Single-trial P300 classification algorithm based on centralized multi-person data fusion CNN.

Introduction: Currently, it is still a challenge to detect single-trial P300 from electroencephalography (EEG) signals. In this paper, to address the typical problems faced by existing single-trial P300 classification, such as complex, time-consuming and low accuracy processes, a single-trial P300 classification algorithm based on multiplayer data fusion convolutional neural network (CNN) is proposed to construct a centralized collaborative brain-computer interfaces (cBCI) for fast and highly accurate classification of P300 EEG signals. Methods: In this paper, two multi-person data fusion methods (parallel data fusion and serial data fusion) are used in the data pre-processing stage to fuse multi-person EEG information stimulated by the same task instructions, and then the fused data is fed as input to the CNN for classification. In building the CNN network for single-trial P300 classification, the Conv layer was first used to extract the features of single-trial P300, and then the Maxpooling layer was used to connect the Flatten layer for secondary feature extraction and dimensionality reduction, thereby simplifying the computation. Finally batch normalisation is used to train small batches of data in order to better generalize the network and speed up single-trial P300 signal classification. Results: In this paper, the above new algorithms were tested on the Kaggle dataset and the Brain-Computer Interface (BCI) Competition III dataset, and by analyzing the P300 waveform features and EEG topography and the four standard evaluation metrics, namely Accuracy, Precision, Recall and F1-score,it was demonstrated that the single-trial P300 classification algorithm after two multi-person data fusion CNNs significantly outperformed other classification algorithms. Discussion: The results show that the single-trial P300 classification algorithm after two multi-person data fusion CNNs significantly outperformed the single-person model, and that the single-trial P300 classification algorithm with two multi-person data fusion CNNs involves smaller models, fewer training parameters, higher classification accuracy and improves the overall P300-cBCI classification rate and actual performance more effectively with a small amount of sample information compared to other algorithms.

Hierarchical fusion detection algorithm for sleep spindle detection.

Background: Sleep spindles are a vital sign implying that human beings have entered the second stage of sleep. In addition, they can effectively reflect a person's learning and memory ability, and clinical research has shown that their quantity and density are crucial markers of brain function. The "gold standard" of spindle detection is based on expert experience; however, the detection cost is high, and the detection time is long. Additionally, the accuracy of detection is influenced by subjectivity. Methods: To improve detection accuracy and speed, reduce the cost, and improve efficiency, this paper proposes a layered spindle detection algorithm. The first layer used the Morlet wavelet and RMS method to detect spindles, and the second layer employed an improved k-means algorithm to improve spindle detection efficiency. The fusion algorithm was compared with other spindle detection algorithms to prove its effectiveness. Results: The hierarchical fusion spindle detection algorithm showed good performance stability, and the fluctuation range of detection accuracy was minimal. The average value of precision was 91.6%, at least five percentage points higher than other methods. The average value of recall could reach 89.1%, and the average value of specificity was close to 95%. The mean values of accuracy and F1-score in the subject sample data were 90.4 and 90.3%, respectively. Compared with other methods, the method proposed in this paper achieved significant improvement in terms of precision, recall, specificity, accuracy, and F1-score. Conclusion: A spindle detection method with high steady-state accuracy and fast detection speed is proposed, which combines the Morlet wavelet with window RMS and an improved k-means algorithm. This method provides a powerful tool for the automatic detection of spindles and improves the efficiency of spindle detection. Through simulation experiments, the sampled data were analyzed and verified to prove the feasibility and effectiveness of this method.

Attention-based 3D convolutional recurrent neural network model for multimodal emotion recognition.

