Motor Imagery Research Articles

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.

Early sleep after action observation plus motor imagery improves gait and balance abilities in older adults

Action observation plus motor imagery (AOMI) is a rehabilitative approach to improve gait and balance performance. However, limited benefits have been reported in older adults. Early sleep after motor practice represents a strategy to enhance the consolidation of trained skills. Here, we investigated the effects of AOMI followed by early sleep on gait and balance performance in older adults. Forty-five older adults (mean age: 70.4 ± 5.2 years) were randomized into three groups performing a 3-week training. Specifically, AOMI-sleep and AOMI-control groups underwent observation and motor imagery of gait and balance tasks between 8:00 and 10:00 p.m. or between 8:00 and 10:00 a.m. respectively, whereas Control group observed landscape video-clips. Participants were assessed for gait performance, static and dynamic balance and fear of falling before and after training and at 1-month follow-up. The results revealed that early sleep after AOMI training sessions improved gait and balance abilities in older adults compared to AOMI-control and Control groups. Furthermore, these benefits were retained at 1-month after the training end. These findings suggested that early sleep after AOMI may represent a safe and easy-applicable intervention to minimize the functional decay in older adults.

Decoding brain signals: A convolutional neural network approach for motor imagery classification

Motor imagery-centered brain-computer interfaces (BCIs) have surfaced as a promising technology with the potential to improve communication and control for people facing motor impairments. Achieving precise classification of motor imagery (MI) tasks from EEG signals is essential for the optimal functioning of BCIs. In this study, we explore the use of convolutional neural networks (CNNs) to achieve robust and precise classification of MI-EEG signals. We utilized well-established EEG datasets, namely BCI Competition IV 2a and BCI Competition IV 2b, to assess our customized CNN architecture. The proposed network achieved an excellent result with an average classification accuracy of 87.3% and 86.29% on the respective datasets. This experiment showcases the network’s capacity to distinguish various motor imagery tasks by utilizing extracted temporal-spatial characteristics from the EEG data. This proposed network creates opportunities for BCIs that can provide enhanced control and communication options for the user.

A novel precisely designed compact convolutional EEG classifier for motor imagery classification

A novel precisely designed compact convolutional EEG classifier for motor imagery classification

Effects of motor imagery and action observation on respiratory function in mild smokers: a randomized single-blind controlled pilot trial

Background: Motor imagery (MI) and action observation (AO) training can activate brain areas involved in planning, adjusting, and automating voluntary movement in a manner similar to that when these activities are being performed. Aim: The main objective of this study was to assess the effects of MI and AO training on respiratory function in mild smokers. Methods: A single-blind placebo-controlled pilot trial was designed. A total of 27 mild smokers were randomized into three groups: MI (n = 9), AO (n = 9), and sham observation (SO; n = 9) groups. The MI and AO groups performed mental training of breathing exercises while the SO group observed a landscape without a human agent. The primary outcomes were pulmonary function parameters (forced expiratory volume during the 1st s [FEV1], forced vital capacity [FVC], FEV1/FVC ratio, maximum voluntary ventilation [MVV], and peak expiratory flow [PEF]), and the secondary outcomes were maximal inspiratory/expiratory pressures (MIP/MEP) and perceived fatigue. All outcome measures were assessed at baseline and post-intervention. Results: Regarding the pulmonary function parameters, only the AO group showed significant within-group differences in FEV1 (mean differences [MD] = 0.37 L (0.17 – 0.56), P = 0.001), FVC (MD = 0.1 L (0.02 – 0.16), P = 0.008), and PEF (MD = 0.74 L/s (0.29 – 1.18), P = 0.002) with a small-to-moderate effect size. No differences were found in FEV1/FVC ratio and MVV. With regard to the maximal static pressures, only the AO group showed significant within-group differences in MEP with a small effect size (MD = 11.22 cm H2O (0.19 – 22.2), P = 0.046). Finally, both AO and MI groups showed significantly greater perceived fatigue with regard to SO group with a large effect size (P < 0.05). Conclusion: AO training has a slight impact on some pulmonary function parameters, such as FEV1, FVC, or PEF, as well as on MEP when applied in isolation and in a single session. Relevance for Patients: Although it is still early to draw some solid conclusions, AO training could be used in combination with respiratory exercises to see if the effect is greater than exercises in isolation. The study of movement representation strategies on pulmonary function is a field that has been sparingly explored so far. This paper offers some interesting data to be considered for further research.

