Brain-computer Interface System Research Articles

Assessing robustness to adversarial attacks in attention-based networks: Case of EEG-based motor imagery classification

The classification of motor imagery (MI) using Electroencephalography (EEG) plays a pivotal role in facilitating communication for individuals with physical limitations through Brain-Computer Interface (BCI) systems. Recent strides in Attention-Based Networks (ATN) have showcased remarkable performance in EEG signal classification, presenting a promising alternative to conventional Convolutional Neural Networks (CNNs). However, while CNNs have been extensively analyzed for their resilience against adversarial attacks, the susceptibility of ATNs in comparable scenarios remains largely unexplored. This paper aims to fill this gap by investigating the robustness of ATNs in adversarial contexts. We propose a high-performing attention-based deep learning model specifically designed for classifying Motor Imagery (MI) brain signals extracted from EEG data. Subsequently, we conduct a thorough series of experiments to assess various attack strategies targeting ATNs employed in EEG-based BCI tasks. Our analysis utilizes the widely recognized BCI Competition 2a dataset to demonstrate the effectiveness of attention mechanisms in BCI endeavors. Despite achieving commendable classification results in terms of accuracy (87.15%) and kappa score (0.8287), our findings reveal the vulnerability of attention-based models to adversarial manipulations (accuracy: 9.07%, kappa score: -0.21), highlighting the imperative for bolstering the robustness of attention architectures for EEG classification tasks.

Efficient dual-frequency SSVEP brain-computer interface system exploiting interocular visual resource disparities

Here, we present a novel dual-frequency steady-state visual evoked potential (SSVEP) brain-computer interface (BCI) system with a unique human–computer interaction (HCI) feature that utilizes distinct visual resource disparities between the two eyes. By leveraging subtle differences in electroencephalography (EEG) responses due to binocular disparity, our system effectively distinguishes identical frequency targets. We conducted a two-phase validation: initial offline experiments confirmed EEG differences from interocular disparity, followed by global optimization to refine the SSVEP coding and spatial arrangement, allowing for 40 distinct targets within 20 frequency combinationes. A tailored binocular-task-related component analysis (bTRCA) algorithm was developed, achieving a maximum individual information transmission rate (ITR) of 338.32 bits/min offline and an average of 224.96 bits/min, with the online version reaching an average ITR of 234.13 bits/min. These results indicate that this system is a leading candidate for dual-frequency SSVEP BCIs, opening new prospects for HCI applications.

A hierarchical recursive feature elimination algorithm to develop brain computer interface application of user behavior for statistical reasoning and decision making

BackgroundWith the aid of a brain computer interface (BCI), users can communicate and receive signals wirelessly or over wired connections to operate smart devices. A BCI classifier's architecture is quite difficult since numerous elements should be combined. These elements are made up of brain signals, which also include high levels of weak sounds that could provide reliable participant recordings of daily activities. We must use computer vision techniques to create a model in order to control those information. The high dimension and volume of signals present the classification classifier with its primary obstacles. New methodDue to this, we extracted and classified the brain activity in this study, and we also presented a novel hierarchical recursive feature elimination method that we refer to as HRFE embracing noisy additions. HRFE makes a variety of categorization suggestions to eliminate bias in classifying BCI systems of different types. We put the HRFE to the test on two BCI signal data sets—specifically, dataset I and BCI contests III—using shallow and deep convolution network classification techniques. Just a grid of assets is used to create electrocorticography (ECoG) signals on the contralateral (right) motor cortex, and these signals are recorded in the BCI contests III database. ResultsUsing ECoG signals, we choose the top 20 features that have the biggest effects on distortion and classification selection. Comparison with existing methodsThe simulation findings show that HRFE has a significant computer vision enhancement when compared to comparable feature selection methods in the literature, particularly for ECoG signal, which achieves about 93% reliability.

