A review of classification algorithms for EEG-based brain–computer interfaces: a 10 year update

Motor imagery of finger movements: Effects on cortical and muscle activities

PURPOSE

The purpose of the current study was to explore the immediate effect of motor imagery (MI) involving finger movement of a given limb on cortical response and muscle activity in healthy subjects.

METHODS

Twenty healthy right-handed adults (7 females and 13 males) with a mean + SD age of 22.05 + 6.08 years participated in the study. The beta-band event-related desynchronization (ERD) at the sensorimotor cortex and muscle activity during finger movement tasks using either the index, middle, or thumb digits on the non-dominant left hand were compared before and after an MI training session. Subjects underwent a pre-MI, MI training, and finally a post-MI session where they either performed or imagined performing a button-pushing action 50 times per session with each of the three digits.

RESULTS

The ERD power in the beta frequency band was lower in pre-MI compared to post-MI and was significantly different between the pre- and post-MI sessions for both the index and middle fingers, but not the thumb. A significant decrease was seen in the mean muscle activity during post-MI compared to pre-MI for all the digits except the thumb.

CONCLUSIONS

The results from the current study suggest that complex MI can result in motor learning and improvement in motor performance, thereby requiring less effort during motion.

GRU-powered sleep stage classification with permutation-based EEG channel selection

We present a new approach to classifying the sleep stage that incorporates a computationally inexpensive method based on permutations for channel selection and takes advantage of deep learning power, specifically the gated recurrent unit (GRU) model, along with other deep learning methods. By systematically permuting the electroencephalographic (EEG) channels, different combinations of EEG channels are evaluated to identify the most informative subset for the classification of the 5-class sleep stage. For analysis, we used an EEG dataset that was collected at the International Institute for Integrative Sleep Medicine (WPI-IIIS) at the University of Tsukuba in Japan. The results of these explorations provide many new insights such as the (1) drastic decrease in performance when channels are fewer than 3, (2) 3-random channels selected by permutation provide the same or better prediction than the 3 channels recommended by the American Academy of Sleep Medicine (AASM), (3) N1 class suffers the most in prediction accuracy as the channels drop from 128 to 3 random or 3 AASM, and (4) no single channel provides acceptable levels of accuracy in the prediction of 5 classes. The results obtained show the GRU's ability to retain essential temporal information from EEG data, which allows capturing the underlying patterns associated with each sleep stage effectively. Using permutation-based channel selection, we enhance or at least maintain as high model efficiency as when using high-density EEG, incorporating only the most informative EEG channels.

Single-trial movement intention detection estimation in patients with Parkinson's disease: a movement-related cortical potential study

Objectives. Parkinson patients often suffer from motor impairments such as tremor and freezing of movement that can be difficult to treat. To unfreeze movement, it has been suggested to provide sensory stimuli. To avoid constant stimulation, episodes with freezing of movement needs to be detected which is a challenge. This can potentially be obtained using a brain-computer interface (BCI) based on movement-related cortical potentials (MRCPs) that are observed in association with the intention to move. The objective in this study was to detect MRCPs from single-trial EEG.Approach. Nine Parkinson patients executed 100 wrist movements and 100 ankle movements while continuous EEG and EMG were recorded. The experiment was repeated in two sessions on separate days. Using temporal, spectral and template matching features, a random forest (RF), linear discriminant analysis, and k-nearest neighbours (kNN) classifier were constructed in offline analysis to discriminate between epochs containing movement-related or idle brain activity to provide an estimation of the performance of a BCI. Three classification scenarios were tested: 1) within-session (using training and testing data from the same session and participant), between-session (using data from the same participant from session one for training and session two for testing), and across-participant (using data from all participants except one for training and testing on the remaining participant).Main results. The within-session classification scenario was associated with the highest classification accuracies which were in the range of 88%-89% with a similar performance across sessions. The performance dropped to 69%-75% and 70%-75% for the between-session and across-participant classification scenario, respectively. The highest classification accuracies were obtained for the RF and kNN classifiers.Significance. The results indicate that it is possible to detect movement intentions in individuals with Parkinson's disease such that they can operate a BCI which may control the delivery of sensory stimuli to unfreeze movement.

Intentional binding for noninvasive BCI control

Objective. Noninvasive brain-computer interfaces (BCIs) allow to interact with the external environment by naturally bypassing the musculoskeletal system. Making BCIs efficient and accurate is paramount to improve the reliability of real-life and clinical applications, from open-loop device control to closed-loop neurorehabilitation.Approach. By promoting sense of agency and embodiment, realistic setups including multimodal channels of communication, such as eye-gaze, and robotic prostheses aim to improve BCI performance. However, how the mental imagery command should be integrated in those hybrid systems so as to ensure the best interaction is still poorly understood. To address this question, we performed a hybrid EEG-based BCI training involving healthy volunteers enrolled in a reach-and-grasp action operated by a robotic arm.Main results. Showed that the hand grasping motor imagery timing significantly affects the BCI accuracy evolution as well as the spatiotemporal brain dynamics. Larger accuracy improvement was obtained when motor imagery is performed just after the robot reaching, as compared to before or during the movement. The proximity with the subsequent robot grasping favored intentional binding, led to stronger motor-related brain activity, and primed the ability of sensorimotor areas to integrate information from regions implicated in higher-order cognitive functions.Significance. Taken together, these findings provided fresh evidence about the effects of intentional binding on human behavior and cortical network dynamics that can be exploited to design a new generation of efficient brain-machine interfaces.

Correcting for ERP latency jitter improves gaze-independent BCI decoding

Objective.Patients suffering from heavy paralysis or Locked-in-Syndrome can regain communication using a Brain-Computer Interface (BCI). Visual event-related potential (ERP) based BCI paradigms exploit visuospatial attention (VSA) to targets laid out on a screen. However, performance drops if the user does not direct their eye gaze at the intended target, harming the utility of this class of BCIs for patients suffering from eye motor deficits. We aim to create an ERP decoder that is less dependent on eye gaze.Approach.ERP component latency jitter plays a role in covert visuospatial attention (VSA) decoding. We introduce a novel decoder which compensates for these latency effects, termed Woody Classifier-based Latency Estimation (WCBLE). We carried out a BCI experiment recording ERP data in overt and covert visuospatial attention (VSA), and introduce a novel special case of covert VSA termed split VSA, simulating the experience of patients with severely impaired eye motor control. We evaluate WCBLE on this dataset and the BNCI2014-009 dataset, within and across VSA conditions to study the dependency on eye gaze and the variation thereof during the experiment.Main results.WCBLE outperforms state-of-the-art methods in the VSA conditions of interest in gaze-independent decoding, without reducing overt VSA performance. Results from across-condition evaluation show that WCBLE is more robust to varying VSA conditions throughout a BCI operation session.Significance. Together, these results point towards a pathway to achieving gaze independence through suited ERP decoding. Our proposed gaze-independent solution enhances decoding performance in those cases where performing overt VSA is not possible.

Temporal-spatial cross attention network for recognizing imagined characters

Previous research has primarily employed deep learning models such as Convolutional Neural Networks (CNNs), and Recurrent Neural Networks (RNNs) for decoding imagined character signals. These approaches have treated the temporal and spatial features of the signals in a sequential, parallel, or single-feature manner. However, there has been limited research on the cross-relationships between temporal and spatial features, despite the inherent association between channels and sampling points in Brain-Computer Interface (BCI) signal acquisition, which holds significant information about brain activity. To address the limited research on the relationships between temporal and spatial features, we proposed a Temporal-Spatial Cross-Attention Network model, named TSCA-Net. The TSCA-Net is comprised of four modules: the Temporal Feature (TF), the Spatial Feature (SF), the Temporal-Spatial Cross (TSCross), and the Classifier. The TF combines LSTM and Transformer to extract temporal features from BCI signals, while the SF captures spatial features. The TSCross is introduced to learn the correlations between the temporal and spatial features. The Classifier predicts the label of BCI data based on its characteristics. We validated the TSCA-Net model using publicly available datasets of handwritten characters, which recorded the spiking activity from two micro-electrode arrays (MEAs). The results showed that our proposed TSCA-Net outperformed other comparison models (EEG-Net, EEG-TCNet, S3T, GRU, LSTM, R-Transformer, and ViT) in terms of accuracy, precision, recall, and F1 score, achieving 92.66 % , 92.77 % , 92.70 % , and 92.58 % , respectively. The TSCA-Net model demonstrated a 3.65 % to 7.49 % improvement in accuracy over the comparison models.

