Brain-computer Interface Techniques Research Articles

A Survey of Artificial Intelligence (AI) and Brain Computer Interface Techniques (BCI) for Translating Brain Signals into Text

A Survey of Artificial Intelligence (AI) and Brain Computer Interface Techniques (BCI) for Translating Brain Signals into Text

Emotional brain network decoded by biological spiking neural network.

Emotional disorders are essential manifestations of many neurological and psychiatric diseases. Nowadays, researchers try to explore bi-directional brain-computer interface techniques to help the patients. However, the related functional brain areas and biological markers are still unclear, and the dynamic connection mechanism is also unknown. To find effective regions related to different emotion recognition and intervention, our research focuses on finding emotional EEG brain networks using spiking neural network algorithm with binary coding. We collected EEG data while human participants watched emotional videos (fear, sadness, happiness, and neutrality), and analyzed the dynamic connections between the electrodes and the biological rhythms of different emotions. The analysis has shown that the local high-activation brain network of fear and sadness is mainly in the parietal lobe area. The local high-level brain network of happiness is in the prefrontal-temporal lobe-central area. Furthermore, the α frequency band could effectively represent negative emotions, while the α frequency band could be used as a biological marker of happiness. The decoding accuracy of the three emotions reached 86.36%, 95.18%, and 89.09%, respectively, fully reflecting the excellent emotional decoding performance of the spiking neural network with self- backpropagation. The introduction of the self-backpropagation mechanism effectively improves the performance of the spiking neural network model. Different emotions exhibit distinct EEG networks and neuro-oscillatory-based biological markers. These emotional brain networks and biological markers may provide important hints for brain-computer interface technique exploration to help related brain disease recovery.

Current Perspective of Brain-Computer Interface Technology on Mild Cognitive Impairment

This work find that the current literature review of brain-computer interface (BCI) in the treatment for the people suffering from mild cognitive impairment (MCI) is a relatively unexploited field compared to other application of BCI techniques. Through our literature retrieval strategy, starting from the theoretical basis of BCI in the treatment of MCI and the pathological basis of MCI, this paper explains the advantages and importance of introducing brain-computer interface into the treatment of MCI. Then, starting from the mode of obtaining brain-related information from BCI, the paper describes the detection methods and possible treatment methods of BCI in the field of treating MCI by applying multiple BCI-related methods such as BCI-based EEG method, BCI-based fMRI method, and BCI-based fNIRS method. This paper also discusses how these BCI-based methods can be combined and interact with each other to provide a better detection and treatment for MCI diagnosis and rehabilitation. Finally, the above several forms of BCI are used to describe the possible future development direction and the existing difficulties of these methods, and why these difficulties may prevent researchers from applying these methods to the actual clinical practices.

Flexible In-Ga-Zn-N-O synaptic transistors for ultralow-power neuromorphic computing and EEG-based brain-computer interfaces.

Designing low-power and flexible artificial neural devices with artificial neural networks is a promising avenue for creating brain-computer interfaces (BCIs). Herein, we report the development of flexible In-Ga-Zn-N-O synaptic transistors (FISTs) that can simulate essential and advanced biological neural functions. These FISTs are optimized to achieve ultra-low power consumption under a super-low or even zero channel bias, making them suitable for wearable BCI applications. The effective tunability of synaptic behaviors promotes the realization of associative and non-associative learning, facilitating Covid-19 chest CT edge detection. Importantly, FISTs exhibit high tolerance to long-term exposure under an ambient environment and bending deformation, indicating their suitability for wearable BCI systems. We demonstrate that an array of FISTs can classify vision-evoked EEG signals with up to ∼87.9% and 94.8% recognition accuracy for EMNIST-Digits and MindBigdata, respectively. Thus, FISTs have enormous potential to significantly impact the development of various BCI techniques.

Forecasting Unplanned Purchase Behavior under Buy-One Get-One-Free Promotions Using Functional Near-Infrared Spectroscopy.