The use of happy faces as visual stimuli improves the performance of the hybrid SSVEP+P300 brain computer interface

BackgroundThis study illustrates a hybrid brain-computer interface (BCI) in which steady-state visual evoked potentials (SSVEP) and event-related potentials (P300) are evoked simultaneously. The goal of this study was to improve the performance of the current hybrid SSVEP+P300 BCI systems by incorporating a happy face into visual stimuli. New methodIn this study, happy and sad faces were added to a visual stimulus to induce stronger cortical signals in a hybrid SSVEP+P300 BCI. Additionally, we developed a paradigm in which SSVEP responses were triggered by non-face stimuli, whereas P300 responses were triggered by face stimuli. We tested four paradigms: happy face paradigm (HF), sad face paradigm (SF), happy face and flicker paradigm (HFF), and sad face and flicker paradigm (SFF). Results and conclusionsThe results demonstrated that the HFF paradigm elicited more robust cortical responses, which resulted in enhanced system accuracy and information transfer rate (ITR). The HFF paradigm has a system communication rate of 25.9 bits per second and an average accuracy of 96.1%. Compared with other paradigms, the HFF paradigm is the best choice for BCI applications because it has the highest ITR and maximum level of comfort.

Playing Tetris with Your Mind: A BCI-Controlled Gaming Experience

Brain-Computer Interface Controlled Tetris: Exploring Feasibility and User Experience. This project explores the feasibility and user experience of controlling a Tetris game using a Brain-Computer Interface (BCI). The BCI system utilizes non-invasive Electroencephalography (EEG) to detect P300 brainwaves generated by user focus on specific on-screen cues corresponding to desired game actions (left, right, rotate, drop).An ESP8266 microcontroller board, EEG sensor, and display are used to create a portable BCI system. Software development leverages open-source libraries for EEG signal processing, classification, and game engine control (onboard or external).The experiment evaluates BCI control performance through metrics like accuracy, speed, and Tetris completion rate. A control group playing traditional Tetris (if applicable) allows for comparison. Subjective user experience is assessed through questionnaires and surveys. The results are expected to demonstrate the feasibility of BCI- controlled Tetris using affordable hardware. Potential trade-offs between BCI control speed and traditional methods will be explored. User experience analysis will provide insights into the challenge, enjoyment, and future directions of BCI gaming. This project contributes to the growing field of BCI applications by demonstrating its potential for novel and mind-controlled gaming experiences. Keywords: Brain-Computer Interface, Electroencephalography, performance.

A review on electroencephalogram based channel selection

The electroencephalogram (EEG) signal is the key signal carrier of the brain-computer interface (BCI) system. The EEG data collected by the whole-brain electrode arrangement is conducive to obtaining higher information representation. Personalized electrode layout, while ensuring the accuracy of EEG signal decoding, can also shorten the calibration time of BCI and has become an important research direction. This paper reviews the EEG signal channel selection methods in recent years, conducts a comparative analysis of the combined effects of different channel selection methods and different classification algorithms, obtains the commonly used channel combinations in motor imagery, P300 and other paradigms in BCI, and explains the application scenarios of the channel selection method in different paradigms are discussed, in order to provide stronger support for a more accurate and portable BCI system.

Online continual decoding of streaming EEG signal with a balanced and informative memory buffer

Electroencephalography (EEG) based Brain Computer Interface (BCI) systems play a significant role in facilitating how individuals with neurological impairments effectively interact with their environment. In real world applications of BCI system for clinical assistance and rehabilitation training, the EEG classifier often needs to learn on sequentially arriving subjects in an online manner. As patterns of EEG signals can be significantly different for different subjects, the EEG classifier can easily erase knowledge of learnt subjects after learning on later ones as it performs decoding in online streaming scenario, namely catastrophic forgetting. In this work, we tackle this problem with a memory-based approach, which considers the following conditions: (1) subjects arrive sequentially in an online manner, with no large scale dataset available for joint training beforehand, (2) data volume from the different subjects could be imbalanced, (3) decoding difficulty of the sequential streaming signal vary, (4) continual classification for a long time is required. This online sequential EEG decoding problem is more challenging than classic cross subject EEG decoding as there is no large-scale training data from the different subjects available beforehand. The proposed model keeps a small balanced memory buffer during sequential learning, with memory data dynamically selected based on joint consideration of data volume and informativeness. Furthermore, for the more general scenarios where subject identity is unknown to the EEG decoder, aka. subject agnostic scenario, we propose a kernel based subject shift detection method that identifies underlying subject changes on the fly in a computationally efficient manner. We develop challenging benchmarks of streaming EEG data from sequentially arriving subjects with both balanced and imbalanced data volumes, and performed extensive experiments with a detailed ablation study on the proposed model. The results show the effectiveness of our proposed approach, enabling the decoder to maintain performance on all previously seen subjects over a long period of sequential decoding. The model demonstrates the potential for real-world applications.

