Test and re-test reliability of optimal stimulation targets and parameters for personalized neuromodulation.

Protocols have been proposed to optimize neuromodulation targets and parameters to increase treatment efficacies for different neuropsychiatric diseases. However, no study has investigated the temporal effects of optimal neuromodulation targets and parameters simultaneously via exploring the test-retest reliability of the optimal neuromodulation protocols. In this study, we employed a publicly available structural and resting-state functional magnetic resonance imaging (fMRI) dataset to investigate the temporal effects of the optimal neuromodulation targets and parameters inferred from our customized neuromodulation protocol and examine the test-retest reliability over scanning time. 57 healthy young subjects were included in this study. Each subject underwent a repeated structural and resting state fMRI scan in two visits with an interval of 6 weeks between two scanning visits. Brain controllability analysis was performed to determine the optimal neuromodulation targets and optimal control analysis was further applied to calculate the optimal neuromodulation parameters for specific brain states transition. Intra-class correlation (ICC) measure was utilized to examine the test-retest reliability. Our results demonstrated that the optimal neuromodulation targets and parameters had excellent test-retest reliability (both ICCs > 0.80). The test-retest reliability of model fitting accuracies between the actual final state and the simulated final state also showed a good test-retest reliability (ICC > 0.65). Our results indicated the validity of our customized neuromodulation protocol to reliably identify the optimal neuromodulation targets and parameters between visits, which may be reliably extended to optimize the neuromodulation protocols to efficiently treat different neuropsychiatric disorders.

Self-Supervised EEG Emotion Recognition Models Based on CNN.

Emotion plays crucial roles in human life. Recently, emotion classification from electroencephalogram (EEG) signal has attracted attention by researchers due to the rapid development of brain computer interface (BCI) techniques and machine learning algorithms. However, recent studies on emotion classification show resource utilization because they use the fully-supervised learning methods. Therefore, in this study, we applied the self-supervised learning methods to improve the efficiency of resources usage. We employed a self-supervised approach to train deep multi-task convolutional neural network (CNN) for EEG-based emotion classification. First, six signal transformations were performed on unlabeled EEG data to construct the pretext task. Second, a multi-task CNN was used to perform signal transformation recognition on the transformed signals together with the original signals. After the signal transformation recognition network was trained, the convolutional layer network was frozen and the fully connected layer was reconstructed as emotion recognition network. Finally, the EEG data with affective labels were used to train the emotion recognition network to clarify the emotion. In this paper, we conduct extensive experiments from the data scaling perspective using the SEED, DEAP affective dataset. Results showed that the self-supervised learning methods can learn the internal representation of data and save computation time compared to the fully-supervised learning methods. In conclusion, our study suggests that the self-supervised machine learning model can improve the performance of emotion classification compared to the conventional fully supervised model.

Multi-domain feature analysis method of MI-EEG signal based on Sparse Regularity Tensor-Train decomposition.

Tensor analysis can comprehensively retain multidomain characteristics, which has been employed in EEG studies. However, existing EEG tensor has large dimension, making it difficult to extract features. Traditional Tucker decomposition and Canonical Polyadic decomposition(CP) decomposition algorithms have problems of low computational efficiency and weak capability to extract features. To solve the above problems, Tensor-Train(TT) decomposition is adopted to analyze the EEG tensor. Meanwhile, sparse regularization term can then be added to TT decomposition, resulting in a sparse regular TT decomposition (SR-TT). The SR-TT algorithm is proposed in this paper, which has higher accuracy and stronger generalization ability than state-of-the-art decomposition methods. The SR-TT algorithm was verified with BCI competition III and BCI competition IV dataset and achieved 86.38% and 85.36% classification accuracies, respectively. Meanwhile, compared with traditional tensor decomposition (Tucker and CP) method, the computational efficiency of the proposed algorithm was improved by 16.49 and 31.08 times in BCI competition III and 20.72 and 29.45 times more efficient in BCI competition IV. Besides, the method can leverage tensor decomposition to extract spatial features, and the analysis is performed by pairs of brain topography visualizations to show the changes of active brain regions under the task condition. In conclusion, the proposed SR-TT algorithm in the paper provides a novel insight for tensor EEG analysis.