Recording the tactile P300 with the cEEGrid for potential use in a brain-computer interface

Brain-computer interfaces (BCIs) are scientifically well established, but they rarely arrive in the daily lives of potential end-users. This could be in part because electroencephalography (EEG), a prevalent method to acquire brain activity for BCI operation, is considered too impractical to be applied in daily life of end-users with physical impairment as an assistive device. Hence, miniaturized EEG systems such as the cEEGrid have been developed. While they promise to be a step toward bridging the gap between BCI development, lab demonstrations, and home use, they still require further validation. Encouragingly, the cEEGrid has already demonstrated its ability to record visually and auditorily evoked event-related potentials (ERP), which are important as input signal for many BCIs. With this study, we aimed at evaluating the cEEGrid in the context of a BCI based on tactually evoked ERPs. To compare the cEEGrid with a conventional scalp EEG, we recorded brain activity with both systems simultaneously. Forty healthy participants were recruited to perform a P300 oddball task based on vibrotactile stimulation at four different positions. This tactile paradigm has been shown to be feasible for BCI repeatedly but has never been tested with the cEEGrid. We found distinct P300 deflections in the cEEGrid data, particularly at vertical bipolar channels. With an average of 63%, the cEEGrid classification accuracy was significantly above the chance level (25%) but significantly lower than the 81% reached with the EEG cap. Likewise, the P300 amplitude was significantly lower (cEEGrid R2-R7: 1.87 μV, Cap Cz: 3.53 μV). These results indicate that a tactile BCI using the cEEGrid could potentially be operated, albeit with lower efficiency. Additionally, participants' somatosensory sensitivity was assessed, but no correlation to the accuracy of either EEG system was shown. Our research contributes to the growing amount of literature comparing the cEEGrid to conventional EEG systems and provides first evidence that the tactile P300 can be recorded behind the ear. A BCI based on a thus simplified EEG system might be more readily accepted by potential end-users, provided the accuracy can be substantially increased, e.g., by training and improved classification.

A systematic evaluation of Euclidean alignment with deep learning for EEG decoding

Objective:Electroencephalography signals are frequently used for various Brain-Computer interface (BCI) tasks. While deep learning (DL) techniques have shown promising results, they are hindered by the substantial data requirements. By leveraging data from multiple subjects, transfer learning enables more effective training of DL models. A technique that is gaining popularity is Euclidean alignment (EA) due to its ease of use, low computational complexity, and compatibility with DL models. However, few studies evaluate its impact on the training performance of shared and individual DL models. In this work, we systematically evaluate the effect of EA combined with DL for decoding BCI signals.Approach:We used EA as a pre-processing step to train shared DL models with data from multiple subjects and evaluated their transferability to new subjects.Main results:Our experimental results show that it improves decoding in the target subject by 4.33% and decreases convergence time by more than 70%. We also trained individual models for each subject to use as a majority-voting ensemble classifier. In this scenario, using EA improved the 3-model ensemble accuracy by 3.71%. However, when compared to the shared model with EA, the ensemble accuracy was 3.62% lower.Significance:EA succeeds in the task of improving transfer learning performance with DL models and, could be used as a standard pre-processing technique.

Atypical low-frequency cortical encoding of speech identifies children with developmental dyslexia

Slow cortical oscillations play a crucial role in processing the speech amplitude envelope, which is perceived atypically by children with developmental dyslexia. Here we use electroencephalography (EEG) recorded during natural speech listening to identify neural processing patterns involving slow oscillations that may characterize children with dyslexia. In a story listening paradigm, we find that atypical power dynamics and phase-amplitude coupling between delta and theta oscillations characterize dyslexic versus other child control groups (typically-developing controls, other language disorder controls). We further isolate EEG common spatial patterns (CSP) during speech listening across delta and theta oscillations that identify dyslexic children. A linear classifier using four delta-band CSP variables predicted dyslexia status (0.77 AUC). Crucially, these spatial patterns also identified children with dyslexia when applied to EEG measured during a rhythmic syllable processing task. This transfer effect (i.e., the ability to use neural features derived from a story listening task as input features to a classifier based on a rhythmic syllable task) is consistent with a core developmental deficit in neural processing of speech rhythm. The findings are suggestive of distinct atypical neurocognitive speech encoding mechanisms underlying dyslexia, which could be targeted by novel interventions.

Optimizing spatial specificity and signal quality in fNIRS: an overview of potential challenges and possible options for improving the reliability of real-time applications

The optical brain imaging method functional near-infrared spectroscopy (fNIRS) is a promising tool for real-time applications such as neurofeedback and brain-computer interfaces. Its combination of spatial specificity and mobility makes it particularly attractive for clinical use, both at the bedside and in patients' homes. Despite these advantages, optimizing fNIRS for real-time use requires careful attention to two key aspects: ensuring good spatial specificity and maintaining high signal quality. While fNIRS detects superficial cortical brain regions, consistently and reliably targeting specific regions of interest can be challenging, particularly in studies that require repeated measurements. Variations in cap placement coupled with limited anatomical information may further reduce this accuracy. Furthermore, it is important to maintain good signal quality in real-time contexts to ensure that they reflect the true underlying brain activity. However, fNIRS signals are susceptible to contamination by cerebral and extracerebral systemic noise as well as motion artifacts. Insufficient real-time preprocessing can therefore cause the system to run on noise instead of brain activity. The aim of this review article is to help advance the progress of fNIRS-based real-time applications. It highlights the potential challenges in improving spatial specificity and signal quality, discusses possible options to overcome these challenges, and addresses further considerations relevant to real-time applications. By addressing these topics, the article aims to help improve the planning and execution of future real-time studies, thereby increasing their reliability and repeatability.

Session-independent subject-adaptive mental imagery BCI using selective filter-bank adaptive Riemannian features

The brain-computer interfaces (BCIs) facilitate the users to exploit information encoded in neural signals, specifically electroencephalogram (EEG), to control devices and for neural rehabilitation. Mental imagery (MI)-driven BCI predicts the user's pre-meditated mental objectives, which could be deployed as command signals. This paper presents a novel learning-based framework for classifying MI tasks using EEG-based BCI. In particular, our work focuses on the variation in inter-session data and the extraction of multi-spectral user-tailored features for robust performance. Thus, the goal is to create a calibration-free subject-adaptive learning framework for various mental imagery tasks not restricted to motor imagery alone. In this regard, critical spectral bands and the best temporal window are first selected from the EEG training trials of the subject based on the Riemannian user learning distance metric (Dscore) that checks for distinct and stable patterns. The filtered covariance matrices of the EEG trials in each spectral band are then transformed towards a reference covariance matrix using the Riemannian transfer learning, enabling the different sessions to be comparable. The evaluation of our proposed Selective Time-window and Multi-scale Filter-Bank with Adaptive Riemannian (STFB-AR) features on four public datasets, including disabled subjects, showed around 15% and 8% improvement in mean accuracy over baseline and fixed filter-bank models, respectively.

Three-stage transfer learning for motor imagery EEG recognition

Motor imagery (MI) paradigms have been widely used in neural rehabilitation and drowsiness state assessment. The progress in brain-computer interface (BCI) technology has emphasized the importance of accurately and efficiently detecting motor imagery intentions from electroencephalogram (EEG). Despite the recent breakthroughs made in developing EEG-based algorithms for decoding MI, the accuracy and efficiency of these models remain limited by technical challenges posed by cross-subject heterogeneity in EEG data processing and the scarcity of EEG data for training. Inspired by the optimal transport theory, this study aims to develop a novel three-stage transfer learning (TSTL) method, which uses the existing labeled data from a source domain to improve classification performance on an unlabeled target domain. Notably, the proposed method comprises three components, namely, the Riemannian tangent space mapping (RTSM), source domain transformer (SDT), and optimal subspace mapping (OSM). The RTSM maps a symmetric positive definite matrix from the Riemannian space to the tangent space to minimize the marginal probability distribution drift. The SDT transforms the source domain to a target domain by finding the optimal transport mapping matrix to reduce the joint probability distribution differences. The OSM finally maps the transformed source domain and original target domain to the same subspace to further mitigate the distribution discrepancy. The performance of the proposed method was validated on two public BCI datasets, and the average accuracy of the algorithm on two datasets was 72.24% and 69.29%. Our results demonstrated the improved performance of EEG-based MI detection in comparison with state-of-the-art algorithms.

Analysis of modulations of mental fatigue on intra-individual variability from single-trial event related potentials

BACKGROUND

Intra-individual variability (IIV), a measure of variance within an individual's performance, has been demonstrated as metrics of brain responses for neural functionality. However, how mental fatigue modulates IIV remains unclear. Consequently, the development of robust mental fatigue detection methods at the single-trial level is challenging.

NEW METHODS

Based on a long-duration flanker task EEG dataset, the modulations of mental fatigue on IIV were explored in terms of response time (RT) and trial-to-trial latency variations of event-related potentials (ERPs). Specifically, latency variations were quantified using residue iteration decomposition (RIDE) to reconstruct latency-corrected ERPs. We compared reconstructed ERPs with raw ERPs by means of temporal principal component analysis (PCA). Furthermore, a single-trial classification pipeline was developed to detect the changes of mental fatigue levels.

