journal = {European Journal of Anaesthesiology},
author = {Georgii, Marie-Therese and Pilge, Stefanie and Schneider, Gerhard and Kreuzer, Matthias},
month = dec,
year = {2020},
pages = {1084--1092},
}
@article{3,
title = {Aperiodic component of the electroencephalogram power spectrum reflects the hypnotic level of anaesthesia},
volume = {134},
issn = {00070912},
url = {https://linkinghub.elsevier.com/retrieve/pii/S0007091224006482},
doi = {10.1016/j.bja.2024.09.027},
language = {en},
number = {2},
urldate = {2026-05-11},
journal = {British Journal of Anaesthesia},
author = {Widmann, Sandra and Ostertag, Julian and Zinn, Sebastian and Pilge, Stefanie and García, Paul S. and Kratzer, Stephan and Schneider, Gerhard and Kreuzer, Matthias},
month = feb,
year = {2025},
pages = {392--401},
}
@article{4,
title = {The spectral exponent of the resting {EEG} indexes the presence of consciousness during unresponsiveness induced by propofol, xenon, and ketamine},
volume = {189},
issn = {10538119},
url = {https://linkinghub.elsevier.com/retrieve/pii/S1053811919300242},
doi = {10.1016/j.neuroimage.2019.01.024},
language = {en},
urldate = {2026-05-11},
journal = {NeuroImage},
author = {Colombo, Michele Angelo and Napolitani, Martino and Boly, Melanie and Gosseries, Olivia and Casarotto, Silvia and Rosanova, Mario and Brichant, Jean-Francois and Boveroux, Pierre and Rex, Steffen and Laureys, Steven and Massimini, Marcello and Chieregato, Arturo and Sarasso, Simone},
month = apr,
year = {2019},
pages = {631--644},
}
