Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group
  • Journal of Near Infrared Spectroscopy
  • Vol. 20,
  • Issue 1,
  • pp. 107-116
  • (2012)

Investigating the Need for Modelling Temporal Dependencies in a Brain-Computer Interface with Real-Time Feedback Based on near Infrared Spectra

Not Accessible

Your library or personal account may give you access

Abstract

Near infrared (NIR) spectroscopy is an emerging non-invasive brain-computer interface (BCI) modality that measures changes in haemoglobin concentrations in neurocortical tissue. Previous NIR spectroscopy studies have not employed real-time feedback with online classification, a combination which would allow users to alter their mental strategy on the fly. In particular, it is unclear whether or not the temporal dependencies of haemodynamic changes ought to be considered in online classification. To answer this quest ion, this study contrasted online classification of prefrontal haemodynamics using NIR spectra processed using two approaches: an artificial neural network (ANN) that considered instantaneous samples of oxy- and deoxy-haemoglobin concentrations (i.e. ignored temporal dependencies) and a hidden Markov model-based (HMM) classifier which modelled a temporal sequence of concentrations (i.e. embodied temporal dependencies). Both classifiers were implemented for online operation with immediate visual feedback via a monitor showing a vertical bar the height of which was contingent on the classifier's output. Ten subjects participated in two study sessions each, one with each type of classifier. Participants were cued to raise and lower the bar in alternating 20s intervals using mental fast singing and focused breathing, respectively. Only the ANN classifier facilitated online classification rates greater than chance (P = 0.0289). The influence of physiological noise on online classification of prefrontal haemodynamics was deemed to be minimal via offline analysis of concurrently measured respiration and blood pulse. Nine of the ten participants reported using the feedback to alter their activation strategy. Mental fatigue, task repetitiveness and the lack of ambient lighting were identified as factors compromising performance in half the participants. The inferior performance of the HMM classifiers suggests that modelling of the temporal dynamics of haemoglobin concentration changes may not be necessary in an online NIR-BCI. Further study of online NIR-BCIs with instantaneous feedback is warranted.

© 2012 IM Publications LLP

PDF Article
More Like This
CGAN-rIRN: a data-augmented deep learning approach to accurate classification of mental tasks for a fNIRS-based brain-computer interface

Yao Zhang, Dongyuan Liu, Tieni Li, Pengrui Zhang, Zhiyong Li, and Feng Gao
Biomed. Opt. Express 14(6) 2934-2954 (2023)

Detection and classification of three-class initial dips from prefrontal cortex

Amad Zafar and Keum-Shik Hong
Biomed. Opt. Express 8(1) 367-383 (2017)

A new broadband near-infrared spectroscopy system for in-vivo measurements of cerebral cytochrome-c-oxidase changes in neonatal brain injury

Gemma Bale, Subhabrata Mitra, Judith Meek, Nicola Robertson, and Ilias Tachtsidis
Biomed. Opt. Express 5(10) 3450-3466 (2014)

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All Rights Reserved