Association between visuo-spatial working memory and gait motor imagery

Although motor imagery and working memory (WM) appear to be closely linked, no previous studies have demonstrated direct evidence for the relationship between motor imagery and WM abilities. This study investigated the association between WM and gait motor imagery and focused on the individual differences in young adults. This study included 33 participants (mean age: 22.2 ± 0.9 years). We used two methods to measure the ability of different WM domains: verbal and visuo-spatial WM. Gait motor imagery accuracy was assessed via the mental chronometry paradigm. We measured the times participants took to complete an actual and imagined walk along a 5 m walkway, with three different path widths. The linear mixed effects model analysis revealed that visuo-spatial WM ability was a significant predictor of the accuracy of gait motor imagery, but not of verbal WM ability. Specifically, individuals with lower visuo-spatial WM ability demonstrated more inaccuracies in the difficult path-width conditions. However, gait motor imagery was not as accurate as actual walking even in the easiest path width or in participants with high visuo-spatial WM ability. Further, visuo-spatial WM ability was significantly correlated with mental walking but not with actual walking. These results suggest that visuo-spatial WM is related to motor imagery rather than actual movement.

Can they touch? A novel mental motor imagery task for the assessment of back pain.

As motor imagery is informed by the anticipated sensory consequences of action, including pain, we reasoned that motor imagery could provide a useful indicator of chronic back pain. We tested the hypothesis that mental motor imagery regarding body movements can provide a reliable assessment of low back pain. Eighty-five subjects with back pain and forty-five age-matched controls were shown two names of body parts and asked to indicate if they could imagine moving so that the named body parts touched. Three types of imagined movements were interrogated: movements of arms, movements of legs and movements requiring flexion and/or rotation of the low back. Accuracy and reaction times were measured. Subjects with back pain were less likely to indicate that they could touch body parts than age-matched controls. The effect was observed only for those movements that required movement of the low back or legs, suggesting that the effect was not attributable to task difficulty or non-specific effects. There was an effect of pain severity. Compared to subjects with mild pain, subjects with severe pain were significantly less likely to indicate that they could move so that named body parts touched. There was a correlation between pain ratings and impaired performance for stimuli that involved the lower but not upper body. As the Can They Touch task is quick, easy to administer and does not require an explicit judgment of pain severity, it may provide useful information to supplement the assessment of subjects with chronic pain.

Variation in brain connectivity during motor imagery and motor execution in stroke patients based on electroencephalography.

The objective of this study was to analyze the changes in connectivity between motor imagery (MI) and motor execution (ME) in the premotor area (PMA) and primary motor cortex (MA) of the brain, aiming to explore suitable forms of treatment and potential therapeutic targets. Twenty-three inpatients with stroke were selected, and 21 right-handed healthy individuals were recruited. EEG signal during hand MI and ME (synergy and isolated movements) was recorded. Correlations between functional brain areas during MI and ME were compared. PMA and MA were significantly and positively correlated during hand MI in all participants. The power spectral density (PSD) values of PMA EEG signals were greater than those of MA during MI and ME in both groups. The functional connectivity correlation was higher in the stroke group than in healthy people during MI, especially during left-handed MI. During ME, functional connectivity correlation in the brain was more enhanced during synergy movements than during isolated movements. The regions with abnormal functional connectivity were in the 18th lead of the left PMA area. Left-handed MI may be crucial in MI therapy, and the 18th lead may serve as a target for non-invasive neuromodulation to promote further recovery of limb function in patients with stroke. This may provide support for the EEG theory of neuromodulation therapy for hemiplegic patients.