It is very important for consumers to recognize their wrong shopping habits such as unplanned purchase behavior (UPB). The traditional methods used for measuring the UPB in qualitative and quantitative studies have some drawbacks because of human perception and memory. We proposed a UPB identification methodology applied with the brain-computer interface technique using a support vector machine (SVM) along with a functional near-infrared spectroscopy (fNIRS). Hemodynamic signals and behavioral data were collected from 33 subjects by performing Task 1 which included the Buy-One-Get-One-Free (BOGOF) and Task 2 which excluded the BOGOF condition. The acquired data were calculated with 6 time-domain features and then classified them using SVM with 10-cross validations. Thereafter, we evaluated whether the results were reliable using the area under the receiver operating characteristic curve (AUC). As a result, we achieved average accuracy greater than 94%, which is reliable because of the AUC values above 0.97. We found that the UPB brain activity was more relevant to Task 1 with the BOGOF condition than with Task 2 in the prefrontal cortex. UPBs were sufficiently derived from self-reported measurement, indicating that the subjects perceived increased impulsivity in the BOGOF condition. Therefore, this study improves the detection and understanding of UPB as a path for a computer-aided detection perspective for rating the severity of UPBs.

Applying Functional Animation to Pictorial Symbols for Supporting P300–Brain–Computer Interface Access to Augmentative and Alternative Communication Devices by Children

Motion techniques, such as highlighting communication items via flip and/or zoom motion, may support BCI-AAC success and comfort for those using P300 brain–computer interface techniques for augmentative and alternative communication device access (BCI-AAC). However, the utilization of motion in P300-BCI-AAC for children is largely unexplored. Therefore, the aims of this investigation are to provide initial insights into the impacts of functional pictorial symbol animation on (a) event-related potentials underlying P300-BCI-AAC performance and (b) children’s user experience. Participants completed two conditions: (1) a static item condition, and (2) a functional animation condition. Following completion, participants completed Likert scale ratings of factors like visual appeal. Animation significantly increased parieto-occipital P200/P300 amplitudes and delayed parieto-occipital N200, P200/P300, and N400 latencies. Further, animation raised ratings of visual appeal to a level approaching significance. Static highlighting significantly increased N400 amplitudes. Findings provide early insights into neural activity underlying P300-BCI-AAC highlighting of pictorial symbols via animation in children and justify further evaluation.

An efficient EEG signal classification technique for Brain–Computer Interface using hybrid Deep Learning

Differently-abled individuals always need support from others for their day-to-day activities. Brain Computer Interface (BCI) has the potential to help those people in carrying out the basic activities. Artificial Intelligence (AI) and Internet of Things (IoT) ecosystem make the BCI technique more useful and efficient. Deep Learning (DL) algorithms can be used for proper identification of body activities from the brain signal. In this paper, we have developed a hybrid DL architecture for the efficient analysis of the EEG signal. To extract the important multi-domain features, we have efficiently selected and used the CNN filters in the proposed architecture. For further improvement of the system performance, the redundant features were removed using the feature reduction technique. In our experiment, we used real-time EEG datasets collected from the physionet biosignal database. The obtained results established that the proposed method outperformed the existing techniques in terms of accuracy, precision, recall, F1-score, and Matthew Correlation Coefficient. Finally, we developed an IoT-based BCI prototype and successfully conducted all the experiments to validate the scheme in a real-time environment.

An Unsupervised Deep-Transfer-Learning-Based Motor Imagery EEG Classification Scheme for Brain-Computer Interface.

Brain–computer interface (BCI) research has attracted worldwide attention and has been rapidly developed. As one well-known non-invasive BCI technique, electroencephalography (EEG) records the brain’s electrical signals from the scalp surface area. However, due to the non-stationary nature of the EEG signal, the distribution of the data collected at different times or from different subjects may be different. These problems affect the performance of the BCI system and limit the scope of its practical application. In this study, an unsupervised deep-transfer-learning-based method was proposed to deal with the current limitations of BCI systems by applying the idea of transfer learning to the classification of motor imagery EEG signals. The Euclidean space data alignment (EA) approach was adopted to align the covariance matrix of source and target domain EEG data in Euclidean space. Then, the common spatial pattern (CSP) was used to extract features from the aligned data matrix, and the deep convolutional neural network (CNN) was applied for EEG classification. The effectiveness of the proposed method has been verified through the experiment results based on public EEG datasets by comparing with the other four methods.

