Steady state visual evoked potentials-based patient interface under breathing constraints - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2015

Steady state visual evoked potentials-based patient interface under breathing constraints

Mathieu Raux
  • Function : Author
  • PersonId : 932023

Abstract

Steady state visual evoked potentials (SSVEP) have been widely utilized in brain computer interfacing (BCI) in last years. In this paper, we present a study exploring the possibilities of SSVEP to manage the communication between patients suffering respiratory disorders and health care providers. By imposing different breathing constraints, five healthy subjects communicated their breathing sensations (breathing well/breathing bad) using a visual frequency tagging paradigm: two visual stimuli with different flickering frequencies (15 and 20 Hz) were simultaneously presented on a screen. Using electroencephalographic (EEG) signals from only three EEG electrodes, two spectral features were extracted by a spatial filter in a sliding window, then classified by an unsupervised algorithm based on k-medians. Average detection success rates were of 70% during breathing discomfort, and of 83% when subjects breathed comfortably. Results suggest that SSVEP-based BCI may be a promising choice to improve patient-caregiver communication in situations of breathing discomfort when verbal communication is difficult

Domains

Bioengineering
Fichier principal
Vignette du fichier
ssevp_ventilation_final_revised.pdf (419.94 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01207793 , version 1 (01-10-2015)

Identifiers

Cite

Xavier Navarro, Sebastien Campion, Fabrizio de Vico Fallani, Pierre Pouget, Thomas Similowski, et al.. Steady state visual evoked potentials-based patient interface under breathing constraints. Neural Engineering (NER), 2015 7th International IEEE/EMBS Conference, 2015, montpellier, France. ⟨10.1109/NER.2015.7146579⟩. ⟨hal-01207793⟩
100 View
169 Download

Altmetric

Share

Gmail Facebook X LinkedIn More