The influence of paradigm interface guided by different visual types on MI-BCI performance

ABSTRACT Visual paradigms of Brain-Computer Interfaces (BCI) for motor imagery (MI) tasks are the basis for communication through (electroencephalogram) EEG signals. During the MI-BCI user training process, this study analyzes and summarises four different visual paradigms and compares their impact on the outcomes of MI-BCI training. Four different visual paradigms are experimentally compared through classification outcomes and subjective evaluation. EEG features were extracted via Common Spatial Patterns (CSP) and passed to a Support Vector Machine (SVM) model for their classification. The results show that all four types of visual paradigms have a significant impact on the outcomes of MI-BCI training, with Paradigm Set II having the most significant impact. This is because paradigm set II offers a paradigm interface with relatively low visual complexity on the basis of action observation, and visual guidance with more clarity and more accurate EEG classification.

Improving cross-subject classification performance of motor imagery signals: a data augmentation-focused deep learning framework

Motor imagery brain-computer interfaces (MI-BCIs) have gained a lot of attention in recent years thanks to their potential to enhance rehabilitation and control of prosthetic devices for individuals with motor disabilities. However, accurate classification of motor imagery signals remains a challenging task due to the high inter-subject variability and non-stationarity in the electroencephalogram (EEG) data. In the context of MI-BCIs, with limited data availability, the acquisition of EEG data can be difficult. In this study, several data augmentation techniques have been compared with the proposed data augmentation technique adaptive cross-subject segment replacement (ACSSR). This technique, in conjunction with the proposed deep learning framework, allows for a combination of similar subject pairs to take advantage of one another and boost the classification performance of MI-BCIs. The proposed framework features a multi-domain feature extractor based on common spatial patterns with a sliding window and a parallel two-branch convolutional neural network. The performance of the proposed methodology has been evaluated on the multi-class BCI Competition IV Dataset 2a through repeated 10-fold cross-validation. Experimental results indicated that the implementation of the ACSSR method (80.47%) in the proposed framework has led to a considerable improvement in the classification performance compared to the classification without data augmentation (77.63%), and other fundamental data augmentation techniques used in the literature. The study contributes to the advancements for the development of effective MI-BCIs by showcasing the ability of the ACSSR method to address the challenges in motor imagery signal classification tasks.

Noise-Factorized Disentangled Representation Learning for Generalizable Motor Imagery EEG Classification.

Motor Imagery (MI) Electroencephalography (EEG) is one of the most common Brain-Computer Interface (BCI) paradigms that has been widely used in neural rehabilitation and gaming. Although considerable research efforts have been dedicated to developing MI EEG classification algorithms, they are mostly limited in handling scenarios where the training and testing data are not from the same subject or session. Such poor generalization capability significantly limits the realization of BCI in real-world applications. In this paper, we proposed a novel framework to disentangle the representation of raw EEG data into three components, subject/session-specific, MI-task-specific, and random noises, so that the subject/session-specific feature extends the generalization capability of the system. This is realized by a joint discriminative and generative framework, supported by a series of fundamental training losses and training strategies. We evaluated our framework on three public MI EEG datasets, and detailed experimental results show that our method can achieve superior performance by a large margin compared to current state-of-the-art benchmark algorithms.

Unraveling the Left-Right Judgment Task in Chronic Low Back Pain: Insights Through Behavioral, Electrophysiological, Motor Imagery, and Bodily Disruption Perspectives