Introduction: Multimodal emotion recognition has become a hot topic in human-computer interaction and intelligent healthcare fields. However, combining information from different human different modalities for emotion computation is still challenging. Methods: In this paper, we propose a three-dimensional convolutional recurrent neural network model (referred to as 3FACRNN network) based on multimodal fusion and attention mechanism. The 3FACRNN network model consists of a visual network and an EEG network. The visual network is composed of a cascaded convolutional neural network-time convolutional network (CNN-TCN). In the EEG network, the 3D feature building module was added to integrate band information, spatial information and temporal information of the EEG signal, and the band attention and self-attention modules were added to the convolutional recurrent neural network (CRNN). The former explores the effect of different frequency bands on network recognition performance, while the latter is to obtain the intrinsic similarity of different EEG samples. Results: To investigate the effect of different frequency bands on the experiment, we obtained the average attention mask for all subjects in different frequency bands. The distribution of the attention masks across the different frequency bands suggests that signals more relevant to human emotions may be active in the high frequency bands γ (31-50 Hz). Finally, we try to use the multi-task loss function Lc to force the approximation of the intermediate feature vectors of the visual and EEG modalities, with the aim of using the knowledge of the visual modalities to improve the performance of the EEG network model. The mean recognition accuracy and standard deviation of the proposed method on the two multimodal sentiment datasets DEAP and MAHNOB-HCI (arousal, valence) were 96.75 ± 1.75, 96.86 ± 1.33; 97.55 ± 1.51, 98.37 ± 1.07, better than those of the state-of-the-art multimodal recognition approaches. Discussion: The experimental results show that starting from the multimodal information, the facial video frames and electroencephalogram (EEG) signals of the subjects are used as inputs to the emotion recognition network, which can enhance the stability of the emotion network and improve the recognition accuracy of the emotion network. In addition, in future work, we will try to utilize sparse matrix methods and deep convolutional networks to improve the performance of multimodal emotion networks.

Corrigendum: Multi-person feature fusion transfer learning-based convolutional neural network for SSVEP-based collaborative BCI.

[This corrects the article DOI: 10.3389/fnins.2022.971039.].

Multi-person feature fusion transfer learning-based convolutional neural network for SSVEP-based collaborative BCI.

Objective: The conventional single-person brain-computer interface (BCI) systems have some intrinsic deficiencies such as low signal-to-noise ratio, distinct individual differences, and volatile experimental effect. To solve these problems, a centralized steady-state visually evoked potential collaborative BCI system (SSVEP-cBCI), which characterizes multi-person electroencephalography (EEG) feature fusion was constructed in this paper. Furthermore, three different feature fusion methods compatible with this new system were developed and applied to EEG classification, and a comparative analysis of their classification accuracy was performed with transfer learning-based convolutional neural network (TL-CNN) approach. Approach: An EEG-based SSVEP-cBCI system was set up to merge different individuals' EEG features stimulated by the instructions for the same task, and three feature fusion methods were adopted, namely parallel connection, serial connection, and multi-person averaging. The fused features were then input into CNN for classification. Additionally, transfer learning (TL) was applied first to a Tsinghua University (THU) benchmark dataset, and then to a collected dataset, so as to meet the CNN training requirement with a much smaller size of collected dataset and increase the classification accuracy. Ten subjects were recruited for data collection, and both datasets were used to gauge the three fusion algorithms' performance. Main results: The results predicted by TL-CNN approach in single-person mode and in multi-person mode with the three feature fusion methods were compared. The experimental results show that each multi-person mode is superior to single-person mode. Within the 3 s time window, the classification accuracy of the single-person CNN is only 90.6%, while the same measure of the two-person parallel connection fusion method can reach 96.6%, achieving better classification effect. Significance: The results show that the three multi-person feature fusion methods and the deep learning classification algorithm based on TL-CNN can effectively improve the SSVEP-cBCI classification performance. The feature fusion method of multi -person parallel feature connection achieves better classification results. Different feature fusion methods can be selected in different application scenarios to further optimize cBCI.

Molecularly imprinted Ni-polyacrylamide-based electrochemical sensor for the simultaneous detection of dopamine and adenine.

Molecularly imprinted polymer (MIP) membranes prepared in situ present several advantages: they maintain the original morphology, adhere strongly to the collector, and exhibit a controllable structure. In this study, a Ni-polyacrylamide (PAM)-MIP matrix was fabricated in situ on glassy carbon via the one-step electro-polymerization of AM monomers in the presence of Ni and template molecules. Ni2+ ions were introduced as oxidants to promote AM polymerization and bulking agents to fabricate a three-dimensional porous PAM-MIP matrix. The Ni-PAM-based MIP sensor exhibited a quantitative dual response toward dopamine (DA) and adenine (Ade) in the pH range of 5.0-9.0. The linear concentration range changed depending on the pH environment, and the concentrations of DA and Ade ranged from 0.6 to 200 µM and from 0.4 to 300 µM, respectively. The ranges of detection limits (S/N = 3) were 0.12-0.37 µM for DA and 0.15-0.36 µM for Ade. In addition, the dual-MIP sensor exhibited high reliability in the detection of DA and Ade in human serum owing to its excellent anti-interference ability and long-term stability. The technique developed in this study is expected to facilitate the construction of multi-target response electrochemical biosensors and the reliable determination of small molecules with high selectivity and stability.