A new training approach for deep learning in EEG biometrics using triplet loss and EMG-driven additive data augmentation

One of the major challenges facing Electroencephalogram (EEG) in biometric systems is the high time-variability of the brainwaves, especially across different sessions. Electrical muscle activity, or Electromyogram (EMG), is considered one of the main noise sources that affect EEG repeatability due to the widely overlapping spectra. In this paper, we introduce a new training approach for deep learning to learn time-permanent and subject-unique embeddings towards an EEG biometric system. The proposed neural network is trained to minimize the triplet loss from EEG frames that are boosted with EMG-driven additive data augmentation. The EMG-like noise is synthetically generated and added to the original EEG batches during training, i.e. online augmentation, to reduce the memory usage specifically in large-scale EEG datasets. This framework is evaluated on a multi-session database that was collected specifically for EEG biometric evaluation under auditory stimulation and relaxation protocols from over 50 users. Adopting this approach showed significantly improved performance over brain-computer interface techniques and conventional deep learning schemes that use cross-entropy loss where a 6–20% improvement in the Correct Recognition Rate (CRR) and 3–7% decrease in the Equal Error Rate (EER) were achieved under cross-session setup. Besides, replicating session nuisance with the proposed framework demonstrated enhanced session-invariant EEG embeddings compared to adversarial learning methods with an average of 7.5% CRR improvements. Finally, our proposed training procedure attained a consistent performance on a longitudinal assessment setup where the time gap between enrollment and testing is almost 1 year. Under this setup, a CRR up to 99.8% and EER as low as 1.24% were achieved over a subset of 12 subjects. The achieved results demonstrate the effectiveness of the proposed training framework in improving the time-permanence of the EEG biometric traits under different protocols and testing scenarios.

Ten-Hour Stable Noninvasive Brain-Computer Interface Realized by Semidry Hydrogel-Based Electrodes.

Noninvasive brain-computer interface (BCI) has been extensively studied from many aspects in the past decade. In order to broaden the practical applications of BCI technique, it is essential to develop electrodes for electroencephalogram (EEG) collection with advanced characteristics such as high conductivity, long-term effectiveness, and biocompatibility. In this study, we developed a silver-nanowire/PVA hydrogel/melamine sponge (AgPHMS) semidry EEG electrode for long-lasting monitoring of EEG signal. Benefiting from the water storage capacity of PVA hydrogel, the electrolyte solution can be continuously released to the scalp-electrode interface during used. The electrolyte solution can infiltrate the stratum corneum and reduce the scalp-electrode impedance to 10 kΩ-15 kΩ. The flexible structure enables the electrode with mechanical stability, increases the wearing comfort, and reduces the scalp-electrode gap to reduce contact impedance. As a result, a long-term BCI application based on measurements of motion-onset visual evoked potentials (mVEPs) shows that the 3-hour BCI accuracy of the new electrode (77% to 100%) is approximately the same as that of conventional electrodes supported by a conductive gel during the first hour. Furthermore, the BCI system based on the new electrode can retain low contact impedance for 10 hours on scalp, which greatly improved the ability of BCI technique.

Multi-source online transfer algorithm based on source domain selection for EEG classification.

The non-stationary nature of electroencephalography (EEG) signals and individual variability makes it challenging to obtain EEG signals from users by utilizing brain-computer interface techniques. Most of the existing transfer learning methods are based on batch learning in offline mode, which cannot adapt well to the changes generated by EEG signals in the online situation. To address this problem, a multi-source online migrating EEG classification algorithm based on source domain selection is proposed in this paper. By utilizing a small number of labeled samples from the target domain, the source domain selection method selects the source domain data similar to the target data from multiple source domains. After training a classifier for each source domain, the proposed method adjusts the weight coefficients of each classifier according to the prediction results to avoid the negative transfer problem. This algorithm was applied to two publicly available motor imagery EEG datasets, namely, BCI Competition Ⅳ Dataset Ⅱa and BNCI Horizon 2020 Dataset 2, and it achieved average accuracies of 79.29 and 70.86%, respectively, which are superior to those of several multi-source online transfer algorithms, confirming the effectiveness of the proposed algorithm.