Bodily disruptions have been consistently demonstrated in individuals with chronic low back pain. The performance on the left-right judgment task has been purposed as an indirect measure of the cortical proprioceptive representation of the body. It has been suggested to be dependent on implicit motor imagery, although the available evidence is conflicting. Hence, the aim of this case-control observational study was to examine the performance (accuracy and reaction times) and event-related potentials while performing the left-right judgment task for back and hand images in individuals with chronic low back pain versus healthy controls, along with its relationship with self-reported measurements and quantitative sensory testing. While self-reported data suggested bodily disruptions in the chronic low back pain sample, this was not supported by quantitative sensory testing. Although both groups displayed the same performance, our results suggested an increased attentional load on participants with chronic low back pain to achieve equal performance, measured by a higher N1 peak amplitude in occipital electrodes, especially when the effect of contextual images arises. The absence of differences in the reaction times for the left-right judgment task between both groups, along with inconsistencies in self-reported and quantitative sensory testing data, could question the involvement of implicit motor imagery in solving the task. In conclusion, our results suggest disrupted attentional processing in participants with chronic low back pain to solve the left-right judgment task. PerspectiveAlthough there are no differences in the performance of the left-right judgment task (hits, reaction times) between chronic low back pain patients and controls, the analysis of event-related potentials revealed that patients require a higher cognitive load, measured by N1 peak amplitude.

Personalized motor imagery prediction model based on individual difference of ERP

Objective. Motor imagery-based brain–computer interaction (MI-BCI) is a novel method of achieving human and external environment interaction that can assist individuals with motor disorders to rehabilitate. However, individual differences limit the utility of the MI-BCI. In this study, a personalized MI prediction model based on the individual difference of event-related potential (ERP) is proposed to solve the MI individual difference. Approach. A novel paradigm named action observation-based multi-delayed matching posture task evokes ERP during a delayed matching posture task phase by retrieving picture stimuli and videos, and generates MI electroencephalogram through action observation and autonomous imagery in an action observation-based motor imagery phase. Based on the correlation between the ERP and MI, a logistic regression-based personalized MI prediction model is built to predict each individual’s suitable MI action. 32 subjects conducted the MI task with or without the help of the prediction model to select the MI action. Then classification accuracy of the MI task is used to evaluate the proposed model and three traditional MI methods. Main results. The personalized MI prediction model successfully predicts suitable action among 3 sets of daily actions. Under suitable MI action, the individual’s ERP amplitude and event-related desynchronization (ERD) intensity are the largest, which helps to improve the accuracy by 14.25%. Significance. The personalized MI prediction model that uses the temporal ERP features to predict the classification accuracy of MI is feasible for improving the individual’s MI-BCI performance, providing a new personalized solution for the individual difference and practical BCI application.

Multilayer network-based channel selection for motor imagery brain–computer interface

Objective. The number of electrode channels in a motor imagery-based brain–computer interface (MI-BCI) system influences not only its decoding performance, but also its convenience for use in applications. Although many channel selection methods have been proposed in the literature, they are usually based on the univariate features of a single channel. This leads to a loss of the interaction between channels and the exchange of information between networks operating at different frequency bands. Approach. We integrate brain networks containing four frequency bands into a multilayer network framework and propose a multilayer network-based channel selection (MNCS) method for MI-BCI systems. A graph learning-based method is used to estimate the multilayer network from electroencephalogram (EEG) data that are filtered by multiple frequency bands. The multilayer participation coefficient of the multilayer network is then computed to select EEG channels that do not contain redundant information. Furthermore, the common spatial pattern (CSP) method is used to extract effective features. Finally, a support vector machine classifier with a linear kernel is trained to accurately identify MI tasks. Main results. We used three publicly available datasets from the BCI Competition containing data on 12 healthy subjects and one dataset containing data on 15 stroke patients to validate the effectiveness of our proposed method. The results showed that the proposed MNCS method outperforms all channels (85.8% vs. 93.1%, 84.4% vs. 89.0%, 71.7% vs. 79.4%, and 72.7% vs. 84.0%). Moreover, it achieved significantly higher decoding accuracies on MI-BCI systems than state-of-the-art methods (paired t-tests, p < 0.05). Significance. The experimental results showed that the proposed MNCS method can select appropriate channels to improve the decoding performance as well as the convenience of the application of MI-BCI systems.

T-TIME: Test-Time Information Maximization Ensemble for Plug-and-Play BCIs.