RESULTS

We found an increased IIV in the RT metric in the fatigue state compared to the alert state. The same sequence of ERPs (N1, P2, N2, P3a, P3b, and slow wave, or SW) was separated from both raw and reconstructed ERPs using PCA, whereas differences between raw and reconstructed ERPs in explained variances for separated ERPs were found owing to IIV. Particularly, a stronger N2 was detected in the fatigue than alert state after RIDE. The single-trial fatigue detection pipeline yielded an acceptable accuracy of 73.3%.

COMPARISON WITH EXISTING METHODS

The IIV has been linked to aging and brain disorders, and as an extension, our finding demonstrates IIV as an efficient indicator of mental fatigue.

CONCLUSIONS

This study reveals significant modulations of mental fatigue on IIV at the behavioral and neural levels and establishes a robust mental fatigue detection pipeline.

Attention-based deep convolutional neural network for classification of generalized and focal epileptic seizures

Epilepsy affects over 50 million people globally. Electroencephalography is critical for epilepsy diagnosis, but manual seizure classification is time-consuming and requires extensive expertise. This paper presents an automated multi-class seizure classification model using EEG signals from the Temple University Hospital Seizure Corpus ver. 1.5.2. 11 features including time-based correlation, time-based eigenvalues, power spectral density, frequency-based correlation, frequency-based eigenvalues, sample entropy, spectral entropy, logarithmic sum, standard deviation, absolute mean, and ratio of Daubechies D4 wavelet transformed coefficients were extracted from 10-second sliding windows across channels. The model combines multi-head self-attention mechanism with a deep convolutional neural network (CNN) to classify seven subtypes of generalized and focal epileptic seizures. The model achieved 0.921 weighted accuracy and 0.902 weighted F1 score in classifying focal onset non-motor, generalized onset non-motor, simple partial, complex partial, absence, tonic, and tonic-clonic seizures. In comparison, a CNN model without multi-head attention achieved 0.767 weighted accuracy. Ablation studies were conducted to validate the importance of transformer encoders and attention. The promising classification results demonstrate the potential of deep learning for handling EEG complexity and improving epilepsy diagnosis. This seizure classification model could enable timely interventions when translated into clinical practice.

Semi-supervised multi-source transfer learning for cross-subject EEG motor imagery classification

Electroencephalogram (EEG) motor imagery (MI) classification refers to the use of EEG signals to identify and classify subjects' motor imagery activities; this task has received increasing attention with the development of brain-computer interfaces (BCIs). However, the collection of EEG data is usually time-consuming and labor-intensive, which makes it difficult to obtain sufficient labeled data from the new subject to train a new model. Moreover, the EEG signals of different individuals exhibit significant differences, leading to a significant drop in the performance of a model trained on the existing subjects when directly classifying EEG signals acquired from new subjects. Therefore, it is crucial to make full use of the EEG data of the existing subjects and the unlabeled EEG data of the new target subject to improve the MI classification performance achieved for the target subject. This research study proposes a semi-supervised multi-source transfer (SSMT) learning model to address the above problems; the model learns informative and domain-invariant representations to address cross-subject MI-EEG classification tasks. In particular, a dynamic transferred weighting schema is presented to obtain the final predictions by integrating the weighted features derived from multi-source domains. The average accuracies achieved on two publicly available EEG datasets reach 83.57 % and 85.09 % , respectively, validating the effectiveness of the SSMT process. The SSMT process reveals the importance of informative and domain-invariant representations in MI classification tasks, as they make full use of the domain-invariant information acquired from each subject.

An Explainable and Generalizable Recurrent Neural Network Approach for Differentiating Human Brain States on EEG Dataset

Electroencephalogram (EEG) is one of the most widely used brain computer interface (BCI) approaches. Despite the success of existing EEG approaches in brain state recognition studies, it is still challenging to differentiate brain states via explainable and generalizable deep learning approaches. In other words, how to explore meaningful and distinguishing features and how to overcome the huge variability and overfitting problem still need to be further studied. To alleviate these challenges, in this work, a multiple random fragment search-based multilayer recurrent neural network (MRFS-MRNN) is proposed to improve the differentiating performance and explore meaningful patterns. Specifically, an explainable MRNN module is proposed to capture the temporal dependences preserved in EEG time series. Besides, a MRFS module is designed to cut multiple random fragments from the entire EEG signal time course to improve the effectiveness of brain state differentiating ability. MRFS-MRNN is concatenatedto effectively overcome the huge variabilities and overfitting problems. Experiment results demonstrate that the proposed MRFS-MRNN model not only has excellent differentiating performance, but also has good explanation and generalization ability. The classification accuracies reach as high as 95.18% for binary classification and 89.19% for four-category classification on the individual level. Similarly, 95.53% and 85.84% classification accuracies are obtained for the binary and four-category classification on the group level. What's more, 94.28% and 85.43% classification accuracies of binary and four-category classifications are achieved for predicting brand new subjects. The experiment results showed that the proposed method outperformed other state-of-the-art (SOTA) models on the same underlying data and improved the explanation and generalization ability.

Riemannian transfer learning based on log-Euclidean metric for EEG classification

Introduction

Brain computer interfaces (BCI), which establish a direct interaction between the brain and the external device bypassing peripheral nerves, is one of the hot research areas. How to effectively convert brain intentions into instructions for controlling external devices in real-time remains a key issue that needs to be addressed in brain computer interfaces. The Riemannian geometry-based methods have achieved competitive results in decoding EEG signals. However, current Riemannian classifiers tend to overlook changes in data distribution, resulting in degenerated classification performance in cross-session and/or cross subject scenarios.

Methods

This paper proposes a brain signal decoding method based on Riemannian transfer learning, fully considering the drift of the data distribution. Two Riemannian transfer learning methods based log-Euclidean metric are developed, such that historical data (source domain) can be used to aid the training of the Riemannian decoder for the current task, or data from other subjects can be used to boost the training of the decoder for the target subject.

Results

The proposed methods were verified on BCI competition III, IIIa, and IV 2a datasets. Compared with the baseline that without transfer learning, the proposed algorithm demonstrates superior classification performance. In contrast to the Riemann transfer learning method based on the affine invariant Riemannian metric, the proposed method obtained comparable classification performance, but is much more computationally efficient.

Discussion

With the help of proposed transfer learning method, the Riemannian classifier obtained competitive performance to existing methods in the literature. More importantly, the transfer learning process is unsupervised and time-efficient, possessing potential for online learning scenarios.

Advancing passive BCIs: a feasibility study of two temporal derivative features and effect size-based feature selection in continuous online EEG-based machine error detection

The emerging integration of Brain-Computer Interfaces (BCIs) in human-robot collaboration holds promise for dynamic adaptive interaction. The use of electroencephalogram (EEG)-measured error-related potentials (ErrPs) for online error detection in assistive devices offers a practical method for improving the reliability of such devices. However, continuous online error detection faces challenges such as developing efficient and lightweight classification techniques for quick predictions, reducing false alarms from artifacts, and dealing with the non-stationarity of EEG signals. Further research is essential to address the complexities of continuous classification in online sessions. With this study, we demonstrated a comprehensive approach for continuous online EEG-based machine error detection, which emerged as the winner of a competition at the 32nd International Joint Conference on Artificial Intelligence. The competition consisted of two stages: an offline stage for model development using pre-recorded, labeled EEG data, and an online stage 3 months after the offline stage, where these models were tested live on continuously streamed EEG data to detect errors in orthosis movements in real time. Our approach incorporates two temporal-derivative features with an effect size-based feature selection technique for model training, together with a lightweight noise filtering method for online sessions without recalibration of the model. The model trained in the offline stage not only resulted in a high average cross-validation accuracy of 89.9% across all participants, but also demonstrated remarkable performance during the online session 3 months after the initial data collection without further calibration, maintaining a low overall false alarm rate of 1.7% and swift response capabilities. Our research makes two significant contributions to the field. Firstly, it demonstrates the feasibility of integrating two temporal derivative features with an effect size-based feature selection strategy, particularly in online EEG-based BCIs. Secondly, our work introduces an innovative approach designed for continuous online error prediction, which includes a straightforward noise rejection technique to reduce false alarms. This study serves as a feasibility investigation into a methodology for seamless error detection that promises to transform practical applications in the domain of neuroadaptive technology and human-robot interaction.