Efficacy Evaluation of Neurofeedback-Based Anxiety Relief.

Anxiety disorder is a mental illness that involves extreme fear or worry, which can alter the balance of chemicals in the brain. This change and evaluation of anxiety state are accompanied by a comprehensive treatment procedure. It is well-known that the treatment of anxiety is chiefly based on psychotherapy and drug therapy, and there is no objective standard evaluation. In this paper, the proposed method focuses on examining neural changes to explore the effect of mindfulness regulation in accordance with neurofeedback in patients with anxiety. We designed a closed neurofeedback experiment that includes three stages to adjust the psychological state of the subjects. A total of 34 subjects, 17 with anxiety disorder and 17 healthy, participated in this experiment. Through the three stages of the experiment, electroencephalography (EEG) resting state signal and mindfulness-based EEG signal were recorded. Power spectral density was selected as the evaluation index through the regulation of neurofeedback mindfulness, and repeated analysis of variance (ANOVA) method was used for statistical analysis. The findings of this study reveal that the proposed method has a positive effect on both types of subjects. After mindfulness adjustment, the power map exhibited an upward trend. The increase in the average power of gamma wave indicates the relief of anxiety. The enhancement of the wave power represents an improvement in the subjects' mindfulness ability. At the same time, the results of ANOVA showed that P < 0.05, i.e., the difference was significant. From the aspect of neurophysiological signals, we objectively evaluated the ability of our experiment to relieve anxiety. The neurofeedback mindfulness regulation can effect on the brain activity pattern of anxiety disorder patients.

Au@SnO2-vertical graphene-based microneedle sensor for in-situ determination of abscisic acid in plants.

For developing electrochemical plant sensors, in-situ detection of hormone levels in living plants is worth attempting. A microneedle array sensor based on Au@SnO2-vertical graphene (VG)/Ta microelectrodes was constructed for analyzing abscisic acid (ABA) in plants. Graphene was vertically grown on Ta wires with a diameter of 0.6 mm by direct current arc plasma jet chemical vapor deposition with SnO2 as the Au catalyst carrier. These VG nanosheets were embedded with core-shell Au@SnO2 nanoparticles, and the formation mechanism of the sensing layer was investigated. Three Au@SnO2-VG microelectrodes, one Ti wire, and one Pt wire were packed into a microneedle array sensor with a three-electrode system. ABA was then quantitatively detected by direct electrocatalytic oxidation, which involves the synergistic catalytic effects of the abundant catalytic active sites of the Au@SnO2 nanoparticles and the excellent conductivity of the VG nanosheets. The microneedle array sensor responds to ABA in the pH range 4-7, the response concentration range was 0.012 (or 0.024)-495.2 µM, and the detection limit varied between 0.002 and 0.005 µM. The small size, wide pH range, low detection limit, and wide linear concentration range allow the microneedle array sensor to be inserted into plants for in-situ detection of ABA.

EEG-Based Anxious States Classification Using Affective BCI-Based Closed Neurofeedback System.