An electroencephalogram (EEG)-based brain-computer interface (BCI) enables direct communication between the human brain and a computer. Due to individual differences and non-stationarity of EEG signals, such BCIs usually require a subject-specific calibration session before each use, which is time-consuming and user-unfriendly. Transfer learning (TL) has been proposed to shorten or eliminate this calibration, but existing TL approaches mainly consider offline settings, where all unlabeled EEG trials from the new user are available. This article proposes Test-Time Information Maximization Ensemble (T-TIME) to accommodate the most challenging online TL scenario, where unlabeled EEG data from the new user arrive in a stream, and immediate classification is performed. T-TIME initializes multiple classifiers from the aligned source data. When an unlabeled test EEG trial arrives, T-TIME first predicts its labels using ensemble learning, and then updates each classifier by conditional entropy minimization and adaptive marginal distribution regularization. Our code is publicized. Extensive experiments on three public motor imagery based BCI datasets demonstrated that T-TIME outperformed about 20 classical and state-of-the-art TL approaches. To our knowledge, this is the first work on test time adaptation for calibration-free EEG-based BCIs, making plug-and-play BCIs possible.

Transfer learning with data alignment and optimal transport for EEG based motor imagery classification

Objective. The non-stationarity of electroencephalogram (EEG) signals and the variability among different subjects present significant challenges in current Brain–Computer Interfaces (BCI) research, which requires a time-consuming specific calibration procedure to address. Transfer Learning (TL) offers a potential solution by leveraging data or models from one or more source domains to facilitate learning in the target domain, so as to address these challenges. Approach. In this paper, a novel Multi-source domain Transfer Learning Fusion (MTLF) framework is proposed to address the calibration problem. Firstly, the method transforms the source domain data with the resting state segment data, in order to decrease the differences between the source domain and the target domain. Subsequently, feature extraction is performed using common spatial pattern. Finally, an improved TL classifier is employed to classify the target samples. Notably, this method does not require the label information of target domain samples, while concurrently reducing the calibration workload. Main results. The proposed MTLF is assessed on Datasets 2a and 2b from the BCI Competition IV. Compared with other algorithms, our method performed relatively the best and achieved mean classification accuracy of 73.69% and 70.83% on Datasets 2a and 2b respectively. Significance. Experimental results demonstrate that the MTLF framework effectively reduces the discrepancy between the source and target domains and acquires better classification performance on two motor imagery datasets.

How Integration of a Brain-Machine Interface and Obstacle Detection System Can Improve Wheelchair Control via Movement Imagery.

This study presents a human-computer interaction combined with a brain-machine interface (BMI) and obstacle detection system for remote control of a wheeled robot through movement imagery, providing a potential solution for individuals facing challenges with conventional vehicle operation. The primary focus of this work is the classification of surface EEG signals related to mental activity when envisioning movement and deep relaxation states. Additionally, this work presents a system for obstacle detection based on image processing. The implemented system constitutes a complementary part of the interface. The main contributions of this work include the proposal of a modified 10-20-electrode setup suitable for motor imagery classification, the design of two convolutional neural network (CNNs) models employed to classify signals acquired from sixteen EEG channels, and the implementation of an obstacle detection system based on computer vision integrated with a brain-machine interface. The models developed in this study achieved an accuracy of 83% in classifying EEG signals. The resulting classification outcomes were subsequently utilized to control the movement of a mobile robot. Experimental trials conducted on a designated test track demonstrated real-time control of the robot. The findings indicate the feasibility of integration of the obstacle detection system for collision avoidance with the classification of motor imagery for the purpose of brain-machine interface control of vehicles. The elaborated solution could help paralyzed patients to safely control a wheelchair through EEG and effectively prevent unintended vehicle movements.

Bidirectional feature pyramid attention-based temporal convolutional network model for motor imagery electroencephalogram classification.

