Brain-Computer Interface Application of a High-Sensitivity Multichannel fNIRS System: Binary Decision Decoding for Positive and Negative Intentions

Lu Bai; Yao Zhang; Dongyuan Liu; Pengrui Zhang; Feng Gao

doi:10.3788/CJL202249.0507209

Chinese Journal of Lasers, Volume. 49, Issue 5, 0507209(2022)

Brain-Computer Interface Application of a High-Sensitivity Multichannel fNIRS System: Binary Decision Decoding for Positive and Negative Intentions

Lu Bai¹, Yao Zhang¹, Dongyuan Liu¹, Pengrui Zhang¹, and Feng Gao^1,2、*

Author Affiliations

¹College of Precision Instruments and Optoelectronics Engineering, Tianjin University, Tianjin 300072, China

²Tianjin Key Laboratory of Biomedical Detecting Techniques and Instruments, Tianjin 300072, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(7)

Fig. 1. High-sensitivity multi-channel fNIRS system used for data collection. (a) Schematic of fNIRS system; (b) arrangement of optical fiber probes and channels

Download full size

Fig. 2. Experimental paradigm

Download full size

Fig. 3. Changes of hemoglobin concentration in channel 4 for subject one in its No. 2 experiment

Download full size

Fig. 4. Block-averaged hemodynamic response of ΔC_HbO and ΔC_HbR under different answers for subject 8 in its No. 8 experiment

Download full size

Fig. 5. Optical topographies of oxyhemoglobin concentration under different answers. (a) No. 2 experiment for subject 2; (b) No. 3 experiment for subject 3

Download full size

Table 1. Features and channel numbers for classification

View table

Table 1. Features and channel numbers for classification

Subject	Feature		Number
Subject	Based on original light intensity	Based on hemoglobin concentration changes	Based on original light intensity	Based on hemoglobin concentration changes
S1	Kurt	Skew	2	3
S2	Kurt	Skew	3	3
S3	Kurt	Kurt	2	3
S4	Kurt	Kurt	6	5
S5	Skew	Activation	3	3
S6	Kurt	Activation	3	3
S7	Skew	Skew	5	3
S8	Kurt	Activation	5	3
S9	Skew	Skew	3	3
S10	Kurt	Kurt	3	4
Mean			3.5	3.3

Table 2. Classification results of SVM models built with original light intensity and hemoglobin concentration changes data

View table

Table 2. Classification results of SVM models built with original light intensity and hemoglobin concentration changes data

Subject	Specificity /%		Sensitivity /%		Balanced accuracy /%
Subject	Based on original light intensity	Based on hemoglobin concentration changes	Based on original light intensity	Based on hemoglobin concentration changes	Based on original light intensity	Based on hemoglobin concentration changes
S1	67.4	77.2	79.8	66.9	73.6	72.0
S2	65.4	66.4	74.7	75.8	70.0	71.1
S3	71.8	66.5	70.4	74.6	71.2	70.6
S4	67.0	62.2	79.2	84.4	73.1	73.3
S5	52.8	69.6	75.6	80.4	64.0	75.0
S6	73.7	66.1	71.1	78.4	72.5	72.3
S7	63.1	79.5	80.7	70.3	71.8	74.9
S8	70.1	71.4	65.5	75.0	67.7	73.2
S9	65.3	73.4	65.9	73.4	65.7	73.4
S10	91.0	86.8	60.9	64.1	75.9	75.6
Mean	68.8±9.7	71.9±7.5	72.4±6.8	74.3±6.1	70.6±3.7	73.1±1.7

Tools

Get Citation

Copy Citation Text

Lu Bai, Yao Zhang, Dongyuan Liu, Pengrui Zhang, Feng Gao. Brain-Computer Interface Application of a High-Sensitivity Multichannel fNIRS System: Binary Decision Decoding for Positive and Negative Intentions[J]. Chinese Journal of Lasers, 2022, 49(5): 0507209

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Received: Jul. 21, 2021

Accepted: Aug. 27, 2021

Published Online: Mar. 9, 2022

The Author Email: Gao Feng (gaofeng@tju.edu.cn)

DOI:10.3788/CJL202249.0507209

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology