Your browser does not support JavaScript!

Graph signal processing analysis of NIRS signals for brain–computer interfaces

Graph signal processing analysis of NIRS signals for brain–computer interfaces

For access to this article, please select a purchase option:

Buy chapter PDF
(plus tax if applicable)
Buy Knowledge Pack
10 chapters for $120.00
(plus taxes if applicable)

IET members benefit from discounts to all IET publications and free access to E&T Magazine. If you are an IET member, log in to your account and the discounts will automatically be applied.

Learn more about IET membership 

Recommend Title Publication to library

You must fill out fields marked with: *

Librarian details
Your details
Why are you recommending this title?
Select reason:
Signal Processing to Drive Human-Computer Interaction: EEG and eye-controlled interfaces — Recommend this title to your library

Thank you

Your recommendation has been sent to your librarian.

Graph signal processing (GSP) is an emerging field in signal processing that aims at analyzing high-dimensional signals using graphs. The GSP analysis is intended to take into account the signals' inner graphical structure and expand traditional signal processing techniques to the graph-network domain. In this chapter, we present a GSP analysis framework for the implementation of brain-computer interfaces (BCI) based on function near-infrared spectroscopy (NIRS) signals. Firstly, a GSP approach for feature extraction is presented based on the Graph Fourier Transform (GFT). The aforementioned approach captures the spatial information of the NIRS signals. The feature extraction method is applied on a publicly available dataset ofNIRS recordings during mental arithmetic task and shows higher classification rates, up to 92.52%, as compared to the classification rates of two state-of-the-art feature extraction methodologies. Moreover, in order to better demonstrate the spatial distribution of the NIRS information and to quantify the smoothness or not of the NIRS signals across the channel montage we present a GSP, Dirichlet energy-based analysis approach of NIRS signals over a graph. The application of the proposed measure on the same NIRS dataset further shows the spatial characteristics of the NIRS data and the efficiency of this GSP approach to capture it. Moreover, Dirichlet energy-based approach shows high classification rates, >97%, when used to extract features from NIRS signals. In sum, the presented methods show the efficacy of the GSP-based analysis of NIRS signals for BCI applications and pave the way for more robust and efficient implementations.

Chapter Contents:

  • 10.1 Introduction
  • 10.2 NIRS dataset
  • 10.3 Materials and methods
  • 10.3.1 Graph signal processing basics
  • 10.3.2 Dirichlet energy over a graph
  • 10.3.3 Graph construction algorithm
  • 10.3.4 Feature extraction
  • GFT feature vector
  • S feature vector
  • CC feature vector
  • 10.3.5 Classification
  • 10.3.6 Implementation issues
  • 10.4 Results
  • 10.5 Discussion
  • 10.6 Summary
  • References

Inspec keywords: medical signal processing; Fourier transforms; feature extraction; brain-computer interfaces; infrared spectra; electroencephalography; graph theory; signal classification

Other keywords: GSP-based analysis; function near-infrared spectroscopy signals; Graph signal processing analysis; GSP approach; NIRS signals; high-dimensional signals; GSP analysis; brain-computer interfaces; graph Fourier transform; feature extraction

Subjects: Combinatorial mathematics; Function theory, analysis; Electrodiagnostics and other electrical measurement techniques; Knowledge engineering techniques; User interfaces; Integral transforms; Bioelectric signals; Integral transforms; Electrical activity in neurophysiological processes; Algebra, set theory, and graph theory; Biology and medical computing; Signal processing and detection; Digital signal processing; Combinatorial mathematics

Preview this chapter:
Zoom in

Graph signal processing analysis of NIRS signals for brain–computer interfaces, Page 1 of 2

| /docserver/preview/fulltext/books/ce/pbce129e/PBCE129E_ch10-1.gif /docserver/preview/fulltext/books/ce/pbce129e/PBCE129E_ch10-2.gif

Related content

This is a required field
Please enter a valid email address