Assessing Time-Resolved fNIRS for Brain-Computer Interface Applications of Mental Communication

Abstract

Brain-computer interfaces (BCIs) are becoming increasingly popular as a tool to improve the quality of life of patients with disabilities. Recently, time-resolved functional near-infrared spectroscopy (TR-fNIRS) based BCIs are gaining traction because of their enhanced depth sensitivity leading to lower signal contamination from the extracerebral layers. This study presents the first account of TR-fNIRS based BCI for "mental communication" on healthy participants. Twenty-one (21) participants were recruited and were repeatedly asked a series of questions where they were instructed to imagine playing tennis for "yes" and to stay relaxed for "no." The change in the mean time-of-flight of photons was used to calculate the change in concentrations of oxy- and deoxyhemoglobin since it provides a good compromise between depth sensitivity and signal-to-noise ratio. Features were extracted from the average oxyhemoglobin signals to classify them as "yes" or "no" responses. Linear-discriminant analysis (LDA) and support vector machine (SVM) classifiers were used to classify the responses using the leave-one-out cross-validation method. The overall accuracies achieved for all participants were 75% and 76%, using LDA and SVM, respectively. The results also reveal that there is no significant difference in accuracy between questions. In addition, physiological parameters [heart rate (HR) and mean arterial pressure (MAP)] were recorded on seven of the 21 participants during motor imagery (MI) and rest to investigate changes in these parameters between conditions. No significant difference in these parameters was found between conditions. These findings suggest that TR-fNIRS could be suitable as a BCI for patients with brain injuries.

Keywords: brain-computer interface; disorders of consciousness; functional near-infrared spectroscopy; motor-imagery; time-resolved measurement.

FIGURE 1

(A) A participant wearing the TR-fNIRS cap with the probes positioned over the SMA and PMC. (B) Study protocol illustrating the rest and response periods. The total time per question was 5:30 min, which consisted of five 30-s answer periods.

FIGURE 2

2D feature space showing the relationship between SS and r for all of the “yes” and “no” responses.

FIGURE 3

Sample time courses of ΔC_HbO2 and ΔC_Hb for one participant and two questions. Each time course was averaged across data from all four channels. The time course on the left was classified as “yes” while the one on the right was classified as “no.” The gray boxes indicate the response periods. The error bars represent the standard error of mean across channels.

FIGURE 4

ΔC_HbO2(red) and ΔC_Hb (blue) for each question averaged across all trials, channels, and participants. Each column represents a different question. The first row (A) shows the signals that were classified as “yes” while the second row (B) shows the signals that were classified as “no.” The gray boxes indicate the response period. The error bars represent the standard error of mean across participants (n = 18).

FIGURE 5

(A) Classification accuracy obtained versus the number of cycles used for classification. The box plot shows the variation in accuracy for all 15 unique combinations of features. The red circles represent the accuracy for the set of features that was selected as optimum (B) Classification accuracy obtained for questions 1–4 using five cycles for classification.

FIGURE 6

ΔC_HbO2 averaged across channels, trials and participants for (A) the “yes” responses and (B) the “no” responses. The solid lines show the signals based on the SVM classifier output while the dashed lines represent the ground truth responses. The error bars represent the standard error of mean across participants (n = 18).