EEG-based brain-computer interface methods with the aim of rehabilitating advanced stage ALS patients

Amyotrophic Lateral Sclerosis (ALS) is a neurodegenerative disease that leads to progressive muscle weakness and paralysis, ultimately resulting in the loss of ability to communicate and control the environment. EEG-based Brain-Computer Interface (BCI) methods have shown promise in providing communication and control with the aim of rehabilitating ALS patients. In particular, P300-based BCI has been widely studied and used for ALS rehabilitation. Other EEG-based BCI methods, such as Motor Imagery (MI) based BCI and Hybrid BCI, have also shown promise in ALS rehabilitation. Nonetheless, EEG-based BCI methods hold great potential for improvement. This review article introduces and reviews FFT, WPD, CSP, CSSP, CSP, and GC feature extraction methods. The Common Spatial Pattern (CSP) is an efficient and common technique for extracting data properties used in BCI systems. In addition, Linear Discriminant Analysis (LDA), Support Vector Machine (SVM), Neural Networks (NN), and Deep Learning (DL) classification methods were introduced and reviewed. SVM is the most appropriate classifier due to its insensitivity to the curse of dimensionality. Also, DL is used in the design of BCI systems and is a good choice for BCI systems based on motor imagery with big datasets. Despite the progress made in the field, there are still challenges to overcome, such as improving the accuracy and reliability of EEG signal detection and developing more intuitive and user-friendly interfaces By using BCI, disabled patients can communicate with their caregivers and control their environment using various devices, including wheelchairs, and robotic arms.

Efficient Feature Learning Model of Motor Imagery EEG Signals with L1-Norm and Weighted Fusion.

Brain-computer interface (BCI) for motor imagery is an advanced technology used in the field of medical rehabilitation. However, due to the poor accuracy of electroencephalogram feature classification, BCI systems often misrecognize user commands. Although many state-of-the-art feature selection methods aim to enhance classification accuracy, they usually overlook the interrelationships between individual features, indirectly impacting the accuracy of feature classification. To overcome this issue, we propose an adaptive feature learning model that employs a Riemannian geometric approach to generate a feature matrix from electroencephalogram signals, serving as the model's input. By integrating the enhanced adaptive L1 penalty and weighted fusion penalty into the sparse learning model, we select the most informative features from the matrix. Specifically, we measure the importance of features using mutual information and introduce an adaptive weight construction strategy to penalize regression coefficients corresponding to each variable adaptively. Moreover, the weighted fusion penalty balances weight differences among correlated variables, reducing the model's overreliance on specific variables and enhancing accuracy. The performance of the proposed method was validated on BCI Competition IV datasets IIa and IIb using the support vector machine. Experimental results demonstrate the effectiveness and superiority of the proposed model compared to the existing models.

Resting state EEG assisted imagined vowel phonemes recognition by native and non-native speakers using brain connectivity measures.

Communication is challenging for disabled individuals, but with advancement of brain-computer interface (BCI) systems, alternative communication systems can be developed. Current BCI spellers, such as P300, SSVEP, and MI, have drawbacks like reliance on external stimuli or conversation irrelevant mental tasks. In contrast to these systems, Imagined speech based BCI systems rely on directly decoding the vowels/words user is thinking, making them more intuitive, user friendly and highly popular among Brain-Computer-Interface (BCI) researchers. However, more research needs to be conducted on how subject-specific characteristics such as mental state, age, handedness, nativeness and resting state activity affects the brain's output during imagined speech. In an overt speech, it is evident that native and non-native speakers' brains function differently. Therefore, this paper explores how nativeness to language affects EEG signals while imagining vowel phonemes, using brain-map analysis and scalogram and also investigates the inclusion of features extracted from resting state EEG with imagined state EEG. The Fourteen-channel EEG for Imagined Speech (FEIS) dataset was used to analyse the EEG signals recorded while imagining vowel phonemes for 16 subjects (nine native English and seven non-native Chinese). For the classification of vowel phonemes, different connectivity measures such as covariance, coherence, and Phase Synchronous Index-PSI were extracted and analysed using statistics based Multivariate Analysis of Variance (MANOVA) approach. Different fusion strategies (difference, concatenation, Common Spatial Pattern-CSP and Canonical Correlation Analysis-CCA) were carried out to incorporate resting state EEG connectivity measures with imagined state connectivity measures for enhancing the accuracy of imagined vowel phoneme recognition. Simulation results revealed that concatenating imagined state and rest state covariance and PSI features provided the maximum accuracy of 92.78% for native speakers and 94.07% for non-native speakers.

