Deep Neural Networks for Automatic Classification of Anesthetic-Induced Unconsciousness

K. Patlatzoglou*, S. Chennu, M. Boly, Q. Noirhomme, V. Bonhomme, J.F. Brichant, O. Gosseries, S. Laureys, S. Wang, V. Yamamoto, J. Su, Y. Yang, E. Jones, L. Iasemidis, T. Mitchell

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingAcademicpeer-review


Despite the common use of anesthetics to modulate consciousness in the clinic, brain-based monitoring of consciousness is uncommon. We combined electroencephalographic measurement of brain activity with deep neural networks to automatically discriminate anesthetic states induced by propofol. Our results with leave-one-participant-out-cross-validation show that convolutional neural networks significantly outperform multilayer perceptrons in discrimination accuracy when working with raw time series. Perceptrons achieved comparable accuracy when provided with power spectral densities. These findings highlight the potential of deep convolutional networks for completely automatic extraction of useful spatio-temporo-spectral features from human EEG.
Original languageEnglish
Title of host publicationBrain Informatics
Subtitle of host publicationInternational Conference, BI 2018, Arlington, TX, USA, December 7–9, 2018, Proceedings
EditorsShouyi Wang, Vicky Yamamoto, Jianzhong Su, Yang Yang, Erick Jones, Leon Iasemidis, Tom Mitchell
Number of pages10
ISBN (Print)9783030055868
Publication statusPublished - 2018
EventInternational Conference on Brain Informatics (BI) - Arlington, United States
Duration: 7 Dec 20189 Dec 2018

Publication series

SeriesLecture Notes in Artificial Intelligence


ConferenceInternational Conference on Brain Informatics (BI)
Country/TerritoryUnited States
Internet address


  • Consciousness
  • Anesthesia
  • EEG
  • Deep learning


Dive into the research topics of 'Deep Neural Networks for Automatic Classification of Anesthetic-Induced Unconsciousness'. Together they form a unique fingerprint.

Cite this