Multi-Task Heterogeneous Ensemble Learning-Based Cross-Subject EEG Classification Under Stroke Patients

Robot-assisted motor training is applied for neurorehabilitation in stroke patients, using motor imagery (MI) as a representative paradigm of brain-computer interfaces to offer real-life assistance to individuals facing movement challenges. However, the effectiveness of training with MI may vary depending on the location of the stroke lesion, which should be considered. This paper introduces a multi-task electroencephalogram-based heterogeneous ensemble learning (MEEG-HEL) specifically designed for cross-subject training. In the proposed framework, common spatial patterns were used for feature extraction, and the features according to stroke lesions are shared and selected through sequential forward floating selection. The heterogeneous ensembles were used as classifiers. Nine patients with chronic ischemic stroke participated, engaging in MI and motor execution (ME) paradigms involving finger tapping. The classification criteria for the multi-task were established in two ways, taking into account the characteristics of stroke patients. In the cross-subject session, the first involved a direction recognition task for two-handed classification, achieving a performance of 0.7419 (±0.0811) in MI and 0.7061 (±0.1270) in ME. The second task focused on motor assessment for lesion location, resulting in a performance of 0.7457 (±0.1317) in MI and 0.6791 (±0.1253) in ME. Comparing the specific-subject session, except for ME on the motor assessment task, performance on both tasks was significantly higher than the cross-subject session. Furthermore, classification performance was similar to or statistically higher in cross-subject sessions compared to baseline models. The proposed MEEG-HEL holds promise in improving the practicality of neurorehabilitation in clinical settings and facilitating the detection of lesions.

EEG source imaging of hand movement-related areas: an evaluation of the reconstruction and classification accuracy with optimized channels

The hand motor activity can be identified and converted into commands for controlling machines through a brain-computer interface (BCI) system. Electroencephalography (EEG) based BCI systems employ electrodes to measure the electrical brain activity projected at the scalp and discern patterns. However, the volume conduction problem attenuates the electric potential from the brain to the scalp and introduces spatial mixing to the signals. EEG source imaging (ESI) techniques can be applied to alleviate these issues and enhance the spatial segregation of information. Despite this potential solution, the use of ESI has not been extensively applied in BCI systems, largely due to accuracy concerns over reconstruction accuracy when using low-density EEG (ldEEG), which is commonly used in BCIs. To overcome these accuracy issues in low channel counts, recent studies have proposed reducing the number of EEG channels based on optimized channel selection. This work presents an evaluation of the spatial and temporal accuracy of ESI when applying optimized channel selection towards ldEEG number of channels. For this, a simulation study of source activity related to hand movement has been performed using as a starting point an EEG system with 339 channels. The results obtained after optimization show that the activity in the concerned areas can be retrieved with a spatial accuracy of 3.99, 10.69, and 14.29 mm (localization error) when using 32, 16, and 8 channel counts respectively. In addition, the use of optimally selected electrodes has been validated in a motor imagery classification task, obtaining a higher classification performance when using 16 optimally selected channels than 32 typical electrode distributions under 10-10 system, and obtaining higher classification performance when combining ESI methods with the optimal selected channels.

A causal perspective on brainwave modeling for brain-computer interfaces

Objective. Machine learning (ML) models have opened up enormous opportunities in the field of brain-computer Interfaces (BCIs). Despite their great success, they usually face severe limitations when they are employed in real-life applications outside a controlled laboratory setting.Approach. Mixing causal reasoning, identifying causal relationships between variables of interest, with brainwave modeling can change one's viewpoint on some of these major challenges which can be found in various stages in the ML pipeline, ranging from data collection and data pre-processing to training methods and techniques.Main results. In this work, we employ causal reasoning and present a framework aiming to breakdown and analyze important challenges of brainwave modeling for BCIs.Significance. Furthermore, we present how general ML practices as well as brainwave-specific techniques can be utilized and solve some of these identified challenges. And finally, we discuss appropriate evaluation schemes in order to measure these techniques' performance and efficiently compare them with other methods that will be developed in the future.

Tracking the Cognitive Band in an Open-Ended Task

Open-ended tasks can be decomposed into the three levels of Newell's Cognitive Band: the Unit-Task level, the Operation level, and the Deliberate-Act level. We analyzed the video game Co-op Space Fortress at these levels, reporting both the match of a cognitive model to subject behavior and the use of electroencephalogram (EEG) to track subject cognition. The Unit Task level in this game involves coordinating with a partner to kill a fortress. At this highest level of the Cognitive Band, there is a good match between subject behavior and the model. The EEG signals were also strong enough to track when Unit Tasks succeeded or failed. The intermediate Operation level in this task involves legs of flight to achieve a kill. The EEG signals associated with these operations are much weaker than the signals associated with the Unit Tasks. Still, it was possible to reconstruct subject play with much better than chance success. There were significant differences in the leg behavior of subjects and models. Model behavior did not provide a good basis for interpreting a subject's behavior at this level. At the lowest Deliberate-Act level, we observed overlapping key actions, which the model did not display. Such overlapping key actions also frustrated efforts to identify EEG signals of motor actions. We conclude that the Unit-task level is the appropriate level both for understanding open-ended tasks and for using EEG to track the performance of open-ended tasks.

Comparison of recognition methods for an asynchronous (un-cued) BCI system: an investigation with 40-class SSVEP dataset

Steady-state visual evoked potential (SSVEP)-based brain-computer Interface (BCI) has demonstrated the potential to manage multi-command targets to achieve high-speed communication. Recent studies on multi-class SSVEP-based BCI have focused on synchronous systems, which rely on predefined time and task indicators; thus, these systems that use passive approaches may be less suitable for practical applications. Asynchronous systems recognize the user's intention (whether or not the user is willing to use systems) from brain activity; then, after recognizing the user's willingness, they begin to operate by switching swiftly for real-time control. Consequently, various methodologies have been proposed to capture the user's intention. However, in-depth investigation of recognition methods in asynchronous BCI system is lacking. Thus, in this work, three recognition methods (power spectral density analysis, canonical correlation analysis (CCA), and support vector machine (SVM)) used widely in asynchronous SSVEP BCI systems were explored to compare their performance. Further, we categorized asynchronous systems into two approaches (1-stage and 2-stage) based upon the recognition process's design, and compared their performance. To do so, a 40-class SSVEP dataset collected from 40 subjects was introduced. Finally, we found that the CCA-based method in the 2-stage approach demonstrated statistically significantly higher performance with a sensitivity of 97.62 ± 02.06%, specificity of 76.50 ± 23.50%, and accuracy of 75.59 ± 10.09%. Thus, it is expected that the 2-stage approach together with CCA-based recognition and FB-CCA classification have good potential to be implemented in practical asynchronous SSVEP BCI systems.

MEET: A Multi-Band EEG Transformer for Brain States Decoding

OBJECTIVE

Electroencephalography (EEG) is among the most widely used and inexpensive neuroimaging techniques. Compared to the CNN or RNN based models, Transformer can better capture the temporal information in EEG signals and focus more on global features of the brain's functional activities. Importantly, according to the multiscale nature of EEG signals, it is crucial to consider the multi-band concept into the design of EEG Transformer architecture.

METHODS

We propose a novel Multi-band EEG Transformer (MEET) to represent and analyze the multiscale temporal time series of human brain EEG signals. MEET mainly includes three parts: 1) transform the EEG signals into multi-band images, and preserve the 3D spatial information between electrodes; 2) design a Band Attention Block to compute the attention maps of the stacked multi-band images and infer the fused feature maps; 3) apply the Temporal Self-Attention and Spatial Self-Attention modules to extract the spatiotemporal features for the characterization and differentiation of multi-frame dynamic brain states.

RESULTS

The experimental results show that: 1) MEET outperforms state-of-the-art methods on multiple open EEG datasets (SEED, SEED-IV, WM) for brain states classification; 2) MEET demonstrates that 5-bands fusion is the best integration strategy; and 3) MEET identifies interpretable brain attention regions.

SIGNIFICANCE

MEET is an interpretable and universal model based on the multiband-multiscale characteristics of EEG.

CONCLUSION

The innovative combination of band attention and temporal/spatial self-attention mechanisms in MEET achieves promising data-driven learning of the temporal dependencies and spatial relationships of EEG signals across the entire brain in a holistic and comprehensive fashion.

Machine learning for hand pose classification from phasic and tonic EMG signals during bimanual activities in virtual reality

Myoelectric prostheses have recently shown significant promise for restoring hand function in individuals with upper limb loss or deficiencies, driven by advances in machine learning and increasingly accessible bioelectrical signal acquisition devices. Here, we first introduce and validate a novel experimental paradigm using a virtual reality headset equipped with hand-tracking capabilities to facilitate the recordings of synchronized EMG signals and hand pose estimation. Using both the phasic and tonic EMG components of data acquired through the proposed paradigm, we compare hand gesture classification pipelines based on standard signal processing features, convolutional neural networks, and covariance matrices with Riemannian geometry computed from raw or xDAWN-filtered EMG signals. We demonstrate the performance of the latter for gesture classification using EMG signals. We further hypothesize that introducing physiological knowledge in machine learning models will enhance their performances, leading to better myoelectric prosthesis control. We demonstrate the potential of this approach by using the neurophysiological integration of the "move command" to better separate the phasic and tonic components of the EMG signals, significantly improving the performance of sustained posture recognition. These results pave the way for the development of new cutting-edge machine learning techniques, likely refined by neurophysiology, that will further improve the decoding of real-time natural gestures and, ultimately, the control of myoelectric prostheses.