PURPOSE: Anxiety disorder is one of the psychiatric disorders that involves extreme fear or worry, which can change the balance of chemicals in the brain. To the best of our knowledge, the evaluation of anxiety state is still based on some subjective questionnaires and there is no objective standard assessment yet. Unlike other methods, our approach focuses on study the neural changes to identify and classify the anxiety state using electroencephalography (EEG) signals. METHODS: We designed a closed neurofeedback experiment that contains three experimental stages to adjust subjects' mental state. The EEG resting state signal was recorded from thirty-four subjects in the first and third stages while EEG-based mindfulness recording was recorded in the second stage. At the end of each stage, the subjects were asked to fill a Visual Analogue Scale (VAS). According to their VAS score, the subjects were classified into three groups: non-anxiety, moderate or severe anxiety groups. RESULTS: After processing the EEG data of each group, support vector machine (SVM) classifiers were able to classify and identify two mental states (non-anxiety and anxiety) using the Power Spectral Density (PSD) as patterns. The highest classification accuracies using Gaussian kernel function and polynomial kernel function are 92.48 ±  1.20% and 88.60  ±  1.32%, respectively. The highest average of the classification accuracies for healthy subjects is 95.31 ±  1.97% and for anxiety subjects is 87.18 ±  3.51%. CONCLUSIONS: The results suggest that our proposed EEG neurofeedback-based classification approach is efficient for developing affective BCI system for detection and evaluation of anxiety disorder states.

Polydopamine/graphene/MnO2 composite-based electrochemical sensor for in situ determination of free tryptophan in plants.

The in vivo detection of small active molecules in plant tissues is essential for the development of precision agriculture. Tryptophan (Trp) is an important precursor material for auxin biosynthesis in plants, and the detection of Trp levels in plants is critical for regulating the plant growth process. In this study, an electrochemical plant sensor was fabricated by electrochemically depositing a polydopamine (PDA)/reduced graphene oxide (RGO)-MnO2 nanocomposite onto a glassy carbon electrode (GCE). PDA/RGO-MnO2/GCE exhibited high electrocatalytic activity for the oxidation of Trp owing to the combined selectivity of PDA and catalytic activity of RGO-MnO2. To address the pH variability of plants, a reliable Trp detection program was proposed for selecting an appropriate quantitative detection model for the pH of the plant or plant tissue of interest. Therefore, a series of linear regression curves was constructed in the pH range of 4.0-7.0 using the PDA/RGO-MnO2/GCE-based sensor. In this pH range, the linear detection range of Trp was 1-300 µM, the sensitivity was 0.39-1.66 µA µM-1, and the detection limit was 0.22-0.39 µM. Moreover, the practical applicability of the PDA/RGO-MnO2/GCE-based sensor was successfully demonstrated by determining Trp in tomato fruit and juice. This sensor stably and reliably detected Trp levels in tomatoes in vitro and in vivo, demonstrating the feasibility of this research strategy for the development of electrochemical sensors for measurements in various plant tissues.

Motor Imagination of Lower Limb Movements at Different Frequencies.

Motor imagination (MI) is the mental process of only imagining an action without an actual movement. Research on MI has made significant progress in feature information detection and machine learning decoding algorithms, but there are still problems, such as a low overall recognition rate and large differences in individual execution effects, which make the development of MI run into a bottleneck. Aiming at solving this bottleneck problem, the current study optimized the quality of the MI original signal by "enhancing the difficulty of imagination tasks," conducted the qualitative and quantitative analyses of EEG rhythm characteristics, and used quantitative indicators, such as ERD mean value and recognition rate. Research on the comparative analysis of the lower limb MI of different tasks, namely, high-frequency motor imagination (HFMI) and low-frequency motor imagination (LFMI), was conducted. The results validate the following: the average ERD of HFMI (-1.827) is less than that of LFMI (-1.3487) in the alpha band, so did (-3.4756 < -2.2891) in the beta band. In the alpha and beta characteristic frequency bands, the average ERD of HFMI is smaller than that of LFMI, and the ERD values of the two are significantly different (p=0.0074 < 0.01; r = 0.945). The ERD intensity STD values of HFMI are less than those of LFMI. which suggests that the ERD intensity individual difference among the subjects is smaller in the HFMI mode than in the LFMI mode. The average recognition rate of HFMI is higher than that of LFMI (87.84% > 76.46%), and the recognition rate of the two modes is significantly different (p=0.0034 < 0.01; r = 0.429). In summary, this research optimizes the quality of MI brain signal sources by enhancing the difficulty of imagination tasks, achieving the purpose of improving the overall recognition rate of the lower limb MI of the participants and reducing the differences of individual execution effects and signal quality among the subjects.