Enhancement of Brain Computer Interface System Based on Artificial Intelligent Technique

With the rapid development of information technology, the EEG has become an increasingly popular method for analyzing brain signals. EEG signals greatly contribute to(BCI) brain-computer communication . BCI outputs are used to restore many functions in relation to individuals with a motor impairment, and these signals are complex random signals produced from hundreds of millions of neurons in EEG mixtures in the brain that contain many data about brain activity. its include artefacts which have a great influence on the diagnosis; Hence, these unwanted signals become the most important problem in EEG signal analysis. Therefore, used four blind source separation techniques (BSS), STONE, FICA, BEFICA and EFICA. The proposed system is using one of the new Antlion (ALO) optimization algorithms to improve the performance of the previous algorithms and find out which ones are the most responsive, by comparing them and choosing the best ones according to the (PSD) standard. With the use of real data, noticed through the results that the EFICA algorithm is the best response to the improvement and the most efficient, as the ratio of (PSD) was the least possible, as ALO worked to reduce thevariance in the distribution of the frequency spectrum because it relied on solving constrained problems using various searches.

Comprehensive Review of Noninvasive Brain-Computer Interfaces for Controlling Robotic Arms

A robotic arm is a mechanical device with a given number of Degrees of Freedom (DoFs) that mimics the functions of a human arm and performs any desired task, such as grasping and moving objects. Current research is directed toward the design of robots and artificial human body parts controlled by brain signals, translating human thoughts into actions. Brain-Computer Interface (BCI) systems have been used to enable people with motor disabilities to control assistive robotic equipment that replaces the lost functions. This paper presents a review of the state-of-the-art of the latest papers dealing with the control of a robotic arm based on Electroencephalogram (EEG). A comparative study of the different methods and techniques used in different blocks of the robotic arm’s noninvasive BCI controlling system is conducted. These blocks include signal acquisition using noninvasive electrodes, signal preprocessing, feature extraction, classification, and command. Additionally, this paper presents a performance comparison of the reviewed controlling systems of robotic arms using EEG signals.

Evaluation of an online SSVEP-BCI with fast system setup

The brain–computer interface (BCI) plays an important role in neural restoration. Current BCI systems generally require complex experimental preparation to perform well, but this time-consuming process may hinder their use in clinical applications. To explore the feasibility of simplifying the BCI system setup, a wearable BCI system based on the steady-state visual evoked potential (SSVEP) was developed and evaluated. Fifteen healthy participants were recruited to test the fast-setup system using dry and wet electrodes in a real-life scenario. In this study, the average system setup time for the dry electrode was 38.40 seconds and that for the wet electrode was 103.40 seconds, which are times appreciably shorter than those in previous BCI experiments, enabling a rapid setup of the BCI system. Although the electroencephalogram (EEG) signal quality was low in this fast-setup BCI experiment, the BCI system achieved an information transfer rate of 138.89 bits/min with an eight-channel wet electrode and an information transfer rate of 70.59 bits/min with an eight-channel dry electrode, showing that the overall performance was close to that in traditional experiments. In addition, the results suggest that the solutions of a multi-channel dry electrode or few-channel wet electrode may be suitable for the fast-setup SSEVP-BCI. This fast-setup SSVEP-BCI has the advantages of simple preparation and stable performance and is thus conducive to promoting the use of the BCI in clinical practice.

A fused multi-subfrequency bands and CBAM SSVEP-BCI classification method based on convolutional neural network.

For the brain-computer interface (BCI) system based on steady-state visual evoked potential (SSVEP), it is difficult to obtain satisfactory classification performance for short-time window SSVEP signals by traditional methods. In this paper, a fused multi-subfrequency bands and convolutional block attention module (CBAM) classification method based on convolutional neural network (CBAM-CNN) is proposed for discerning SSVEP-BCI tasks. This method extracts multi-subfrequency bands SSVEP signals as the initial input of the network model, and then carries out feature fusion on all feature inputs. In addition, CBAM is embedded in both parts of the initial input and feature fusion for adaptive feature refinement. To verify the effectiveness of the proposed method, this study uses the datasets of Inner Mongolia University of Technology (IMUT) and Tsinghua University (THU) to evaluate the performance of the proposed method. The experimental results show that the highest accuracy of CBAM-CNN reaches 0.9813 percentage point (pp). Within 0.1-2s time window, the accuracy of CBAM-CNN is 0.0201-0.5388 (pp) higher than that of CNN, CCA-CWT-SVM, CCA-SVM, CCA-GNB, FBCCA, and CCA. Especially in the short-time window range of 0.1-1s, the performance advantage of CBAM-CNN is more significant. The maximum information transmission rate (ITR) of CBAM-CNN is 503.87bit/min, which is 227.53bit/min-503.41bit/min higher than the above six EEG decoding methods. The study further results show that CBAM-CNN has potential application value in SSVEP decoding.

