PSAEEGNet: pyramid squeeze attention mechanism-based CNN for single-trial EEG classification in RSVP task.

Introduction: Accurate classification of single-trial electroencephalogram (EEG) is crucial for EEG-based target image recognition in rapid serial visual presentation (RSVP) tasks. P300 is an important component of a single-trial EEG for RSVP tasks. However, single-trial EEG are usually characterized by low signal-to-noise ratio and limited sample sizes. Methods: Given these challenges, it is necessary to optimize existing convolutional neural networks (CNNs) to improve the performance of P300 classification. The proposed CNN model called PSAEEGNet, integrates standard convolutional layers, pyramid squeeze attention (PSA) modules, and deep convolutional layers. This approach arises the extraction of temporal and spatial features of the P300 to a finer granularity level. Results: Compared with several existing single-trial EEG classification methods for RSVP tasks, the proposed model shows significantly improved performance. The mean true positive rate for PSAEEGNet is 0.7949, and the mean area under the receiver operating characteristic curve (AUC) is 0.9341 (p < 0.05). Discussion: These results suggest that the proposed model effectively extracts features from both temporal and spatial dimensions of P300, leading to a more accurate classification of single-trial EEG during RSVP tasks. Therefore, this model has the potential to significantly enhance the performance of target recognition systems based on EEG, contributing to the advancement and practical implementation of target recognition in this field.

Automatic epileptic seizure detection based on EEG using a moth-flame optimization of one-dimensional convolutional neural networks.

Introduction: Frequent epileptic seizures can cause irreversible damage to the brains of patients. A potential therapeutic approach is to detect epileptic seizures early and provide artificial intervention to the patient. Currently, extracting electroencephalogram (EEG) features to detect epileptic seizures often requires tedious methods or the repeated adjustment of neural network hyperparameters, which can be time- consuming and demanding for researchers. Methods: This study proposes an automatic detection model for an EEG based on moth-flame optimization (MFO) optimized one-dimensional convolutional neural networks (1D-CNN). First, according to the characteristics and need for early epileptic seizure detection, a data augmentation method for dividing an EEG into small samples is proposed. Second, the hyperparameters are tuned based on MFO and trained for an EEG. Finally, the softmax classifier is used to output EEG classification from a small-sample and single channel. Results: The proposed model is evaluated with the Bonn EEG dataset, which verifies the feasibility of EEG classification problems that involve up to five classes, including healthy, preictal, and ictal EEG from various brain regions and individuals. Discussion: Compared with existing advanced optimization algorithms, such as particle swarm optimization, genetic algorithm, and grey wolf optimizer, the superiority of the proposed model is further verified. The proposed model can be implemented into an automatic epileptic seizure detection system to detect seizures in clinical applications.

Surface Electromyography and Electroencephalogram-Based Gait Phase Recognition and Correlations Between Cortical and Locomotor Muscle in the Seven Gait Phases.

The classification of gait phases based on surface electromyography (sEMG) and electroencephalogram (EEG) can be used to the control systems of lower limb exoskeletons for the rehabilitation of patients with lower limb disorders. In this study, the slope sign change (SSC) and mean power frequency (MPF) features of EEG and sEMG were used to recognize the seven gait phases [loading response (LR), mid-stance (MST), terminal stance (TST), pre-swing (PSW), initial swing (ISW), mid-swing (MSW), and terminal swing (TSW)]. Previous researchers have found that the cortex is involved in the regulation of treadmill walking. However, corticomuscular interaction analysis in a high level of gait phase granularity remains lacking in the time-frequency domain, and the feasibility of gait phase recognition based on EEG combined with sEMG is unknown. Therefore, the time-frequency cross mutual information (TFCMI) method was applied to research the theoretical basis of gait control in seven gait phases using beta-band EEG and sEMG data. We firstly found that the feature set comprising SSC of EEG as well as SSC and MPF of sEMG was robust for the recognition of seven gait phases under three different walking speeds. Secondly, the distribution of TFCMI values in eight topographies (eight muscles) was different at PSW and TSW phases. Thirdly, the differences of corticomuscular interaction between LR and MST and between TST and PSW of eight muscles were not significant. These insights enrich previous findings of the authors who have carried out gait phase recognition and provide a theoretical basis for gait recognition based on EEG and sEMG.

Different sEMG and EEG Features Analysis for Gait phase Recognition.