Sandwich-type electrochemical immunosensor for carcinoembryonic antigen detection based on the cooperation of a gold-vertical graphene electrode and gold@silica-methylene blue.

In this study, a sandwich-type electrochemical (EC) immunosensor was proposed to detect a carcinoembryonic antigen (CEA) based on Au-graphene and Au@SiO2-methylene blue (MB). The Au nanoparticles (NPs)-vertical graphene (VG) electrode efficiently amplifies the response signal by immobilizing a large amount of the coating antibody (Ab) and is characterized by excellent electrocatalytic activity. The MB nanodot-loaded Au@SiO2 carriers with core-shell nanostructure and detection Ab were used to construct the Ab-Au@SiO2-MB label, which improved the sensitivity due to the high EC signal of MB nanodots and the high labeling effect between the detection Ab and MB probe. A novel double-Ab sandwich strategy was developed to further improve the sensitivity and stability based on the same specificity of the coating and detection Abs for the recognition of CEA. Under optimal conditions, the developed EC sensor exhibited a wide linear range from 1 fg mL-1 to 100 ng mL-1, with an ultralow detection limit of 0.8 fg mL-1 (S/N = 3). The feasibility in the clinical application of the EC sensor was verified by the in vitro detection of CEA in human serum.

Fabrication of chitosan/Au-TiO2 nanotube-based dry electrodes for electroencephalography recording.

In this paper, we describe a method for fabricating dry electrodes for use in recording electroencephalograms (EEGs), which are based on the use of chitosan (Ch), gold (Au) particles, and titanium dioxide (TiO2) nanotube arrays deposited on titanium (Ti) thin sheets. The samples were characterized by scanning electron microscopy, X-ray diffraction, electrochemical impedance spectroscopy, and EEG signal collection. The TiO2 nanotube arrays were grown on the Ti thin sheet by an electrochemical anodic oxidation method. The Au particles were deposited on the bottom and surface layers of the TiO2 nanotube array using an electrochemistry-based multi-potential step technology. The fabricated dry Ch/Au-TiO2 electrodes have an efficient conversion interface for ion current/electron current, a high biocompatible contact surface, and a fast electron transfer channel. To confirm that the Ch/Au-TiO2 layer can be used in dry EEG electrodes, the impedance spectra of the electrodes in solution and skin were analyzed. The mean impedance values for skin were found to be approximately 169±33.0kΩ at 2.15Hz and 67.4±8.9kΩ at 100Hz. In addition, EEG signals from the forehead and sites with hair were collected using both the dry Ch/Au-TiO2 electrode and a wet Ag/AgCl electrode for comparison purposes. It was found that high quality EEG signal recordings could be obtained using the dry electrodes. The fact that electrolytes are not required means that the electrodes are suitable for use in long-term bio-potential testing.

Visual attention recognition based on nonlinear dynamical parameters of EEG.

Varieties of neurophysiological measures have been utilized in visual attention studies. The linear parameters like power spectrum are the most commonly used features in the existing studies. In this paper, however, nonlinear parameters including approximate entropy, sample entropy and multiscale entropy were tested. All subjects were instructed to perform tasks with three different attention levels (i.e. attention, no attention and rest) in two experiments. Nonlinear features were extracted from the EEG signals. Then, statistical analyses and classification with support vector machine (SVM) were performed. A comparison between the classification results based on the linear feature / and the sample entropy was performed for further analysis. The results suggest that sample entropy stands out in the dynamical parameters with the accuracies of 76.19% and 85.24% in recognition of three levels of attention for the two experiments respectively. And the further comparison shows that the sample entropy performs even better than the / power ratio. It is suggested that nonlinear dynamical parameters may be indispensable for a robust attention recognition system.

[Research progress on application of brain-computer-interface in mobile peripheral control].

Brain computer interface (BCI) is an information channel independent of routine brain output ways such as peripheral nerves and muscle organization. As a special human-computer interface mode, it provides a direct communication pathway between the brain and external devices so as to exert control over those devices by ways other than primitive human communication. Controlling over mobile peripheral devices such as intelligent wheelchairs or nursing robots is a very important application of BCI technology in the future. This paper describes the newest progress of the above mentioned technology, analyzes and compares key techniques involved, and forecasts future development in this field.