Fatigue in children using motor imagery and P300 brain-computer interfaces

BACKGROUND

Brain-computer interface (BCI) technology offers children with quadriplegic cerebral palsy unique opportunities for communication, environmental exploration, learning, and game play. Research in adults demonstrates a negative impact of fatigue on BCI enjoyment, while effects on BCI performance are variable. To date, there have been no pediatric studies of BCI fatigue. The purpose of this study was to assess the effects of two different BCI paradigms, motor imagery and visual P300, on the development of self-reported fatigue and an electroencephalography (EEG) biomarker of fatigue in typically developing children.

METHODS

Thirty-seven typically-developing school-aged children were recruited to a prospective, crossover study. Participants attended three sessions: (A) motor imagery-BCI, (B) visual P300-BCI, and (C) video viewing (control). The motor imagery task involved an imagined left- or right-hand squeeze. The P300 task involved attending to one square on a 3 × 3 grid during a random single flash sequence. Each paradigm had respective calibration periods and a similar visual counting game. Primary outcomes were self-reported fatigue and the power of the EEG alpha band both collected during resting-state periods pre- and post-task. Self-reported fatigue was measured using a 10-point visual analog scale. EEG alpha band power was calculated as the integrated power spectral density from 8 to 12 Hz of the EEG spectrum.

RESULTS

Thirty-two children completed the protocol (age range 7-16, 63% female). Self-reported fatigue and EEG alpha band power increased across all sessions (F(1,155) = 33.9, p < 0.001; F = 5.0(1,149), p = 0.027 respectively). No differences in fatigue development were observed between session types. There was no correlation between self-reported fatigue and EEG alpha band power change. BCI performance varied between participants and paradigms as expected but was not associated with self-reported fatigue or EEG alpha band power.

CONCLUSION

Short periods (30-mintues) of BCI use can increase self-reported fatigue and EEG alpha band power to a similar degree in children performing motor imagery and P300 BCI paradigms. Performance was not associated with our measures of fatigue; the impact of fatigue on useability and enjoyment is unclear. Our results reflect the variability of fatigue and the BCI experience more broadly in children and warrant further investigation.

Multimodal decoding of error processing in a virtual reality flight simulation

Technological advances in head-mounted displays (HMDs) facilitate the acquisition of physiological data of the user, such as gaze, pupil size, or heart rate. Still, interactions with such systems can be prone to errors, including unintended behavior or unexpected changes in the presented virtual environments. In this study, we investigated if multimodal physiological data can be used to decode error processing, which has been studied, to date, with brain signals only. We examined the feasibility of decoding errors solely with pupil size data and proposed a hybrid decoding approach combining electroencephalographic (EEG) and pupillometric signals. Moreover, we analyzed if hybrid approaches can improve existing EEG-based classification approaches and focused on setups that offer increased usability for practical applications, such as the presented game-like virtual reality flight simulation. Our results indicate that classifiers trained with pupil size data can decode errors above chance. Moreover, hybrid approaches yielded improved performance compared to EEG-based decoders in setups with a reduced number of channels, which is crucial for many out-of-the-lab scenarios. These findings contribute to the development of hybrid brain-computer interfaces, particularly in combination with wearable devices, which allow for easy acquisition of additional physiological data.

Artifact removal and motor imagery classification in EEG using advanced algorithms and modified DNN

This paper presents an advanced approach for EEG artifact removal and motor imagery classification using a combination of Four Class Iterative Filtering and Filter Bank Common Spatial Pattern Algorithm with a Modified Deep Neural Network (DNN) classifier. The research aims to enhance the accuracy and reliability of BCI systems by addressing the challenges posed by EEG artifacts and complex motor imagery tasks. The methodology begins by introducing FCIF, a novel technique for ocular artifact removal, utilizing iterative filtering and filter banks. FCIF's mathematical formulation allows for effective artifact mitigation, thereby improving the quality of EEG data. In tandem, the FC-FBCSP algorithm is introduced, extending the Filter Bank Common Spatial Pattern approach to handle four-class motor imagery classification. The Modified DNN classifier enhances the discriminatory power of the FC-FBCSP features, optimizing the classification process. The paper showcases a comprehensive experimental setup, featuring the utilization of BCI Competition IV Dataset 2a & 2b. Detailed preprocessing steps, including filtering and feature extraction, are presented with mathematical rigor. Results demonstrate the remarkable artifact removal capabilities of FCIF and the classification prowess of FC-FBCSP combined with the Modified DNN classifier. Comparative analysis highlights the superiority of the proposed approach over baseline methods and the method achieves the mean accuracy of 98.575%.

Improved empirical mode decomposition bagging RCSP combined with Fisher discriminant method for EEG feature extraction and classification

Background

Traditional Common Spatial Pattern (CSP) algorithms for Electroencephalogram (EEG) signal classification are sensitive to noise and can produce low accuracy in small sample datasets.

New method

To solve the problem, an improved Empirical Mode Decomposition (EMD) Bagging Regularized CSP (RCSP) algorithm is proposed. It filters EEG signals through improved EMD, inhibits high-frequency noise, retains effective information in the characteristic frequency band, and uses Bagging algorithm for data reconstruction. Feature extraction is performed with regularization of spatial patterns and Fisher linear discriminant analysis for feature classification. T-test is used for classification.

Results

The improved EMD Bagging RCSP algorithm has improved accuracy and robustness compared to CSP and its derivatives. The average classification rate is increased by about 6%, demonstrating the effectiveness and correctness of the proposed algorithm.Comparison with existing methods: The proposed algorithm outperforms CSP and its derivatives by retaining effective information and inhibiting high-frequency noise in small sample EEG datasets.

Conclusions

The proposed EMD Bagging RCSP algorithm provides a reliable and effective method for EEG signal classification and can be used in various applications, including brain-computer interfaces and clinical EEG diagnosis.

Reproducible machine learning research in mental workload classification using EEG

This study addresses concerns about reproducibility in scientific research, focusing on the use of electroencephalography (EEG) and machine learning to estimate mental workload. We established guidelines for reproducible machine learning research using EEG and used these to assess the current state of reproducibility in mental workload modeling. We first started by summarizing the current state of reproducibility efforts in machine learning and in EEG. Next, we performed a systematic literature review on Scopus, Web of Science, ACM Digital Library, and Pubmed databases to find studies about reproducibility in mental workload prediction using EEG. All of this previous work was used to formulate guidelines, which we structured along the widely recognized Cross-Industry Standard Process for Data Mining (CRISP-DM) framework. By using these guidelines, researchers can ensure transparency and comprehensiveness of their methodologies, therewith enhancing collaboration and knowledge-sharing within the scientific community, and enhancing the reliability, usability and significance of EEG and machine learning techniques in general. A second systematic literature review extracted machine learning studies that used EEG to estimate mental workload. We evaluated the reproducibility status of these studies using our guidelines. We highlight areas studied and overlooked and identify current challenges for reproducibility. Our main findings include limitations on reporting performance on unseen test data, open sharing of data and code, and reporting of resources essential for training and inference processes.

A comparative study of stereo-dependent SSVEP targets and their impact on VR-BCI performance

Steady-state visual evoked potential brain-computer interfaces (SSVEP-BCI) have attracted significant attention due to their ease of deployment and high performance in terms of information transfer rate (ITR) and accuracy, making them a promising candidate for integration with consumer electronics devices. However, as SSVEP characteristics are directly associated with visual stimulus attributes, the influence of stereoscopic vision on SSVEP as a critical visual attribute has yet to be fully explored. Meanwhile, the promising combination of virtual reality (VR) devices and BCI applications is hampered by the significant disparity between VR environments and traditional 2D displays. This is not only due to the fact that screen-based SSVEP generally operates under static, stable conditions with simple and unvaried visual stimuli but also because conventional luminance-modulated stimuli can quickly induce visual fatigue. This study attempts to address these research gaps by designing SSVEP paradigms with stereo-related attributes and conducting a comparative analysis with the traditional 2D planar paradigm under the same VR environment. This study proposed two new paradigms: the 3D paradigm and the 3D-Blink paradigm. The 3D paradigm induces SSVEP by modulating the luminance of spherical targets, while the 3D-Blink paradigm employs modulation of the spheres' opacity instead. The results of offline 4-object selection experiments showed that the accuracy of 3D and 2D paradigm was 85.67 and 86.17% with canonical correlation analysis (CCA) and 86.17 and 91.73% with filter bank canonical correlation analysis (FBCCA), which is consistent with the reduction in the signal-to-noise ratio (SNR) of SSVEP harmonics for the 3D paradigm observed in the frequency-domain analysis. The 3D-Blink paradigm achieved 75.00% of detection accuracy and 27.02 bits/min of ITR with 0.8 seconds of stimulus time and task-related component analysis (TRCA) algorithm, demonstrating its effectiveness. These findings demonstrate that the 3D and 3D-Blink paradigms supported by VR can achieve improved user comfort and satisfactory performance, while further algorithmic optimization and feature analysis are required for the stereo-related paradigms. In conclusion, this study contributes to a deeper understanding of the impact of binocular stereoscopic vision mechanisms on SSVEP paradigms and promotes the application of SSVEP-BCI in diverse VR environments.