This research focuses on the gait phase recognition using different sEMG and EEG features. Seven healthy volunteers, 23-26 years old, were enrolled in this experiment. Seven phases of gait were divided by three-dimensional trajectory of lower limbs during treadmill walking and classified by Library for Support Vector Machines (LIBSVM). These gait phases include loading response, mid-stance, terminal Stance, pre-swing, initial swing, mid-swing, and terminal swing. Different sEMG and EEG features were assessed in this study. Gait phases of three kinds of walking speed were analyzed. Results showed that the slope sign change (SSC) and mean power frequency (MPF) of sEMG signals and SSC of EEG signals achieved higher accuracy of gait phase recognition than other features, and the accuracy are 95.58% (1.4 km/h), 97.63% (2.0 km/h) and 98.10% (2.6 km/h) respectively. Furthermore, the accuracy of gait phase recognition in the speed of 2.6 km/h is better than other walking speeds.

Efficacy and safety of aerosol inhalation of recombinant human interferon α1b (IFNα1b) injection for noninfluenza viral pneumonia, a multicenter, randomized, double-blind, placebo-controlled trial.

BACKGROUND: To investigate the efficacy and safety of aerosol inhalation of recombinant human interferon α1b (IFNα1b) injection for noninfluenza viral pneumonia. METHODS: One hundred sixty-four patients with noninfluenza viral pneumonia were divided into IFNα1b and control groups. The IFNα1b group received routine treatment + aerosol inhalation of recombinant human IFNα1b injection (50 µg × 2 injections, bid). The control group received routine treatment + IFN analog (two injections, bid). Overall response rate (ORR) of five kinds clinical symptoms. Further outcomes were daily average score and the response rate of each of the symptoms above. RESULTS: A total of 163 patients were included in the full analysis set (FAS) and 151 patients were included in the per-protocol set (PPS). After 7 days of treatment, ORR of clinical symptoms was higher in IFNα1b group than that in control group for both the FAS and PPS. Moreover, after 7 days of treatment, the daily score of three efficacy indexes including expectoration, respiratory rate, and pulmonary rales were improved. The ORRs for expectoration and pulmonary rales were higher in the IFNα1b group than in the control group (P < 0.05). There were no significant differences of the ORRs for coughing, chest pain and respiratory rate between the two groups (P > 0.05). The incidence of adverse events was 6.5% (n = 5) in IFNα1b group and 3.5% (n = 3) in control group (P > 0.05). CONCLUSION: Aerosol inhalation of recombinant human IFNα1b is safe and it can improve the clinical symptoms of noninfluenza viral pneumonia.

An EEG/EMG/EOG-Based Multimodal Human-Machine Interface to Real-Time Control of a Soft Robot Hand.

Brain-computer interface (BCI) technology shows potential for application to motor rehabilitation therapies that use neural plasticity to restore motor function and improve quality of life of stroke survivors. However, it is often difficult for BCI systems to provide the variety of control commands necessary for multi-task real-time control of soft robot naturally. In this study, a novel multimodal human-machine interface system (mHMI) is developed using combinations of electrooculography (EOG), electroencephalography (EEG), and electromyogram (EMG) to generate numerous control instructions. Moreover, we also explore subject acceptance of an affordable wearable soft robot to move basic hand actions during robot-assisted movement. Six healthy subjects separately perform left and right hand motor imagery, looking-left and looking-right eye movements, and different hand gestures in different modes to control a soft robot in a variety of actions. The results indicate that the number of mHMI control instructions is significantly greater than achievable with any individual mode. Furthermore, the mHMI can achieve an average classification accuracy of 93.83% with the average information transfer rate of 47.41 bits/min, which is entirely equivalent to a control speed of 17 actions per minute. The study is expected to construct a more user-friendly mHMI for real-time control of soft robot to help healthy or disabled persons perform basic hand movements in friendly and convenient way.

Decoding Voluntary Movement of Single Hand Based on Analysis of Brain Connectivity by Using EEG Signals.

