Our staff editors continue to share brilliant, thoughtful, and meaningful topics and articles in this recommended series.
This week, we would like to share the latest articles about the application of electroencephalogram (EEG) in neurodegenerative diseases.
Title: EEG and sEMG Fusion-Based Precise Detection of Lower Limb Voluntary Movement using CNN-LSTM model
Authors: Xiaodong Zhang, Hanzhe Li, Runlin Dong, Zhufeng Lu, Cunxin Li
Type: ORIGINAL RESEARCH article
The EEG and sEMG fusion have been widely used in detection of human movement intention for human-robot interaction, but the internal relationship of EEG and sEMG signal is not clear so that their fusion still has some shortcomings. A precise fusion method of EEG and sEMG using CNN-LSTM model will be investigated to detect lower limb voluntary movement in this paper. At first, the EEG and sEMG signals processing of each stage was analyzed, so that the response time difference between EEG and sEMG can be estimated to detect lower limb voluntary movement, and it can be calculated by the symbolic transfer entropy. Secondly, the data fusion and feature of EEG and sEMG was both used for obtaining data matrix of model, and a hybrid CNN-LSTM was established for the EEG and sEMG based decoding model of lower limb voluntary movement, so that the estimated value of time difference was about 24 ~ 26 ms, and the calculated value was between 25-45 ms. Finally, the offline experimental results showed that the accuracy of data fusion was significantly higher than feature fusion-based accuracy in 5-fold cross-validation, and the average accuracy of EEG and sEMG data fusion was more than 95%, the improved the average accuracy for eliminating the response time difference between EEG and sEMG was about 0.7±0.26% in data fusion. In the meantime, the online average accuracy of data fusion-based CNN-LSTM was more than 87% in all subjects. The above results demonstrated that the time difference had an influence on the EEG and sEMG fusion to detect lower limb voluntary movement, and the proposed CNN-LSTM model can achieve a high performance. This work provides a stable and reliable basis for human-robot interaction of lower limb exoskeleton.
Access this article: https://doi.org/10.3389/fnins.2022.954387
Title: Enhanced temporal complexity of EEG signals in older individuals with high cognitive functions
Authors: Yuta Iinuma, Sou Nobukawa, Kimiko Mizukami, Megumi Kawaguchi, Teruya Yamanishi, Tetsuya Takahashi
Type: ORIGINAL RESEARCH article
Recent studies suggest that the maintenance of cognitive function in the later life of older people is an essential factor contributing to mental well-being and physical health. Particularly, the risk of depression, sleep disorders, and Alzheimer's disease significantly increases in patients with mild cognitive impairment. To develop early treatment and prevention strategies for cognitive decline, it is necessary to individually identify the current state of cognitive function since the progression of cognitive decline varies among individuals. Therefore, the development of biomarkers that allow easier measurement of cognitive function in older individuals is relevant for hyperaged societies. One of the methods used to estimate cognitive function focuses on the temporal complexity of electroencephalography (EEG) signals. The characteristics of temporal complexity depend on the time scale, which reflects the range of neuron functional interactions. To capture the dynamics, composed of multiple time scales, multiscale entropy (MSE) analysis is effective for comprehensively assessing the neural activity underlying cognitive function in the brain. Thus, we hypothesized that EEG complexity analysis could serve to assess a wide range of cognitive functions in older adults. To validate our hypothesis, we divided older participants into two groups based on their cognitive function test scores: a high cognitive function group and a low cognitive function group, and applied MSE analysis to the measured EEG data of all participants. The results of the repeated-measures analysis of covariance using age and sex as a covariate in the MSE profile showed a significant difference between the high and low cognitive function groups (F=10.18, p=0.003) and the interaction of the group $\times$ electrodes (F=3.93, p=0.002). Subsequently, the results of the post-hoc $t$-test showed high complexity on a slower time scale in the frontal, parietal, and temporal lobes in the high cognitive function group. This high complexity on a slow time scale reflects the activation of long-distance neural interactions among various brain regions to achieve high cognitive functions. This finding could facilitate the development of a tool for diagnosis of cognitive decline in older individuals.
