Multiclass classification of hemodynamic responses for performance improvement of functional near-infrared spectroscopy-based brain-computer interface

Jaeyoung Shin, Jichai Jeong

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

We improved the performance of a functional near-infrared spectroscopy (fNIRS)-based brain-computer interface based on relatively short task duration and multiclass classification. A custom-built eight-channel fNIRS system was used over the motor cortex areas in both hemispheres to measure the hemodynamic responses evoked by four different motor tasks (overt execution of arm lifting and knee extension for both sides) instead of finger tapping. The hemodynamic responses were classified using the naive Bayes classifier. Among the mean, max, slope, variance, and median of the signal amplitude and the time lag of the signal, several signal features are chosen to obtain highest classification accuracy. Ten runs of threefold cross-validation were conducted, which yielded classification accuracies of 87.1%±2.4% to 95.5%±2.4%, 77.5%±1.9% to 92.4%±3.2%, and 73.8%±3.5% to 91.5%±1.4% for the binary, ternary, and quaternary classifications, respectively. Eight seconds of task duration for obtaining sufficient quaternary classification accuracy was suggested. The bit transfer rate per minute (BPM) based on the quaternary classification accuracy was investigated. A BPM can be achieved from 2.81 to 5.40 bits/min.

Original languageEnglish
Pages (from-to)67009
Number of pages1
JournalJournal of Biomedical Optics
Volume19
Issue number6
DOIs
Publication statusPublished - 2014 Jun 1

Fingerprint

hemodynamic responses
Brain-Computer Interfaces
Brain computer interface
Near infrared spectroscopy
Near-Infrared Spectroscopy
Hemodynamics
brain
infrared spectroscopy
Motor Cortex
cortexes
classifiers
hemispheres
Fingers
Knee
Arm
Classifiers
time lag
slopes

ASJC Scopus subject areas

  • Medicine(all)

Cite this

@article{6d0a31e5daac4a96b704269c73445386,
title = "Multiclass classification of hemodynamic responses for performance improvement of functional near-infrared spectroscopy-based brain-computer interface",
abstract = "We improved the performance of a functional near-infrared spectroscopy (fNIRS)-based brain-computer interface based on relatively short task duration and multiclass classification. A custom-built eight-channel fNIRS system was used over the motor cortex areas in both hemispheres to measure the hemodynamic responses evoked by four different motor tasks (overt execution of arm lifting and knee extension for both sides) instead of finger tapping. The hemodynamic responses were classified using the naive Bayes classifier. Among the mean, max, slope, variance, and median of the signal amplitude and the time lag of the signal, several signal features are chosen to obtain highest classification accuracy. Ten runs of threefold cross-validation were conducted, which yielded classification accuracies of 87.1{\%}±2.4{\%} to 95.5{\%}±2.4{\%}, 77.5{\%}±1.9{\%} to 92.4{\%}±3.2{\%}, and 73.8{\%}±3.5{\%} to 91.5{\%}±1.4{\%} for the binary, ternary, and quaternary classifications, respectively. Eight seconds of task duration for obtaining sufficient quaternary classification accuracy was suggested. The bit transfer rate per minute (BPM) based on the quaternary classification accuracy was investigated. A BPM can be achieved from 2.81 to 5.40 bits/min.",
author = "Jaeyoung Shin and Jichai Jeong",
year = "2014",
month = "6",
day = "1",
doi = "10.1117/1.JBO.19.6.067009",
language = "English",
volume = "19",
pages = "67009",
journal = "Journal of Biomedical Optics",
issn = "1083-3668",
publisher = "SPIE",
number = "6",

}

TY - JOUR

T1 - Multiclass classification of hemodynamic responses for performance improvement of functional near-infrared spectroscopy-based brain-computer interface

AU - Shin, Jaeyoung

AU - Jeong, Jichai

PY - 2014/6/1

Y1 - 2014/6/1

N2 - We improved the performance of a functional near-infrared spectroscopy (fNIRS)-based brain-computer interface based on relatively short task duration and multiclass classification. A custom-built eight-channel fNIRS system was used over the motor cortex areas in both hemispheres to measure the hemodynamic responses evoked by four different motor tasks (overt execution of arm lifting and knee extension for both sides) instead of finger tapping. The hemodynamic responses were classified using the naive Bayes classifier. Among the mean, max, slope, variance, and median of the signal amplitude and the time lag of the signal, several signal features are chosen to obtain highest classification accuracy. Ten runs of threefold cross-validation were conducted, which yielded classification accuracies of 87.1%±2.4% to 95.5%±2.4%, 77.5%±1.9% to 92.4%±3.2%, and 73.8%±3.5% to 91.5%±1.4% for the binary, ternary, and quaternary classifications, respectively. Eight seconds of task duration for obtaining sufficient quaternary classification accuracy was suggested. The bit transfer rate per minute (BPM) based on the quaternary classification accuracy was investigated. A BPM can be achieved from 2.81 to 5.40 bits/min.

AB - We improved the performance of a functional near-infrared spectroscopy (fNIRS)-based brain-computer interface based on relatively short task duration and multiclass classification. A custom-built eight-channel fNIRS system was used over the motor cortex areas in both hemispheres to measure the hemodynamic responses evoked by four different motor tasks (overt execution of arm lifting and knee extension for both sides) instead of finger tapping. The hemodynamic responses were classified using the naive Bayes classifier. Among the mean, max, slope, variance, and median of the signal amplitude and the time lag of the signal, several signal features are chosen to obtain highest classification accuracy. Ten runs of threefold cross-validation were conducted, which yielded classification accuracies of 87.1%±2.4% to 95.5%±2.4%, 77.5%±1.9% to 92.4%±3.2%, and 73.8%±3.5% to 91.5%±1.4% for the binary, ternary, and quaternary classifications, respectively. Eight seconds of task duration for obtaining sufficient quaternary classification accuracy was suggested. The bit transfer rate per minute (BPM) based on the quaternary classification accuracy was investigated. A BPM can be achieved from 2.81 to 5.40 bits/min.

UR - http://www.scopus.com/inward/record.url?scp=84922203514&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84922203514&partnerID=8YFLogxK

U2 - 10.1117/1.JBO.19.6.067009

DO - 10.1117/1.JBO.19.6.067009

M3 - Article

VL - 19

SP - 67009

JO - Journal of Biomedical Optics

JF - Journal of Biomedical Optics

SN - 1083-3668

IS - 6

ER -