Research about decoding neurophysiological signals mainly aims to elucidate the details of human motion control from the perspective of neural activity. We performed brain connectivity analysis with EEG to propose a brain functional network (BFN) and used a feature extraction algorithm for decoding the voluntary hand movement of a subject. By analyzing the characteristic parameters obtained from the BFN, we extracted the most important electrode nodes and frequencies for identifying the direction of movement of a hand. The results demonstrated that the most sensitive EEG components were for frequencies delta, theta, and gamma1 from electrodes F4, F8, C3, Cz, C4, CP4, T3, and T4. Finally, we proposed a model for decoding voluntary movement of the right hand by using a hierarchical linear model (HLM). Through a voluntary hand movement experiment in a spiral trajectory, the Poisson coefficient between the measurement trajectory and the decoding trajectory was used as a test standard to compare the HLM with the traditional multiple linear regression model. It was found that the decoding model based on the HLM obtained superior results. This paper contributes a feature extraction method based on brain connectivity analysis that can mine more comprehensive feature information related to a specific mental state of a subject. The decoding model based on the HLM possesses a strong structure for data manipulation that facilitates precise decoding.

Non-invasive decoding of hand movements from electroencephalography based on a hierarchical linear regression model.

A non-invasive brain-computer interface (BCI) is an assistive technology with basic communication and control capabilities that decodes continuous electroencephalography (EEG) signals generated by the human brain and converts them into commands to control external devices naturally. However, the decoding efficiency is limited at present because it is unclear which decoding parameters can be used to effectively improve the overall decoding performance. In this paper, five subjects performed experiments involving self-initiated upper-limb movements during three experimental phases. The decoding method based on a hierarchical linear regression (HLR) model was devised to investigate the influence of decoding efficiency according to the characteristic parameters of brain functional networks. Then the optimal set of channels and most sensitive frequency bands were selected using the p value from a Kruskal-Wallis test in the experimental phases. Eventually, the trajectories of free movement and conical helix movement could be decoded using HLR. The experimental result showed that the Pearson correlation coefficient (R) between the measured and decoded paths is 0.66 with HLR, which was higher than the value of 0.46 obtained with the multiple linear regression model. The HLR from a decoding efficiency perspective holds promise for the development of EEG-based BCI to aid in the restoration of hand movements in post-stroke rehabilitation.

Development of a Novel Motor Imagery Control Technique and Application in a Gaming Environment.

We present a methodology for a hybrid brain-computer interface (BCI) system, with the recognition of motor imagery (MI) based on EEG and blink EOG signals. We tested the BCI system in a 3D Tetris and an analogous 2D game playing environment. To enhance player's BCI control ability, the study focused on feature extraction from EEG and control strategy supporting Game-BCI system operation. We compared the numerical differences between spatial features extracted with common spatial pattern (CSP) and the proposed multifeature extraction. To demonstrate the effectiveness of 3D game environment at enhancing player's event-related desynchronization (ERD) and event-related synchronization (ERS) production ability, we set the 2D Screen Game as the comparison experiment. According to a series of statistical results, the group performing MI in the 3D Tetris environment showed more significant improvements in generating MI-associated ERD/ERS. Analysis results of game-score indicated that the players' scores presented an obvious uptrend in 3D Tetris environment but did not show an obvious downward trend in 2D Screen Game. It suggested that the immersive and rich-control environment for MI would improve the associated mental imagery and enhance MI-based BCI skills.

Benzoyl Peroxide as an Efficient Dopant for Spiro-OMeTAD in Perovskite Solar Cells.

Although organic small molecule spiro-OMeTAD is widely used as a hole-transport material in perovskite solar cells, its limited electric conductivity poses a bottleneck in the efficiency improvement of perovskite solar cells. Here, a low-cost and easy-fabrication technique is developed to enhance the conductivity and hole-extraction ability of spiro-OMeTAD by doping it with commercially available benzoyl peroxide (BPO). The experimental results show that the conductivity increases several orders of magnitude, from 6.2×10-6  S cm-1 for the pristine spiro-OMeTAD to 1.1×10-3  S cm-1 at 5 % BPO doping and to 2.4×10-2  S cm-1 at 15 % BPO doping, which considerably outperform the conductivity of 4.62×10-4  S cm-1 for the currently used oxygen-doped spiro-OMeTAD. The fluorescence spectra suggest that the BPO-doped spiro-OMeTAD-OMeTAD layer is able to efficiently extract holes from CH3 NH3 PbI3 and thus greatly enhances the charge transfer. The BPO-doped spiro-OMeTAD is used in the fabrication of perovskite solar cells, which exhibit enhancement in the power conversion efficiency.

Volitional and Real-Time Control Cursor Based on Eye Movement Decoding Using a Linear Decoding Model.