Access this article: https://doi.org/10.3389/fnins.2022.878495
Title: Deep Learning-Based Self-Induced Emotion Recognition Using EEG
Authors: Yerim Ji, Suh-Yeon Dong
Type: ORIGINAL RESEARCH article
Emotion recognition from electroencephalogram (EEG) signals requires accurate and efficient signal processing and feature extraction. Deep learning technology has enabled the automatic extraction of raw EEG signal features that contribute to classifying emotions more accurately. Despite such advances, classification of emotions from EEG signals, especially recorded during recalling specific memories or imagining emotional situations has not yet been investigated. In addition, high-density EEG signal classification using deep neural networks faces challenges, such as high computational complexity, redundant channels, and low accuracy. To address these problems, we evaluate the effects of using a simple channel selection method for classifying self-induced emotions based on deep learning. The experiments demonstrate that selecting key channels based on signal statistics can reduce the computational complexity by 89% without decreasing the classification accuracy. The channel selection method with the highest accuracy was the kurtosis-based method, which achieved accuracies of 79.03% and 79.36% for the valence and arousal scales, respectively. The experimental results show that the proposed framework outperforms conventional methods, even though it uses fewer channels. Our proposed method can be beneficial for the effective use of EEG signals in practical applications.
Access this article: https://doi.org/10.3389/fnins.2022.985709
Title: Neurophysiological and Subjective Analysis of VR Emotion Induction Paradigm
Authors: Ming Li, Junjun Pan, Yang Gao, Yang Shen, Fang Luo, Ju Dai, Aimin Hao, Hong Qin
The ecological validity of emotion-inducing scenarios is essential for emotion research. In contrast to the classical passive induction paradigm, immersive VR fully engages the psychological and physiological components of the subject, which is considered an ecologically valid paradigm for studying emotion. Several studies investigate the emotional responses to different VR tasks or games using subjective scales. However, little research regards VR as an eliciting material, especially when systematically analyzing emotional processes in VR from a neurophysiological perspective. To fill this gap and scientifically evaluate VR's ability to be used as an active method for emotion elicitation, we investigate the dynamic relationship between explicit information (subjective evaluations) and implicit information (objective neurophysiological data). A total of 28 participants are enlisted to watch eight VR videos while their SAM/IPQ scores and EEG data are recorded simultaneously. In ecologically valid scenarios, the subjective results demonstrate that VR has significant advantages for evoking emotion in arousal-valence. This conclusion is backed by our examination of objective neurophysiological evidence that VR videos effectively induce high-arousal emotions. In addition, we obtain features of critical channels and frequency oscillations associated with emotional valence, thereby validating previous research in more lifelike circumstances. In particular, we discover hemispheric asymmetry in the occipital region under high and low emotional arousal, which adds to our understanding of neural features and the dynamics of emotional arousal. As a result, we successfully integrate EEG and VR to demonstrate that VR is more pragmatic for evoking natural feelings and is beneficial for emotional research. Our research has set a precedent for new methodologies of using VR induction paradigms to acquire a more reliable explanation of affective computing.
Access this article: https://doi.org/10.1109/TVCG.2022.3203099
Title: Automatic seizure detection based on Gray Level Co-occurrence Matrix of STFT imaged-EEG
Authors: Haniye Shayeste, Babak Mohammadzadeh Asl
Type: Research article
Epilepsy is a complicated neurological disorder, that features recurrent seizures. During a seizure, the repetitive action potentials lead to a high-frequency burst, and a hyper synchronization happens among the activities of a population of cortical neurons concurrently. In this study, we propose a repetition-based seizure detection method, using Gray Level Co-occurrence Matrix (GLCM) and Electroencephalogram (EEG) signals. In this method, imaged-EEGs are made from the Short Time Fourier Transform (STFT) coefficients of two sequential epochs. The GLCMs of these images are calculated and one feature named diagGLCM is extracted from each co-occurrence matrix. This feature demonstrates the repetition of the STFT coefficient of a frequency band in two sequential epochs, which are used for a classification task between ictal (during a seizure) and inter-ictal (between seizures) time segments. Using one frequency band’s STFT repetition, a patient-specific seizure detection algorithm is developed with an accuracy of 99.56%, a sensitivity of 99.52%, and specificity of 99.62%. This accurate classification shows the efficiency of the extracted feature which can be a biomarker for the ictal repetitive action potentials.
Access this article: https://doi.org/10.1016/j.bspc.2022.104109