EEG emotion recognition based on cross-frequency granger causality feature extraction and fusion in the left and right hemispheres

Abstract

EEG emotion recognition based on Granger causality (GC) brain networks mainly focus on the EEG signal from the same-frequency bands, however, there are still some causality relationships between EEG signals in the cross-frequency bands. Considering the functional asymmetric of the left and right hemispheres to emotional response, this paper proposes an EEG emotion recognition scheme based on cross-frequency GC feature extraction and fusion in the left and right hemispheres. Firstly, we calculate the GC relationship of EEG signals according to the frequencies and hemispheres, and mainly focus on the causality of the cross-frequency EEG signals in left and right hemispheres. Then, to remove the redundant connections of the GC brain network, an adaptive two-stage decorrelation feature extraction scheme is proposed under the condition of maintaining the best emotion recognition performance. Finally, a multi-GC feature fusion scheme is designed to balance the recognition accuracy and feature number of each GC feature, which comprehensively considers the influence of the recognition accuracy and computational complexity. Experimental results on the DEAP emotion dataset show that the proposed scheme can achieve an average accuracy of 84.91% for four classifications, which improved the classification accuracy by up to 8.43% compared with that of the traditional same-frequency band GC features.

Bibliometric Analysis of Granger Causality Studies

“…Granger causality test revealed that there was connectivity from the right precentral gyrus to the left precentral gyrus and dorsal anterior cingulate cortex, which affected the internet gaming disorder severity. Zhang et al [ 78 ] proposed the cross-frequency Granger causality feature extraction and fusion in both hemispheres for EEG emotion recognition. This proposed Granger causality had higher accuracy than the same-frequency band Granger causality features.…”

Granger-Causality-Based Multi-Frequency Band EEG Graph Feature Extraction and Fusion for Emotion Recognition

“…Then, each 60 s EEG signal was divided into segments with a window length of and an overlap time of . Our previous work [ 34 ] has shown that the best recognition performance can be achieved when the and are 3 s and 1.5 s, respectively. EEG signals of four frequency bands were extracted using STFT.…”

“…The higher the GC value, the stronger the causality between the corresponding EEG signals. On the contrary, smaller GC values indicate weaker causal relationships or connections caused by noise [ 34 ]. Figure 2 c shows that the number of causal values in the original GC matrix of EEG signals at each frequency band is .…”

EEG signal motion recognition based on GSA-SVM

“…Sun R, Cheng AS K, Chan C, et al ( 2015): This study explores the possibility of using electroencephalography to track gaze position, which may help develop virtual eye tracking technology [21] . Chen D, Huang H, Bao X, et al ( 2023): This study proposes an attention recognition method based on EEG signals, which combines time-domain, frequency-domain, and nonlinear dynamic features [22] . Wang P, Cao X, Zhou Y, et al ( 2023): This review article summarizes recent research on decoding neural activity related to limb movement trajectories, which may help develop assistive and rehabilitation strategies for motion impaired users [23] .…”