Machine learning-based feature combination analysis for odor-dependent hemodynamic responses of rat olfactory bulb

Rodents have a well-developed sense of smell and are used to detect explosives, mines, illegal substances, hidden currency, and contraband, but it is impossible to keep their concentration constantly. Therefore, there is an ongoing effort to infer odors detected by animals without behavioral readings with brain-computer interface (BCI) technology. However, the invasive BCI technique has the disadvantage that long-term studies are limited by the immune response and electrode movement. On the other hand, near-infrared spectroscopy (NIRS)-based BCI technology is a non-invasive method that can measure neuronal activity without worrying about the immune response or electrode movement. This study confirmed that the NIRS-based BCI technology can be used as an odor detection and identification from the rat olfactory system. In addition, we tried to present features optimized for machine learning models by extracting six features, such as slopes, peak, variance, mean, kurtosis, and skewness, from the hemodynamic response, and analyzing the importance of individuals or combinations. As a result, the feature with the highest F1-Score was indicated as slopes, and it was investigated that the combination of the features including slopes and mean was the most important for odor inference. On the other hand, the inclusion of other features with a low correlation with slopes had a positive effect on the odor inference, but most of them resulted in insignificant or rather poor performance. The results presented in this paper are expected to serve as a basis for suggesting the development direction of the hemodynamic response-based bionic nose in the future.

Investigating Feature Ranking Methods for Sub-Band and Relative Power Features in Motor Imagery Task Classification.

Interpreting the brain commands is now easier using brain-computer interface (BCI) technologies. Motor imagery (MI) signal detection is one of the BCI applications, where the movements of the hand and feet can be recognized via brain commands that can be further used to handle emergency situations. Design of BCI techniques encountered challenges of BCI illiteracy, poor signal to noise ratio, intersubject variability, complexity, and performance. The automated models designed for emergency should have lesser complexity and higher performance. To deal with the challenges related to the complexity performance tradeoff, the frequency features of brain signal are utilized in this study. Feature matrix is created from the power of brain frequencies, and newly proposed relative power features are used. Analysis of the relative power of alpha sub-band to beta, gamma, and theta sub-band has been done. These proposed relative features are evaluated with the help of different classifiers. For motor imagery classification, the proposed approach resulted in a maximum accuracy of 93.51% compared to other existing approaches. To check the significance of newly added features, feature ranking approaches, namely, mutual information, chi-square, and correlation, are used. The ranking of features shows that the relative power features are significant for MI task classification. The chi-square provides the best tradeoff between accuracy and feature space. We found that the addition of relative power features improves the overall performance. The proposed models could also provide quick response having reduced complexity.

Motor imagery classification in brain-machine interface with machine learning algorithms: Classical approach to multi-layer perceptron model

Motor Imagery classification is a major topic in Brain-Computer Interface (BCI) because of its value for clinical restoration of impaired motor ability. Compared to the classical approaches combined with Machine Learning (ML) algorithms are primarily investigated during the past decade, the number of studies that employ Deep Learning methods on BCI applications is relatively limited. In this study, we aim to provide a comprehensive comparison between traditional classification methods and suggests the significance of Deep Learning-based BCI techniques, especially Multi-Layered Perceptron. In summary, with Electroencephalography (EEG) signals during motor imagery tasks, Support Vector Machine (SVM) showed the shortest training time and prediction speed with comparable accuracy among traditional ML-based algorithms. Notably, our subject-independent generalized MLP model succsfully classified the signals with ≈90% accuracy and half the classification time compared to traditional ML-based models. (1) This result suggests that this standard MLP model can become a bridging wireframe for further advanced subject-independent BCI with sophisticated optimization. (2) This result suggests the possibility of a much accurate and robust generalized BCI (subject independent) if this model integrates sophisticated optimization.