Brain-machine interface based on deep learning to control asynchronously a lower-limb robotic exoskeleton: a case-of-study

BACKGROUND

This research focused on the development of a motor imagery (MI) based brain-machine interface (BMI) using deep learning algorithms to control a lower-limb robotic exoskeleton. The study aimed to overcome the limitations of traditional BMI approaches by leveraging the advantages of deep learning, such as automated feature extraction and transfer learning. The experimental protocol to evaluate the BMI was designed as asynchronous, allowing subjects to perform mental tasks at their own will.

METHODS

A total of five healthy able-bodied subjects were enrolled in this study to participate in a series of experimental sessions. The brain signals from two of these sessions were used to develop a generic deep learning model through transfer learning. Subsequently, this model was fine-tuned during the remaining sessions and subjected to evaluation. Three distinct deep learning approaches were compared: one that did not undergo fine-tuning, another that fine-tuned all layers of the model, and a third one that fine-tuned only the last three layers. The evaluation phase involved the exclusive closed-loop control of the exoskeleton device by the participants' neural activity using the second deep learning approach for the decoding.

RESULTS

The three deep learning approaches were assessed in comparison to an approach based on spatial features that was trained for each subject and experimental session, demonstrating their superior performance. Interestingly, the deep learning approach without fine-tuning achieved comparable performance to the features-based approach, indicating that a generic model trained on data from different individuals and previous sessions can yield similar efficacy. Among the three deep learning approaches compared, fine-tuning all layer weights demonstrated the highest performance.

CONCLUSION

This research represents an initial stride toward future calibration-free methods. Despite the efforts to diminish calibration time by leveraging data from other subjects, complete elimination proved unattainable. The study's discoveries hold notable significance for advancing calibration-free approaches, offering the promise of minimizing the need for training trials. Furthermore, the experimental evaluation protocol employed in this study aimed to replicate real-life scenarios, granting participants a higher degree of autonomy in decision-making regarding actions such as walking or stopping gait.

TBEEG: A Two-Branch Manifold Domain Enhanced Transformer Algorithm for Learning EEG Decoding

The electroencephalogram-based (EEG) brain-computer interface (BCI) has garnered significant attention in recent research. However, the practicality of EEG remains constrained by the lack of efficient EEG decoding technology. The challenge lies in effectively translating intricate EEG into meaningful, generalizable information. EEG signal decoding primarily relies on either time domain or frequency domain information. There lacks a method capable of simultaneously and effectively extracting both time and frequency domain features, as well as efficiently fuse these features. Addressing these limitations, a two-branch Manifold Domain enhanced transformer algorithm is designed to holistically capture EEG's spatio-temporal information. Our method projects the time-domain information of EEG signals into the Riemannian spaces to fully decode the time dependence of EEG signals. Using wavelet transform, the time domain information is converted into frequency domain information, and the spatial information contained in the frequency domain information of EEG signal is mined through the spectrogram. The effectiveness of the proposed TBEEG algorithm is validated on BCIC-IV-2a dataset and MAMEM-SSVEP-II datasets.

EEG rhythm separation and time-frequency analysis of fast multivariate empirical mode decomposition for motor imagery BCI

Motor imagery electroencephalogram (EEG) is widely employed in brain-computer interface (BCI) systems. As a time-frequency analysis method for nonlinear and non-stationary signals, multivariate empirical mode decomposition (MEMD) and its noise-assisted version (NA-MEMD) has been widely used in the preprocessing step of BCI systems for separating EEG rhythms corresponding to specific brain activities. However, when applied to multichannel EEG signals, MEMD or NA-MEMD often demonstrate low robustness to noise and high computational complexity. To address these issues, we have explored the advantages of our recently proposed fast multivariate empirical mode decomposition (FMEMD) and its noise-assisted version (NA-FMEMD) for analyzing motor imagery data. We emphasize that FMEMD enables a more accurate estimation of EEG frequency information and exhibits a more noise-robust decomposition performance with improved computational efficiency. Comparative analysis with MEMD on simulation data and real-world EEG validates the above assertions. The joint average frequency measure is employed to automatically select intrinsic mode functions that correspond to specific frequency bands. Thus, FMEMD-based classification architecture is proposed. Using FMEMD as a preprocessing algorithm instead of MEMD can improve the classification accuracy by 2.3% on the BCI Competition IV dataset. On the Physiobank Motor/Mental Imagery dataset and BCI Competition IV Dataset 2a, FMEMD-based architecture also attained a comparable performance to complex algorithms. The results indicate that FMEMD proficiently extracts feature information from small benchmark datasets while mitigating dimensionality constraints resulting from computational complexity. Hence, FMEMD or NA-FMEMD can be a powerful time-frequency preprocessing method for BCI.

Manifold Learning-Based Common Spatial Pattern for EEG Signal Classification

EEG signal classification using Riemannian manifolds has shown great potential. However, the huge computational cost associated with Riemannian metrics poses challenges for applying Riemannian methods, particularly in high-dimensional feature data. To address these, we propose an efficient ensemble method called MLCSP-TSE-MLP, which aims to reduce the computational cost while achieving superior performance. MLCSP of the ensemble utilizes a Riemannian graph embedding strategy to learn intrinsic low-dimensional sub-manifolds, enhancing discrimination. TSE uses the Euclidean mean as the reference point for tangent space mapping and reducing computational cost. Finally, the ensemble incorporates the MLP classifier to offer improved classification performance. Classification results conducted on three datasets demonstrate that MLCSP-TSE-MLP achieves significant superior performance compared to various competing methods. Notably, the MLCSP-TSE module achieves a remarkable increase in training speed and exhibits much lower test time compared to traditional Riemannian methods. Based on these results, we believe that the proposed MLCSP-TSE-MLP is a powerful tool for handling high-dimensional data and holds great potential for practical applications.

EEG-based finger movement classification with intrinsic time-scale decomposition

Introduction

Brain-computer interfaces (BCIs) are systems that acquire the brain's electrical activity and provide control of external devices. Since electroencephalography (EEG) is the simplest non-invasive method to capture the brain's electrical activity, EEG-based BCIs are very popular designs. Aside from classifying the extremity movements, recent BCI studies have focused on the accurate coding of the finger movements on the same hand through their classification by employing machine learning techniques. State-of-the-art studies were interested in coding five finger movements by neglecting the brain's idle case (i.e., the state that brain is not performing any mental tasks). This may easily cause more false positives and degrade the classification performances dramatically, thus, the performance of BCIs. This study aims to propose a more realistic system to decode the movements of five fingers and the no mental task (NoMT) case from EEG signals.

Methods

In this study, a novel praxis for feature extraction is utilized. Using Proper Rotational Components (PRCs) computed through Intrinsic Time Scale Decomposition (ITD), which has been successfully applied in different biomedical signals recently, features for classification are extracted. Subsequently, these features were applied to the inputs of well-known classifiers and their different implementations to discriminate between these six classes. The highest classifier performances obtained in both subject-independent and subject-dependent cases were reported. In addition, the ANOVA-based feature selection was examined to determine whether statistically significant features have an impact on the classifier performances or not.

Results

As a result, the Ensemble Learning classifier achieved the highest accuracy of 55.0% among the tested classifiers, and ANOVA-based feature selection increases the performance of classifiers on five-finger movement determination in EEG-based BCI systems.

Discussion

When compared with similar studies, proposed praxis achieved a modest yet significant improvement in classification performance although the number of classes was incremented by one (i.e., NoMT).

Evaluation of a headphones-fitted EEG system for the recording of auditory evoked potentials and mental workload assessment

Advancements in portable neuroimaging technologies open up new opportunities to gain insight into the neural dynamics and cognitive processes underlying day-to-day behaviors. In this study, we evaluated the relevance of a headphone- mounted electroencephalogram (EEG) system for monitoring mental workload. The participants (N = 12) were instructed to pay attention to auditory alarms presented sporadically while performing the Multi-Attribute Task Battery (MATB) whose difficulty was staged across three conditions to manipulate mental workload. The P300 Event-Related Potentials (ERP) elicited by the presentation of auditory alarms were used as probes of attentional resources available. The amplitude and latency of P300 ERPs were compared across experimental conditions. Our findings indicate that the P300 ERP component can be captured using a headphone-mounted EEG system. Moreover, neural responses to alarm could be used to classify mental workload with high accuracy (over 80%) at a single-trial level. Our analyses indicated that the signal-to-noise ratio acquired by the sponge-based sensors remained stable throughout the recordings. These results highlight the potential of portable neuroimaging technology for the development of neuroassistive applications while underscoring the current limitations and challenges associated with the integration of EEG sensors in everyday-life wearable technologies. Overall, our study contributes to the growing body of research exploring the feasibility and validity of wearable neuroimaging technologies for the study of human cognition and behavior in real-world settings.