Dual-Encoder VAE-GAN With Spatiotemporal Features for Emotional EEG Data Augmentation.

The current data scarcity problem in EEG-based emotion recognition tasks leads to difficulty in building high-precision models using existing deep learning methods. To tackle this problem, a dual encoder variational autoencoder-generative adversarial network (DEVAE-GAN) incorporating spatiotemporal features is proposed to generate high-quality artificial samples. First, EEG data for different emotions are preprocessed as differential entropy features under five frequency bands and divided into segments with a 5s time window. Secondly, each feature segment is processed in two forms: the temporal morphology data and the spatial morphology data distributed according to the electrode position. Finally, the proposed dual encoder is trained to extract information from these two features, concatenate the two pieces of information as latent variables, and feed them into the decoder to generate artificial samples. To evaluate the effectiveness, a systematic experimental study was conducted in this work on the SEED dataset. First, the original training dataset is augmented with different numbers of generated samples; then, the augmented training datasets are used to train the deep neural network to construct the sentiment model. The results show that the augmented datasets generated by the proposed method have an average accuracy of 97.21% on all subjects, which is a 5% improvement compared to the original dataset, and the similarity between the generated data and the original data distribution is proved. These results demonstrate that our proposed model can effectively learn the distribution of raw data to generate high-quality artificial samples, which can effectively train a high-precision affective model.

A new feature selection approach for driving fatigue EEG detection with a modified machine learning algorithm.

This study aims to identify new electroencephalography (EEG) features for the detection of driving fatigue. The most common EEG feature in driving fatigue detection is the power spectral density (PSD) of five frequency bands, i.e., alpha, beta, gamma, delta, and theta bands. PSD has proved to be useful, however its flaw is that it covers much implicit information of the time domain. In this study we propose a new approach, which combines ensemble empirical mode decomposition (EEMD) and PSD, to explore new EEG features for driving fatigue detection. Through EEMD we get a series of intrinsic mode function (IMF) components, from which we can extract PSD features. We used six features to compare with the proposed features, including the PSD of five frequency bands, PSD of empirical mode decomposition (EMD)-IMF components, PSD, permutation entropy (PE), sample entropy (SE), and fuzzy entropy (FE) of EEMD-IMF components, and common spatial pattern. Feature overlap ratio and multiple machine learning methods were applied to evaluate these feature extraction approaches. The results show that the classification accuracy and overlap ratio of experiments based on IMF's energy spectrum is far superior to other features. Through channel optimization and a comparison of accuracy, we conclude that our new feature selection approach has a better performance based on the modified hierarchical extreme learning machine algorithm with Particle Swarm Optimization (PSO-H-ELM) classifier, which has the highest average accuracy of 97.53%.

Effect of BCI-Controlled Pedaling Training System With Multiple Modalities of Feedback on Motor and Cognitive Function Rehabilitation of Early Subacute Stroke Patients.

Brain-computer interfaces (BCIs) are currently integrated into traditional rehabilitation interventions after stroke. Although BCIs bring many benefits to the rehabilitation process, their effects are limited since many patients cannot concentrate during training. Despite this outcome post-stroke motor-attention dual-task training using BCIs has remained mostly unexplored. This study was a randomized placebo-controlled blinded-endpoint clinical trial to investigate the effects of a BCI-controlled pedaling training system (BCI-PT) on the motor and cognitive function of stroke patients during rehabilitation. A total of 30 early subacute ischemic stroke patients with hemiplegia and cognitive impairment were randomly assigned to the BCI-PT or traditional pedaling training. We used single-channel Fp1 to collect electroencephalography data and analyze the attention index. The BCI-PT system timely provided visual, auditory, and somatosensory feedback to enhance the patient's participation to pedaling based on the real-time attention index. After 24 training sessions, the attention index of the experimental group was significantly higher than that of the control group. The lower limbs motor function (FMA-L) increased by an average of 4.5 points in the BCI-PT group and 2.1 points in the control group (P = 0.022) after treatments. The difference was still significant after adjusting for the baseline indicators ( ß = 2.41 , 95%CI: 0.48-4.34, P = 0.024). We found that BCI-PT significantly improved the patient's lower limb motor function by increasing the patient's participation. (clinicaltrials.gov: NCT04612426).