Event Abstract

Back to Event

PCI & auditory ERPs for the quantification of the level of consciousness: an EEG-based methods comparison study applied to disorders of consciousness.

Federico Raimondo1, 2, 3, 4*, Audrey Wolff1, Leandro R. Sanz1, Silvia Casarotto5, Matteo Fecchio5, Séverine Blandiaux1, Olivier Bodart1, Alice Barra1, Angela Comanducci5, Renate Rutiku5, Jitka Annen1, Mario Rosanova5, Marcello Massimini5, Jacobo D. Sitt2, 3, Steven Laureys1 and Olivia Gosseries1
  • 1 GIGA Consciousness, University of Liège, Belgium
  • 2 INSERM U1127 Institut du Cerveau et de la Moelle épinière, France
  • 3 Sorbonne Université, France
  • 4 Applied Artificial Intelligence Lab, Department of Computer Science, University of Buenos Aires, Argentina
  • 5 Dipartimento di Scienze biomediche e cliniche Luigi Sacco, Università degli Studi di Milano, Italy

Aims Patients with disorders of consciousness (DOC) suffer from lack of awareness at different levels[1]. Clinical categories range from reflexive behaviour (Unresponsive Wakefulness Syndrome, UWS), to more complex purposeful interaction with the environment such as visual pursuit (Minimally Conscious State minus, MCS-), response to command (MCS+) or capacity to communicate and/or functionally use objects (Emergence from the Minimally Conscious State, EMCS). A challenge exists when diagnosing DOC patients. The gold standard diagnosis is performed by repeated and standardised behavioural assessments[2] (Coma Recovery Scale-Revised). Objective neuroimaging techniques are utilized to support clinical diagnosis. Compared to neuroimaging, EEG-based systems have the advantage of being repeatable and cheap. Two of the most reliable methods are the Perturbational Complexity Index[3] (PCI), driven by the Integrated Information Theory[4] (IIT), and a synergy of EEG-extracted markers[5], [6] from a standardised auditory oddball paradigm (EEG-ERP) that combines information content, information sharing and global workspace theory[7] (GWT) markers. This study, which is part of the Human Brain Project (SP3), aims to confront the results from these two methods when applied to the diagnosis of Disorders of Consciousness. Methods A multi-centre group of 25 DOC patients (i.e., UWS, MCS and EMCS) were subject to both EEG recordings. The PCI value was computed by compressing the spatiotemporal pattern of cortical responses to the perturbation of the cortex with Transcranial Magnetic Stimulation. We then extracted 120 markers, corresponding to quantification of power spectrum and complexity in individual EEG sensors and information sharing between EEG sensors. Finally, we contrasted the obtained PCI values, the values of the EEG-extracted markers and the predicted individual probability of being (minimally) conscious using machine learning. Results When we analysed the relationship between PCI and the most informative EEG-extracted markers, we found a significant correlation with Weighted Symbolic Mutual Information (r=0.42, p=0.034) but not with Alpha Power (r=0.37, p=0.06) and Kolmogorov Complexity (r=0.1, p=0.63). For the multivariate approach, PCI and EEG markers provided a consistent diagnosis for 19 patients (76%) and correlated positively (r=0.59, p=0.002). All UWS (N=5), all EMCS (N=4) and 9 MCS patients were correctly diagnosed by both methods. However, 1 MCS patient was misdiagnosed by both methods, and 6 MCS patients had inconsistent results between the measures. Conclusions PCI correlated positively with the combination of EEG markers in severely brain-injured patients, but not with all the markers independently. Although the EEG-ERP method was designed to probe the GWT, the combination of EEG markers used with machine learning is driven by multiple theories, including IIT and GWT. Case mismatches can be explained by the diverging purposes of the methods: while the PCI is designed to probe capacity for consciousness, EEG-ERP characterises the current state of consciousness. These findings suggest that the shared common background is also evident in the results, providing a validation of the methods and a link between these theories.

References

[1] J. T. Giacino, J. J. Fins, S. Laureys, and N. D. Schiff, “Disorders of consciousness after acquired brain injury: The state of the science,” Nat. Rev. Neurol., vol. 10, no. 2, pp. 99–114, 2014. [2] J. T. Giacino, K. Kalmar, and J. Whyte, “The JFK Coma Recovery Scale-Revised: Measurement characteristics and diagnostic utility,” Arch. Phys. Med. Rehabil., vol. 85, no. 12, pp. 2020–2029, 2004. [3] A. G. Casali et al., “A Theoretically Based Index of Consciousness Independent of Sensory Processing and Behavior,” Sci. Transl. Med., vol. 5, no. 198, pp. 198ra105-198ra105, 2013. [4] G. Tononi, M. Boly, M. Massimini, and C. Koch, “Integrated information theory: From consciousness to its physical substrate,” Nat. Rev. Neurosci., vol. 17, no. 7, pp. 450–461, 2016. [5] J. D. Sitt et al., “Large scale screening of neural signatures of consciousness in patients in a vegetative or minimally conscious state,” Brain, vol. 137, pp. 2258–2270, 2014. [6] D. Engemann et al., “Automated Measurement and Prediction of Consciousness in Vegetative and Minimally Conscious Patients,” ICML Workshop Stat. Mach. Learn. Neurosci. Stamlins 2015, 2015. [7] S. Dehaene, M. Kerszberg, and J. P. Changeux, “A neuronal model of a global workspace in effortful cognitive tasks,” Proc Natl Acad Sci USA, p. 6, 1998.

Keywords: EEG, disorders of consciousness, TMS-EEG, machine learning, PCI

Conference: Belgian Brain Congress 2018 — Belgian Brain Council, LIEGE, Belgium, 19 Oct - 19 Oct, 2018.

Presentation Type: e-posters

Topic: NOVEL STRATEGIES FOR NEUROLOGICAL AND MENTAL DISORDERS: SCIENTIFIC BASIS AND VALUE FOR PATIENT-CENTERED CARE

Citation: Raimondo F, Wolff A, Sanz LR, Casarotto S, Fecchio M, Blandiaux S, Bodart O, Barra A, Comanducci A, Rutiku R, Annen J, Rosanova M, Massimini M, Sitt JD, Laureys S and Gosseries O (2019). PCI & auditory ERPs for the quantification of the level of consciousness: an EEG-based methods comparison study applied to disorders of consciousness.. Front. Neurosci. Conference Abstract: Belgian Brain Congress 2018 — Belgian Brain Council. doi: 10.3389/conf.fnins.2018.95.00093

Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.

The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.

Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.

For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.

Received: 31 Aug 2018; Published Online: 17 Jan 2019.

* Correspondence: Mr. Federico Raimondo, GIGA Consciousness, University of Liège, Liège, Liège, Belgium, federaimondo@gmail.com