Electroencephalography oscillations can predict the cortical response following theta burst stimulation

BACKGROUND

Continuous theta burst stimulation and intermittent theta burst stimulation are clinically popular models of repetitive transcranial magnetic stimulation. However, they are limited by high variability between individuals in cortical excitability changes following stimulation. Although electroencephalography oscillations have been reported to modulate the cortical response to transcranial magnetic stimulation, their association remains unclear. This study aims to explore whether machine learning models based on EEG oscillation features can predict the cortical response to transcranial magnetic stimulation.

METHOD

Twenty-three young, healthy adults attended two randomly assigned sessions for continuous and intermittent theta burst stimulation. In each session, ten minutes of resting-state electroencephalography were recorded before delivering brain stimulation. Participants were classified as responders or non-responders based on changes in resting motor thresholds. Support vector machines and multi-layer perceptrons were used to establish predictive models of individual responses to transcranial magnetic stimulation.

RESULT

Among the evaluated algorithms, support vector machines achieved the best performance in discriminating responders from non-responders for intermittent theta burst stimulation (accuracy: 91.30%) and continuous theta burst stimulation (accuracy: 95.66%). The global clustering coefficient and global characteristic path length in the beta band had the greatest impact on model output.

CONCLUSION

These findings suggest that EEG features can serve as markers of cortical response to transcranial magnetic stimulation. They offer insights into the association between neural oscillations and variability in individuals' responses to transcranial magnetic stimulation, aiding in the optimization of individualized protocols.

EEGNet-based multi-source domain filter for BCI transfer learning

Deep learning has great potential on decoding EEG in brain-computer interface. While common deep learning algorithms cannot directly train models with data from multiple individuals because of the inter-individual differences in EEG. Collecting enough data for each subject to satisfy the training of deep learning would result in an increase in training cost. This study proposes a novel transfer learning, EEGNet-based multi-source domain filter for transfer learning (EEGNet-MDFTL), to reduce the amount of training data and improve the performance of BCI. The EEGNet-MDFTL uses bagging ensemble learning to learn domain-invariant features from the multi-source domain and utilizes model loss value to filter the multi-source domain. Compared with baseline methods, the accuracy of the EEGNet-MDFTL reaches 91.96%, higher than two state-of-the-art methods, which demonstrates source domain filter can select similar source domains to improve the accuracy of the model, and remains a high level even when the data amount is reduced to 1/8, proving that ensemble learning learns enough domain invariant features from the multi-source domain to make the model insensitive to data amount. The proposed EEGNet-MDFTL is effective in improving the decoding performance with a small amount of data, which is helpful to save the BCI training cost.

Brain-Computer Interfaces, Completely Locked-In State in Neurodegenerative Diseases, and End-of-Life Decisions

In the future, policies surrounding end-of-life decisions will be faced with the question of whether competent people in a completely locked-in state should be enabled to make end-of-life decisions via brain-computer interfaces (BCI). This article raises ethical issues with acting through BCIs in the context of these decisions, specifically self-administration requirements within assisted suicide policies. We argue that enabling patients to end their life even once they have entered completely locked-in state might, paradoxically, prolong and uphold their quality of life.

Cortical activation and BCI performance during brief tactile imagery: A comparative study with motor imagery

Brain-computer interfaces (BCIs) rely heavily on motor imagery (MI) for operation, yet tactile imagery (TI) presents a novel approach that may be advantageous in situations where visual feedback is impractical. The current study aimed to compare the cortical activity and digit classification performance induced by TI and MI to assess the viability of TI for use in BCIs. Twelve right-handed participants engaged in trials of TI and MI, focusing on their left and right index digits. Event-related desynchronization (ERD) in the mu and beta bands was analyzed, and classification accuracy was determined through an artificial neural network (ANN). Comparable ERD patterns were observed in both TI and MI, with significant decreases in ERD during imagery tasks. The ANN demonstrated high classification accuracy, with TI achieving a mean±SD of 79.30 ± 3.91 % and MI achieving 81.10 ± 2.96 %, with no significant difference between the two (p = 0.11). The study found that TI induces substantial ERD comparable to MI and maintains high classification accuracy, supporting its potential as an effective mental strategy for BCIs. This suggests that TI could be a valuable alternative in BCI applications, particularly for individuals unable to rely on visual cues.

Exploring EEG-based motor imagery decoding: a dual approach using spatial features and spectro-spatial Deep Learning model IFNet

Introduction

In recent years, the decoding of motor imagery (MI) from electroencephalography (EEG) signals has become a focus of research for brain-machine interfaces (BMIs) and neurorehabilitation. However, EEG signals present challenges due to their non-stationarity and the substantial presence of noise commonly found in recordings, making it difficult to design highly effective decoding algorithms. These algorithms are vital for controlling devices in neurorehabilitation tasks, as they activate the patient's motor cortex and contribute to their recovery.

Methods

This study proposes a novel approach for decoding MI during pedalling tasks using EEG signals. A widespread approach is based on feature extraction using Common Spatial Patterns (CSP) followed by a linear discriminant analysis (LDA) as a classifier. The first approach covered in this work aims to investigate the efficacy of a task-discriminative feature extraction method based on CSP filter and LDA classifier. Additionally, the second alternative hypothesis explores the potential of a spectro-spatial Convolutional Neural Network (CNN) to further enhance the performance of the first approach. The proposed CNN architecture combines a preprocessing pipeline based on filter banks in the frequency domain with a convolutional neural network for spectro-temporal and spectro-spatial feature extraction.

Results and discussion

To evaluate the approaches and their advantages and disadvantages, EEG data has been recorded from several able-bodied users while pedalling in a cycle ergometer in order to train motor imagery decoding models. The results show levels of accuracy up to 80% in some cases. The CNN approach shows greater accuracy despite higher instability.

Towards unlocking motor control in spinal cord injured by applying an online EEG-based framework to decode motor intention, trajectory and error processing

Brain-computer interfaces (BCIs) can translate brain signals directly into commands for external devices. Electroencephalography (EEG)-based BCIs mostly rely on the classification of discrete mental states, leading to unintuitive control. The ERC-funded project "Feel Your Reach" aimed to establish a novel framework based on continuous decoding of hand/arm movement intention, for a more natural and intuitive control. Over the years, we investigated various aspects of natural control, however, the individual components had not yet been integrated. Here, we present a first implementation of the framework in a comprehensive online study, combining (i) goal-directed movement intention, (ii) trajectory decoding, and (iii) error processing in a unique closed-loop control paradigm. Testing involved twelve able-bodied volunteers, performing attempted movements, and one spinal cord injured (SCI) participant. Similar movement-related cortical potentials and error potentials to previous studies were revealed, and the attempted movement trajectories were overall reconstructed. Source analysis confirmed the involvement of sensorimotor and posterior parietal areas for goal-directed movement intention and trajectory decoding. The increased experiment complexity and duration led to a decreased performance than each single BCI. Nevertheless, the study contributes to understanding natural motor control, providing insights for more intuitive strategies for individuals with motor impairments.

Ecological decoding of visual aesthetic preference with oscillatory electroencephalogram features-A mini-review

In today's digital information age, human exposure to visual artifacts has reached an unprecedented quasi-omnipresence. Some of these cultural artifacts are elevated to the status of artworks which indicates a special appreciation of these objects. For many persons, the perception of such artworks coincides with aesthetic experiences (AE) that can positively affect health and wellbeing. AEs are composed of complex cognitive and affective mental and physiological states. More profound scientific understanding of the neural dynamics behind AEs would allow the development of passive Brain-Computer-Interfaces (BCI) that offer personalized art presentation to improve AE without the necessity of explicit user feedback. However, previous empirical research in visual neuroaesthetics predominantly investigated functional Magnetic Resonance Imaging and Event-Related-Potentials correlates of AE in unnaturalistic laboratory conditions which might not be the best features for practical neuroaesthetic BCIs. Furthermore, AE has, until recently, largely been framed as the experience of beauty or pleasantness. Yet, these concepts do not encompass all types of AE. Thus, the scope of these concepts is too narrow to allow personalized and optimal art experience across individuals and cultures. This narrative mini-review summarizes the state-of-the-art in oscillatory Electroencephalography (EEG) based visual neuroaesthetics and paints a road map toward the development of ecologically valid neuroaesthetic passive BCI systems that could optimize AEs, as well as their beneficial consequences. We detail reported oscillatory EEG correlates of AEs, as well as machine learning approaches to classify AE. We also highlight current limitations in neuroaesthetics and suggest future directions to improve EEG decoding of AE.

