Determining Optimal Feature-Combination for LDA Classification of Functional Near-Infrared Spectroscopy Signals in Brain-Computer Interface Application

Abstract

In this study, we determine the optimal feature-combination for classification of functional near-infrared spectroscopy (fNIRS) signals with the best accuracies for development of a two-class brain-computer interface (BCI). Using a multi-channel continuous-wave imaging system, mental arithmetic signals are acquired from the prefrontal cortex of seven healthy subjects. After removing physiological noises, six oxygenated and deoxygenated hemoglobin (HbO and HbR) features-mean, slope, variance, peak, skewness and kurtosis-are calculated. All possible 2-and 3-feature combinations of the calculated features are then used to classify mental arithmetic vs. rest using linear discriminant analysis (LDA). It is found that the combinations containing mean and peak values yielded significantly higher (p < 0.05) classification accuracies for both HbO and HbR than did all of the other combinations, across all of the subjects. These results demonstrate the feasibility of achieving high classification accuracies using mean and peak values of HbO and HbR as features for classification of mental arithmetic vs. rest for a two-class BCI.

Integration of Teaching Processes and Learning Assessment in the Prefrontal Cortex during a Video Game Teaching–learning Task

“…Indeed, if the PFC activity were to facilitate the teaching–learning process, its feedback to the teacher during the teaching–learning task might act similarly as a brain–computer interface (BCI) neuro-feedback training. Recently, several studies had reported NIRS-based BCI focusing on the PFC activity (Naseer et al, 2016a,b). In order to evaluate whether the PFC signals during teaching–learning tasks are suitable for NIRS-based BCI training, it is a prerequisite to calculate the classification accuracies of pattern recognition and discrimination in BCI by cross-validation in future studies (Naseer et al, 2014).…”

“…The NIRS system used in this study consisted of six emitters and six detectors, resulting in sixteen channels, each consisting of one emitter-detector pair. In general, an emitter-detector distance around 2.5–3.5 cm is applied, because a distance below 2 cm might result in only superficial layer signal capture, while a distance above 4 cm might result in a weak signal (Gratton et al, 2006; Naseer et al, 2016a). In our study, this distance was set to 3.0 cm.…”

Subject-independent decoding of affective states using functional near-infrared spectroscopy

“…This approach was used in order to test fNIRS based MDB of affective state with the least assumptions possible, to avoid introducing spurious artifact in the signal and to test the feasibility for posterior real-time analysis (Coyle et al, 2007, Sitaram et al, 2007. In comparison, recent studies identified that combining the mean hemoglobin concentration with other temporal and time-frequency features improves the decoding accuracies reaching values close to 90% in within-subject decoding (Tai and Chau, 2009, Naseer et al, 2016a, Naseer et al, 2016b. Therefore, future studies should also evaluate the effect of different feature extraction techniques to the inter-participants MBD of affective states.…”

Cross-Subject EEG Feature Selection for Emotion Recognition Using Transfer Recursive Feature Elimination

“…In particular, the useful information from the continuous measurement of cortical activities can be extracted and selected to indicate the variations of the human cognitive sate. In recent study, Naseer et al (2016) built a novel brain-computer interface system, where the linear discrimination analysis model is used to classify the functional near-infrared spectroscopy (fNIR) features. Based on the optimal feature combination, the optimal recognition rate of two mental states is achieved.…”