SSVEP-assisted RSVP brain–computer interface paradigm for multi-target classification

Brain–computer Interface (BCI) is actively involved in optimizing the communication medium between the human brain and external devices.Objective. Rapid serial visual presentation (RSVP) is a robust and highly efficient BCI technique in recognizing target objects but suffers from limited target selections. Hybrid BCI systems that combine steady-state visual evoked potential (SSVEP) and RSVP can mitigate this limitation and allow users to operate on multiple targets.Approach. This study proposes a novel hybrid SSVEP-RSVP BCI to improve the performance of classifying the target/non-target objects in a multi-target scenario. In this paradigm, SSVEP stimulation helps in identifying the user’s focus location and RSVP stimuli that elicit event-related potentials differentiate target and non-target objects. Main results. The proposed model achieved an offline accuracy of 81.59% by using 12 electroencephalography (EEG) channels and an online (real-time) accuracy of 78.10% when only four EEG channels are considered. Further, the biomarkers of physiological states are analyzed to assess the cognitive states (mental fatigue and user attention) of the participants based on resting theta and alpha band powers. The results indicate an inverse relationship between the BCI performance and the resting EEG power, validating that the subjects’ performance is affected by physiological states for long-term use of the BCI. Significance. Our findings demonstrate that the combination of SSVEP and RSVP stimuli improves the BCI performance and further enhances the possibility of performing multiple user command tasks, which are inevitable in real-world applications. Additionally, the cognitive state biomarkers discussed imply the need for an efficient and attractive experimental paradigm that reduces the physiological state disparities and provide enhanced BCI performance.

A Virtual Cloud based Brain Connectivity Analysis Using Iot (Internet of Things)

The confidential and secure information of the user’s death storage cannot reach to the particular person or it takes minimum a day or hours to back up all the messages and information. There is lack of time efficiency to reveal all the messages. Here introducing a new concept called BCI (Brain Computer Interface) which is used to connect the human brain and computer. Since it is a BCI technique it consists of both hardware and software. In hardware the major role plays by EEG (Electroenceplogram) sensor, it is to check whether the brain is active or not. On the software side, the cloud is created for storing the information. An individual who want to share the information to his/her favorite or particular person can type the message and save the information in the cloud with his/her mail id and phone number which an individual want to convey. The wearable sensor have to wear like a band, as it may not affect the human brain or human body since it has the low frequency range of 7Hz or below 7Hz. Once the individual is death the stored information in the cloud can sent to that registered mail id and phone number without any delay. The time efficiency is very fast.

Review of pathological index detection and new rehabilitation technique of drug addicts

There are two major research issues with regard to detoxification; one is pathological testing of drug users and the other is rehabilitation methods and techniques. Over the years, domestic and foreign researchers have done a lot of work on pathological changes in the brain and rehabilitation techniques for drug users. This article discusses the research status of these two aspects. At present, the evaluation of brain function in drug addicts is still dominated by a single electroencephalography (EEG), near-infrared spectroscopy (NIRS), or magnetic resonance imaging scan. The multimodal physiological data acquisition method based on EEG–NIRS technique is relatively advantageous for actual physiological data acquisition. The traditional drug rehabilitation method is based on medication and psychological counseling. In recent years, psychological correction (e.g., emotional ventilation, intelligent physical and mental decompression, virtual reality technique and drug addiction suppression system, sports training, and rehabilitation) and physical therapy (transcranial magnetic stimulation) have gradually spread. These rehabilitations focus on comprehensive treatment from the psychological and physical aspects. In recent years, new intervention ideas such as brain–computer interface technique have been continuously proposed. In this review, we have introduced multimodal brain function detection and rehabilitation intervention, which have theoretical and practical significance in drug rehabilitation research.