A Data Augmentation Method for Motor Imagery EEG Signals Based on DCGAN-GP Network.

Motor imagery electroencephalography (EEG) signals have garnered attention in brain-computer interface (BCI) research due to their potential in promoting motor rehabilitation and control. However, the limited availability of labeled data poses challenges for training robust classifiers. In this study, we propose a novel data augmentation method utilizing an improved Deep Convolutional Generative Adversarial Network with Gradient Penalty (DCGAN-GP) to address this issue. We transformed raw EEG signals into two-dimensional time-frequency maps and employed a DCGAN-GP network to generate synthetic time-frequency representations resembling real data. Validation experiments were conducted on the BCI IV 2b dataset, comparing the performance of classifiers trained with augmented and unaugmented data. Results demonstrated that classifiers trained with synthetic data exhibit enhanced robustness across multiple subjects and achieve higher classification accuracy. Our findings highlight the effectiveness of utilizing a DCGAN-GP-generated synthetic EEG data to improve classifier performance in distinguishing different motor imagery tasks. Thus, the proposed data augmentation method based on a DCGAN-GP offers a promising avenue for enhancing BCI system performance, overcoming data scarcity challenges, and bolstering classifier robustness, thereby providing substantial support for the broader adoption of BCI technology in real-world applications.

Classification of Motor Imagery EEG signals using high resolution time-frequency representations and convolutional neural network

A Motor Imagery (MI) based Brain Computer Interface (BCI) system aims to provide neuro-rehabilitation for the motor disabled people and patients with brain injuries (e.g., stroke patients) etc. The aim of this work is to classify the left and right hand MI tasks by utilizing the occurrence of event related desynchronization and synchronization (ERD\\ERS) in the Electroencephalogram (EEG) during these tasks. This study proposes to use a set of Complex Morlet Wavelets (CMW) having frequency dependent widths to generate high-resolution time-frequency representations (TFR) of the MI EEG signals present in the channels C3 and C4. A novel method for the selection of the value of number of cycles relative to the center frequency of the CMW is studied here for extracting the MI task features. The generated TFRs are given as input to a Convolutional neural network (CNN) for classifying them into left or right hand MI tasks. The proposed framework attains a classification accuracy of 82.2% on the BCI Competition IV dataset 2a, showing that the TFRs generated in this work give a higher classification accuracy than the baseline methods and other existing algorithms.

A Modified Hybrid Brain-Computer Interface Speller Based on Steady-State Visual Evoked Potentials and Electromyogram.

To enhance the information transfer rate (ITR) of a steady-state visual evoked potential (SSVEP)-based speller, more characters with flickering symbols should be used. Increasing the number of symbols might reduce the classification accuracy. A hybrid brain-computer interface (BCI) improves the overall performance of a BCI system by taking advantage of two or more control signals. In a simultaneous hybrid BCI, various modalities work with each other simultaneously, which enhances the ITR. In our proposed speller, simultaneous combination of electromyogram (EMG) and SSVEP was applied to increase the ITR. To achieve 36 characters, only nine stimulus symbols were used. Each symbol allowed the selection of four characters based on four states of muscle activity. The SSVEP detected which symbol the subject was focusing on and the EMG determined the target character out of the four characters dedicated to that symbol. The frequency rate for character encoding was applied in the EMG modality and latency was considered in the SSVEP modality. Online experiments were carried out on 10 healthy subjects. The average ITR of this hybrid system was 96.1 bit/min with an accuracy of 91.2%. The speller speed was 20.9 char/min. Different subjects had various latency values. We used an average latency of 0.2 s across all subjects. Evaluation of each modality showed that the SSVEP classification accuracy varied for different subjects, ranging from 80% to 100%, while the EMG classification accuracy was approximately 100% for all subjects. Our proposed hybrid BCI speller showed improved system speed compared with state-of-the-art systems based on SSVEP or SSVEP-EMG, and can provide a user-friendly, practical system for speller applications.