Analyzing and computing humans by means of the brain using Brain-Computer Interfaces - understanding the user - previous evidence, self-relevance and the user's self-concept as potential superordinate human factors of relevance

Brain-computer interfaces (BCIs) are well-known instances of how technology can convert a user's brain activity taken from non-invasive electroencephalography (EEG) into computer commands for the purpose of computer-assisted communication and interaction. However, not all users are attaining the accuracy required to use a BCI consistently, despite advancements in technology. Accordingly, previous research suggests that human factors could be responsible for the variance in BCI performance among users. Therefore, the user's internal mental states and traits including motivation, affect or cognition, personality traits, or the user's satisfaction, beliefs or trust in the technology have been investigated. Going a step further, this manuscript aims to discuss which human factors could be potential superordinate factors that influence BCI performance, implicitly, explicitly as well as inter- and intraindividually. Based on the results of previous studies that used comparable protocols to examine the motivational, affective, cognitive state or personality traits of healthy and vulnerable EEG-BCI users within and across well-investigated BCIs (P300-BCIs or SMR-BCIs, respectively), it is proposed that the self-relevance of tasks and stimuli and the user's self-concept provide a huge potential for BCI applications. As potential key human factors self-relevance and the user's self-concept (self-referential knowledge and beliefs about one's self) guide information processing and modulate the user's motivation, attention, or feelings of ownership, agency, and autonomy. Changes in the self-relevance of tasks and stimuli as well as self-referential processing related to one's self (self-concept) trigger changes in neurophysiological activity in specific brain networks relevant to BCI. Accordingly, concrete examples will be provided to discuss how past and future research could incorporate self-relevance and the user's self-concept in the BCI setting - including paradigms, user instructions, and training sessions.

Unleashing the potential of fNIRS with machine learning: classification of fine anatomical movements to empower future brain-computer interface

In this study, we explore the potential of using functional near-infrared spectroscopy (fNIRS) signals in conjunction with modern machine-learning techniques to classify specific anatomical movements to increase the number of control commands for a possible fNIRS-based brain-computer interface (BCI) applications. The study focuses on novel individual finger-tapping, a well-known task in fNIRS and fMRI studies, but limited to left/right or few fingers. Twenty-four right-handed participants performed the individual finger-tapping task. Data were recorded by using sixteen sources and detectors placed over the motor cortex according to the 10-10 international system. The event's average oxygenated Δ HbO and deoxygenated Δ HbR hemoglobin data were utilized as features to assess the performance of diverse machine learning (ML) models in a challenging multi-class classification setting. These methods include LDA, QDA, MNLR, XGBoost, and RF. A new DL-based model named "Hemo-Net" has been proposed which consists of multiple parallel convolution layers with different filters to extract the features. This paper aims to explore the efficacy of using fNRIS along with ML/DL methods in a multi-class classification task. Complex models like RF, XGBoost, and Hemo-Net produce relatively higher test set accuracy when compared to LDA, MNLR, and QDA. Hemo-Net has depicted a superior performance achieving the highest test set accuracy of 76%, however, in this work, we do not aim at improving the accuracies of models rather we are interested in exploring if fNIRS has the neural signatures to help modern ML/DL methods in multi-class classification which can lead to applications like brain-computer interfaces. Multi-class classification of fine anatomical movements, such as individual finger movements, is difficult to classify with fNIRS data. Traditional ML models like MNLR and LDA show inferior performance compared to the ensemble-based methods of RF and XGBoost. DL-based method Hemo-Net outperforms all methods evaluated in this study and demonstrates a promising future for fNIRS-based BCI applications.

What are wereallydecoding? Unveiling biases in EEG-based decoding of the spatial focus of auditory attention

Objective.Spatial auditory attention decoding (Sp-AAD) refers to the task of identifying the direction of the speaker to which a person is attending in a multi-talker setting, based on the listener's neural recordings, e.g. electroencephalography (EEG). The goal of this study is to thoroughly investigate potential biases when training such Sp-AAD decoders on EEG data, particularly eye-gaze biases and latent trial-dependent confounds, which may result in Sp-AAD models that decode eye-gaze or trial-specific fingerprints rather than spatial auditory attention.Approach.We designed a two-speaker audiovisual Sp-AAD protocol in which the spatial auditory and visual attention were enforced to be either congruent or incongruent, and we recorded EEG data from sixteen participants undergoing several trials recorded at distinct timepoints. We trained a simple linear model for Sp-AAD based on common spatial patterns filters in combination with either linear discriminant analysis (LDA) or k-means clustering, and evaluated them both across- and within-trial.Main results.We found that even a simple linear Sp-AAD model is susceptible to overfitting to confounding signal patterns such as eye-gaze and trial fingerprints (e.g. due to feature shifts across trials), resulting in artificially high decoding accuracies. Furthermore, we found that changes in the EEG signal statistics across trials deteriorate the trial generalization of the classifier, even when the latter is retrained on the test trial with an unsupervised algorithm.Significance.Collectively, our findings confirm that there exist subtle biases and confounds that can strongly interfere with the decoding of spatial auditory attention from EEG. It is expected that more complicated non-linear models based on deep neural networks, which are often used for Sp-AAD, are even more vulnerable to such biases. Future work should perform experiments and model evaluations that avoid and/or control for such biases in Sp-AAD tasks.

Modeling Complex EEG Data Distribution on the Riemannian Manifold Toward Outlier Detection and Multimodal Classification

OBJECTIVE

The usage of Riemannian geometry for Brain-computer interfaces (BCIs) has gained momentum in recent years. Most of the machine learning techniques proposed for Riemannian BCIs consider the data distribution on a manifold to be unimodal. However, the distribution is likely to be multimodal rather than unimodal since high-data variability is a crucial limitation of electroencephalography (EEG). In this paper, we propose a novel data modeling method for considering complex data distributions on a Riemannian manifold of EEG covariance matrices, aiming to improve BCI reliability.

METHODS

Our method, Riemannian spectral clustering (RiSC), represents EEG covariance matrix distribution on a manifold using a graph with proposed similarity measurement based on geodesic distances, then clusters the graph nodes through spectral clustering. This allows flexibility to model both a unimodal and a multimodal distribution on a manifold. RiSC can be used as a basis to design an outlier detector named outlier detection Riemannian spectral clustering (odenRiSC) and a multimodal classifier named multimodal classifier Riemannian spectral clustering (mcRiSC). All required parameters of odenRiSC/mcRiSC are selected in data-driven manner. Moreover, there is no need to pre-set a threshold for outlier detection and the number of modes for multimodal classification.

RESULTS

The experimental evaluation revealed odenRiSC can detect EEG outliers more accurately than existing methods and mcRiSC outperformed the standard unimodal classifier, especially on high-variability datasets.

CONCLUSION

odenRiSC/mcRiSC are anticipated to contribute to making real-life BCIs outside labs and neuroergonomics applications more robust.

SIGNIFICANCE

RiSC can work as a robust EEG outlier detector and multimodal classifier.

A Few-Shot Transfer Learning Approach for Motion Intention Decoding from Electroencephalographic Signals

In this study, a few-shot transfer learning approach was introduced to decode movement intention from electroencephalographic (EEG) signals, allowing to recognize new tasks with minimal adaptation. To this end, a dataset of EEG signals recorded during the preparation of complex sub-movements was created from a publicly available data collection. The dataset was divided into two parts: the source domain dataset (including 5 classes) and the support (target domain) dataset, (including 2 classes) with no overlap between the two datasets in terms of classes. The proposed methodology consists in projecting EEG signals into the space-frequency-time domain, in processing such projections (rearranged in channels × frequency frames) by means of a custom EEG-based deep neural network (denoted as EEGframeNET5), and then adapting the system to recognize new tasks through a few-shot transfer learning approach. The proposed method achieved an average accuracy of 72.45 ± 4.19% in the 5-way classification of samples from the source domain dataset, outperforming comparable studies in the literature. In the second phase of the study, a few-shot transfer learning approach was proposed to adapt the neural system and make it able to recognize new tasks in the support dataset. The results demonstrated the system's ability to adapt and recognize new tasks with an average accuracy of 80 ± 0.12% in discriminating hand opening/closing preparation and outperforming reported results in the literature. This study suggests the effectiveness of EEG in capturing information related to the motor preparation of complex movements, potentially paving the way for BCI systems based on motion planning decoding. The proposed methodology could be straightforwardly extended to advanced EEG signal processing in other scenarios, such as motor imagery or neural disorder classification.