A multi stage EEG data classification using k-means and feed forward neural network

ObjectiveElectroencephalography based brain-computer interface techniques are widely used these days as they bring human intentions into reality. Various researchers have proposed different approaches to decode EEG signals for applications like spellers, emotion recognition, lie detection, brain games etc. The ability to analyze concealed behavior is very important for legal and security purposes. MethodThis research aims to identify the concealed behavior of an individual. This paper presents a three-stage “Concealed Information Test” using wavelet transform, k-means clustering and multi-layer feed-forward neural network. The test has been developed by analyzing ERP component (P300) of EEG data during a mock crime scene. The wavelet transforms extracts time and frequency information from raw EEG data. K-means clustering clusters the wavelet coefficients into three clusters. As, neural network models the nonlinear time series data viz EEG, hence it has been utilized for classification of clustered data. ResultsA hybrid 3-stage classification approach is proposed by combining the advantages of all above-mentioned approaches. EEG data is recorded for a “Concealed Information Test”, for implementing the proposed framework. ConclusionThe performance of the proposed system is improved from existing approaches, by providing an accuracy of 83.1%.

Development of a Smart Helmet for Strategical BCI Applications.

Conducting electrophysiological measurements from human brain function provides a medium for sending commands and messages to the external world, as known as a brain–computer interface (BCI). In this study, we proposed a smart helmet which integrated the novel hygroscopic sponge electrodes and a combat helmet for BCI applications; with the smart helmet, soldiers can carry out extra tasks according to their intentions, i.e., through BCI techniques. There are several existing BCI methods which are distinct from each other; however, mutual issues exist regarding comfort and user acceptability when utilizing such BCI techniques in practical applications; one of the main challenges is the trade-off between using wet and dry electroencephalographic (EEG) electrodes. Recently, several dry EEG electrodes without the necessity of conductive gel have been developed for EEG data collection. Although the gel was claimed to be unnecessary, high contact impedance and low signal-to-noise ratio of dry EEG electrodes have turned out to be the main limitations. In this study, a smart helmet with novel hygroscopic sponge electrodes is developed and investigated for long-term usage of EEG data collection. The existing electrodes and EEG equipment regarding BCI applications were adopted to examine the proposed electrode. In the impedance test of a variety of electrodes, the sponge electrode showed performance averaging 118 kΩ, which was comparable with the best one among existing dry electrodes, which averaged 123 kΩ. The signals acquired from the sponge electrodes and the classic wet electrodes were analyzed with correlation analysis to study the effectiveness. The results indicated that the signals were similar to each other with an average correlation of 90.03% and 82.56% in two-second and ten-second temporal resolutions, respectively, and 97.18% in frequency responses. Furthermore, by applying the proposed differentiable power algorithm to the system, the average accuracy of 21 subjects can reach 91.11% in the steady-state visually evoked potential (SSVEP)-based BCI application regarding a simulated military mission. To sum up, the smart helmet is capable of assisting the soldiers to execute instructions with SSVEP-based BCI when their hands are not available and is a reliable piece of equipment for strategical applications.

Mental imagery for brain-computer interface control and communication in non-responsive individuals.

People who survive severe brain damage may eventually develop a prolonged consciousness disorder. Others can regain full consciousness but remain unable to speak or move because of the severity of the lesions, as for those with locked-in syndrome (LIS). Brain-computer interface techniques can be useful to disentangle these states by detecting neurophysiological correlates of conscious processing of information to enable communication with these individuals after the diagnosis. The goal of our study was to evaluate with a user-centered design approach the usability of a mental imagery task to detect signs of voluntary information processing and enabling communication in a group of severely disabled individuals. Five individuals with LIS participated in the study. Participants were instructed to imagine hand, arm or feet movements during electroencephalography (EEG) to detect patterns of event-related synchronization/desynchronization associated with each task. After the user-centered design, usability was evaluated (i.e., efficiency, effectiveness and satisfaction). Two participants achieved significant levels of accuracy in 2 different tasks. The associated workload and levels of satisfaction perceived by the users were moderate and were mainly related to the time demand of the task. Results showed lack of effectiveness of the task to detect voluntary brain activity and thus detect consciousness or communicate with non-responsive individuals. The application must be modified to be sufficiently satisfying for the intended end-users and suggestions are made in this regard.