Categorized EEG neurofeedback performance unveils simultaneous fMRI deep brain activation

Sivan Kinreich*, Ilana Podlipsky, Nathan Intrator, Talma Hendler

*Corresponding author for this work

Research output: Contribution to journalConference articlepeer-review

Abstract

Decades of Electroencephalogram-NeuroFeedback (EEG-NF) practice have proven that people can be effectively trained to selectively regulate their brain activity, thus potentially improving performance. A common protocol of EEG-NF training aims to guide people via a closed-loop operation shifting from high-amplitude of alpha (8-14Hz) to high-amplitude of theta (4-7 Hz) oscillations resulting in greater theta/alpha ratio (T/A). The induction of such a shift in EEG oscillations has been shown to be useful in reaching a state of relaxation in psychiatric conditions of anxiety and mood disorders. However, the clinical implication of this practice remains elusive and is considered to have relatively low therapeutic yield, possibly due to its poor specificity to a unique brain mechanism. The current project aims to use simultaneous acquisition of Functional Magnetic Resonance Imaging (fMRI) and EEG in order to unfold in high spatial and temporal resolutions, respectively the neural modulations induced via T/A EEG-NF. We used real time EEG preprocessing and analysis during the simultaneous T/A EEG-NF/fMRI. A data driven algorithm was implemented off-line to categorize individual scans into responders and non-responders to the EEG-NF practice via a temporal signature of T/A continuous modulation. Comparing the two groups along with their parasympathetic Heart-Rate reactivity profile verified the relaxed state of the responders. Projection of responders variations in the T/A power to the fMRI whole brain maps revealed networks of correlated and inversely correlated activity reflecting induced relaxation, uniquely among responders.

Original languageEnglish
Pages (from-to)108-115
Number of pages8
JournalLecture Notes in Computer Science
Volume7263 LNAI
DOIs
StatePublished - 2012
EventInternational Workshop on Machine Learning and Interpretation in Neuroimaging, MLINI 2011, Held at Neural Information Processing, NIPS 2011 - Sierra Nevada, Spain
Duration: 16 Dec 201117 Dec 2011

Keywords

  • limbic network
  • neuro-feedback
  • simultaneous fMRI/EEG
  • theta /alpha ratio

Fingerprint

Dive into the research topics of 'Categorized EEG neurofeedback performance unveils simultaneous fMRI deep brain activation'. Together they form a unique fingerprint.

Cite this