As an interactive method gaining popularity, brain-computer interfaces (BCIs) aim to facilitate communication between the brain and external devices. Among the various research topics in BCIs, the classification of motor imagery using electroencephalography (EEG) signals has the potential to greatly improve the quality of life for people with disabilities. This technology assists them in controlling computers or other devices like prosthetic limbs, wheelchairs, and drones. However, the current performance of EEG signal decoding is not sufficient for real-world applications based on Motor Imagery EEG (MI-EEG). To address this issue, this study proposes an attention-based bidirectional feature pyramid temporal convolutional network model for the classification task of MI-EEG. The model incorporates a multi-head self-attention mechanism to weigh significant features in the MI-EEG signals. It also utilizes a temporal convolution network (TCN) to separate high-level temporal features. The signals are enhanced using the sliding-window technique, and channel and time-domain information of the MI-EEG signals is extracted through convolution. Additionally, a bidirectional feature pyramid structure is employed to implement attention mechanisms across different scales and multiple frequency bands of the MI-EEG signals. The performance of our model is evaluated on the BCI Competition IV-2a dataset and the BCI Competition IV-2b dataset, and the results showed that our model outperformed the state-of-the-art baseline model, with an accuracy of 87.5 and 86.3% for the subject-dependent, respectively. In conclusion, the BFATCNet model offers a novel approach for EEG-based motor imagery classification in BCIs, effectively capturing relevant features through attention mechanisms and temporal convolutional networks. Its superior performance on the BCI Competition IV-2a and IV-2b datasets highlights its potential for real-world applications. However, its performance on other datasets may vary, necessitating further research on data augmentation techniques and integration with multiple modalities to enhance interpretability and generalization. Additionally, reducing computational complexity for real-time applications is an important area for future work.

Analysis of Minimal Channel Electroencephalography for Wearable Brain–Computer Interface

Electroencephalography (EEG)-based brain—computer interface (BCI) is a non-invasive technology with potential in various healthcare applications, including stroke rehabilitation and neuro-feedback training. These applications typically require multi-channel EEG. However, setting up a multi-channel EEG headset is time-consuming, potentially resulting in patient reluctance to use the system despite its potential benefits. Therefore, we investigated the appropriate number of electrodes required for a successful BCI application in wearable devices using various numbers of EEG channels. EEG multi-frequency features were extracted using the “filter bank” feature extraction technique. A support vector machine (SVM) was used to classify a left/right-hand opening/closing motor imagery (MI) task. Nine electrodes around the center of the scalp (F3, Fz, F4, C3, Cz, C4, P3, Pz, and P4) provided high classification accuracy with a moderate setup time; hence, this system was selected as the minimal number of required channels. Spherical spline interpolation (SSI) was also applied to investigate the feasibility of generating EEG signals from limited channels on an EEG headset. We found classification accuracies of interpolated groups only, and combined interpolated and collected groups were significantly lower than the measured groups. The results indicate that SSI may not provide additional EEG data to improve classification accuracy of the collected minimal channels. The conclusion is that other techniques could be explored or a sufficient number of EEG channels must be collected without relying on generated data. Our proposed method, which uses a filter bank feature, session-dependent training, and the exploration of many groups of EEG channels, offers the possibility of developing a successful BCI application using minimal channels on an EEG device.

Motor imagery as a potential tool to alleviate choking under pressure in sports performance

Abstract Choking under pressure is a popular issue in sports, having adverse effects on athletes’ performance. It can occur for various reasons, such as debilitative anxiety, increased fear of failure, lack of confidence, lack of preparation, and the presence of distractions. This review investigates the utility of motor imagery as an intervention for choking. Motor imagery, a cognitive process encompassing mental rehearsal of physical movements without physical execution, has been demonstrated to enhance motor skill acquisition and execution among athletes. Beyond its efficacy in enhancing motor skills, motor imagery imparts psychological advantages to athletes by reducing anxiety and bolstering confidence, both pivotal factors in averting performance breakdowns under pressure. In summary, motor imagery holds promise as an intervention for alleviating choking under pressure in sports performance. Nonetheless, further investigations are imperative to ascertain optimal approaches for integrating motor imagery into sports training and competitive settings.