The aim of this study is to build a linear decoding model that reveals the relationship between the movement information and the EOG (electrooculogram) data to online control a cursor continuously with blinks and eye pursuit movements. First of all, a blink detection method is proposed to reject a voluntary single eye blink or double-blink information from EOG. Then, a linear decoding model of time series is developed to predict the position of gaze, and the model parameters are calibrated by the RLS (Recursive Least Square) algorithm; besides, the assessment of decoding accuracy is assessed through cross-validation procedure. Additionally, the subsection processing, increment control, and online calibration are presented to realize the online control. Finally, the technology is applied to the volitional and online control of a cursor to hit the multiple predefined targets. Experimental results show that the blink detection algorithm performs well with the voluntary blink detection rate over 95%. Through combining the merits of blinks and smooth pursuit movements, the movement information of eyes can be decoded in good conformity with the average Pearson correlation coefficient which is up to 0.9592, and all signal-to-noise ratios are greater than 0. The novel system allows people to successfully and economically control a cursor online with a hit rate of 98%.

Magnetron sputtered zinc oxide nanorods as thickness-insensitive cathode interlayer for perovskite planar-heterojunction solar cells.

Suitable electrode interfacial layers are essential to the high performance of perovskite planar heterojunction solar cells. In this letter, we report magnetron sputtered zinc oxide (ZnO) film as the cathode interlayer for methylammonium lead iodide (CH3NH3PbI3) perovskite solar cell. Scanning electron microscopy and X-ray diffraction analysis demonstrate that the sputtered ZnO films consist of c-axis aligned nanorods. The solar cells based on this ZnO cathode interlayer showed high short circuit current and power conversion efficiency. Besides, the performance of the device is insensitive to the thickness of ZnO cathode interlayer. Considering the high reliability and maturity of sputtering technique both in lab and industry, we believe that the sputtered ZnO films are promising cathode interlayers for perovskite solar cells, especially in large-scale production.

Nonlinear EEG decoding based on a particle filter model.

While the world is stepping into the aging society, rehabilitation robots play a more and more important role in terms of both rehabilitation treatment and nursing of the patients with neurological diseases. Benefiting from the abundant contents of movement information, electroencephalography (EEG) has become a promising information source for rehabilitation robots control. Although the multiple linear regression model was used as the decoding model of EEG signals in some researches, it has been considered that it cannot reflect the nonlinear components of EEG signals. In order to overcome this shortcoming, we propose a nonlinear decoding model, the particle filter model. Two- and three-dimensional decoding experiments were performed to test the validity of this model. In decoding accuracy, the results are comparable to those of the multiple linear regression model and previous EEG studies. In addition, the particle filter model uses less training data and more frequency information than the multiple linear regression model, which shows the potential of nonlinear decoding models. Overall, the findings hold promise for the furtherance of EEG-based rehabilitation robots.

Identification of four novel serum protein biomarkers in sepsis patients encoded by target genes of sepsis-related miRNAs.

The goal of the present study was to identify novel protein biomarkers from the target genes of six serum miRNAs that we identified previously in patients with sepsis. The target genes were predicted by bioinformatics analysis; the levels of the respective proteins in the sera of patients with sepsis were detected by ELISA. ACVR2A (activin A receptor, type IIA), FOXO1 (forkhead box O1), IHH (Indian hedgehog), STK4 (serine/threonine kinase 4) and DUSP3 (dual specificity phosphatase 3) were predicted to be the targets of the six miRNAs, and their encoded proteins were used for biomarker identification. Levels of ACVR2A (P<0.01) and FOXO1 (P<0.01) were significantly different among normal controls, patients with sepsis, patients with severe sepsis and patients with septic shock. Furthermore, levels of ACVR2A (P=0.025), FOXO1 (P<0.001), IHH (P=0.001) and STK4 (P=0.001) were differentially expressed in survivors and non-survivors. DUSP3 levels were not significantly different between any groups. Conjoin analysis of the four differentially expressed proteins showed that the area under the curve of the predictive probabilities was 0.875 [95% CI (confidence interval): 0.785-0.965], which was higher than the SOFA (Sequential Organ Failure Assessment) and APACHE II (Acute Physiology and Chronic Health Evaluation II) scores. When the value of predictive probabilities was 0.449, the four proteins yielded a sensitivity of 68% and a specificity of 91%. Dynamic changes in ACVR2A, FOXO1 and IHH levels showed differential expression between survivors and non-survivors at all time points. On the basis of a combined analysis of the four identified proteins, their predictive value of 28-day mortality of patients with sepsis was better than the SOFA or APACHE II scores.