Neural Correlates of Motor/Tactile Imagery and Tactile Sensation in a BCI paradigm: A High-Density EEG Source Imaging Study.

Complementary to brain-computer interface (BCI) based on motor imagery (MI) task, sensory imagery (SI) task provides a way for BCI construction using brain activity from somatosensory cortex. The underlying neurophysiological correlation between SI and MI was unclear and difficult to measure through behavior recording. In this study, we investigated the underlying neurodynamic of motor/tactile imagery and tactile sensation tasks through a high-density electroencephalogram (EEG) recording, and EEG source imaging was used to systematically explore the cortical activation differences and correlations between the tasks. In the experiment, participants were instructed to perform the left and right hand tasks in MI paradigm, sensory stimulation (SS) paradigm and SI paradigm. The statistical results demonstrated that the imagined MI and SI tasks differed from each other within ipsilateral sensorimotor scouts, frontal and right temporal areas in α bands, whereas real SS and imagined SI showed a similar activation pattern. The similarity between SS and SI may provide a way to train the BCI system, while the difference between MI and SI may provide a way to integrate the discriminative information between them to enhance BCI performance. The combination of the tasks and its underlying neurodynamic would provide a new approach for BCI designation for a wider application. BCI studies concentrate on the hybrid decoding method combining MI or SI with SS, but the underlining neurophysiological correlates between them were unclear. MI and SI differed from each other within the ipsilateral sensorimotor cortex in alpha bands. This is a first study to investigate the neurophysiological relationship between MI and SI through an EEG source imaging approach from high-density EEG recording.

Attention-ProNet: A Prototype Network with Hybrid Attention Mechanisms Applied to Zero Calibration in Rapid Serial Visual Presentation-Based Brain-Computer Interface.

The rapid serial visual presentation-based brain-computer interface (RSVP-BCI) system achieves the recognition of target images by extracting event-related potential (ERP) features from electroencephalogram (EEG) signals and then building target classification models. Currently, how to reduce the training and calibration time for classification models across different subjects is a crucial issue in the practical application of RSVP. To address this issue, a zero-calibration (ZC) method termed Attention-ProNet, which involves meta-learning with a prototype network integrating multiple attention mechanisms, was proposed in this study. In particular, multiscale attention mechanisms were used for efficient EEG feature extraction. Furthermore, a hybrid attention mechanism was introduced to enhance model generalization, and attempts were made to incorporate suitable data augmentation and channel selection methods to develop an innovative and high-performance ZC RSVP-BCI decoding model algorithm. The experimental results demonstrated that our method achieved a balance accuracy (BA) of 86.33% in the decoding task for new subjects. Moreover, appropriate channel selection and data augmentation methods further enhanced the performance of the network by affording an additional 2.3% increase in BA. The model generated by the meta-learning prototype network Attention-ProNet, which incorporates multiple attention mechanisms, allows for the efficient and accurate decoding of new subjects without the need for recalibration or retraining.

Self-supervised contrastive learning for EEG-based cross-subject motor imagery recognition.

Objective. The extensive application of electroencephalography (EEG) in brain-computer interfaces (BCIs) can be attributed to its non-invasive nature and capability to offer high-resolution data. The acquisition of EEG signals is a straightforward process, but the datasets associated with these signals frequently exhibit data scarcity and require substantial resources for proper labeling. Furthermore, there is a significant limitation in the generalization performance of EEG models due to the substantial inter-individual variability observed in EEG signals.Approach. To address these issues, we propose a novel self-supervised contrastive learning framework for decoding motor imagery (MI) signals in cross-subject scenarios. Specifically, we design an encoder combining convolutional neural network and attention mechanism. In the contrastive learning training stage, the network undergoes training with the pretext task of data augmentation to minimize the distance between pairs of homologous transformations while simultaneously maximizing the distance between pairs of heterologous transformations. It enhances the amount of data utilized for training and improves the network's ability to extract deep features from original signals without relying on the true labels of the data.Main results. To evaluate our framework's efficacy, we conduct extensive experiments on three public MI datasets: BCI IV IIa, BCI IV IIb, and HGD datasets. The proposed method achieves cross-subject classification accuracies of 67.32%, 82.34%, and 81.13%on the three datasets, demonstrating superior performance compared to existing methods.Significance. Therefore, this method has great promise for improving the performance of cross-subject transfer learning in MI-based BCI systems.