EMPT: a sparsity Transformer for EEG-based motor imagery recognition.

Introduction: Transformer network is widely emphasized and studied relying on its excellent performance. The self-attention mechanism finds a good solution for feature coding among multiple channels of electroencephalography (EEG) signals. However, using the self-attention mechanism to construct models on EEG data suffers from the problem of the large amount of data required and the complexity of the algorithm. Methods: We propose a Transformer neural network combined with the addition of Mixture of Experts (MoE) layer and ProbSparse Self-attention mechanism for decoding the time-frequency-spatial domain features from motor imagery (MI) EEG of spinal cord injury patients. The model is named as EEG MoE-Prob-Transformer (EMPT). The common spatial pattern and the modified s-transform method are employed for achieving the time-frequency-spatial features, which are used as feature embeddings to input the improved transformer neural network for feature reconstruction, and then rely on the expert model in the MoE layer for sparsity mapping, and finally output the results through the fully connected layer. Results: EMPT achieves an accuracy of 95.24% on the MI EEG dataset for patients with spinal cord injury. EMPT has also achieved excellent results in comparative experiments with other state-of-the-art methods. Discussion: The MoE layer and ProbSparse Self-attention inside the EMPT are subjected to visualisation experiments. The experiments prove that sparsity can be introduced to the Transformer neural network by introducing MoE and kullback-leibler divergence attention pooling mechanism, thereby enhancing its applicability on EEG datasets. A novel deep learning approach is presented for decoding EEG data based on MI.

Imagery-Focused Therapy for Visual Hallucinations: A Case Series.

INTRODUCTION: Visual hallucinations (VH) are more common than previously thought and are linked to higher levels of distress and disability in people with a psychotic illness. Despite this, scant attention has been given to VHs in the clinical literature, and the few therapy case series of cognitive behavioural therapy (CBT) published to date have not demonstrated reliable change. In other areas of clinical research, problematic mental imagery has been found to be more strongly related to negative affect in psychological disorders than negative linguistic thinking, and imagery focused techniques have commonly been found to improve the outcomes in CBT trials. Given VHs have many similarities with visual mental imagery and many of the distressing beliefs associated with VHs targeted in CBT are maintained by accompanying mental imagery (i.e., imaging a hallucinated figure attacking them), it seems plausible that an imagery-focused approach to treating VHs may be most effective. METHODS: The current study is a multiple baseline case series (N = 11) of a 10-session imagery-focused therapy for VH in a transdiagnostic sample. RESULTS: The study had good attendance and feedback, no adverse events and only one [seemly unrelated] drop-out, suggesting good feasibility, safety and acceptability. The majority of clients reported reduction on both full-scale measures (administered at 3 baselines, midtherapy, posttherapy and 3-month follow-up) and weekly measures of VH severity and distress, ranging from medium to large effect sizes. CONCLUSIONS: The case series suggests that an imagery-focused approach to treating VHs may be beneficial, with a recommendation for more rigorous clinical trials to follow.

Multiclass motor imagery classification with Riemannian geometry and temporal-spectral selection.

Motor imagery (MI) based brain-computer interfaces (BCIs) decode the users' intentions from electroencephalography (EEG) to achieve information control and interaction between the brain and external devices. In this paper, firstly, we apply Riemannian geometry to the covariance matrix extracted by spatial filtering to obtain robust and distinct features. Then, a multiscale temporal-spectral segmentation scheme is developed to enrich the feature dimensionality. In order to determine the optimal feature configurations, we utilize a linear learning-based temporal window and spectral band (TWSB) selection method to evaluate the feature contributions, which efficiently reduces the redundant features and improves the decoding efficiency without excessive loss of accuracy. Finally, support vector machines are used to predict the classification labels based on the selected MI features. To evaluate the performance of our model, we test it on the publicly available BCI Competition IV dataset 2a and 2b. The results show that the method has an average accuracy of 79.1% and 83.1%, which outperforms other existing methods. Using TWSB feature selection instead of selecting all features improves the accuracy by up to about 6%. Moreover, the TWSB selection method can effectively reduce the computational burden. We believe that the framework reveals more interpretable feature information of motor imagery EEG signals, provides neural responses discriminative with high accuracy, and facilitates the performance of real-time MI-BCI.

On the design of deep learning-based control algorithms for visually guided UAVs engaged in power tower inspection tasks.

This paper focuses on the design of Convolution Neural Networks to visually guide an autonomous Unmanned Aerial Vehicle required to inspect power towers. The network is required to precisely segment images taken by a camera mounted on a UAV in order to allow a motion module to generate collision-free and inspection-relevant manoeuvres of the UAV along different types of towers. The images segmentation process is particularly challenging not only because of the different structures of the towers but also because of the enormous variability of the background, which can vary from the uniform blue of the sky to the multi-colour complexity of a rural, forest, or urban area. To be able to train networks that are robust enough to deal with the task variability, without incurring into a labour-intensive and costly annotation process of physical-world images, we have carried out a comparative study in which we evaluate the performances of networks trained either with synthetic images (i.e., the synthetic dataset), physical-world images (i.e., the physical-world dataset), or a combination of these two types of images (i.e., the hybrid dataset). The network used is an attention-based U-NET. The synthetic images are created using photogrammetry, to accurately model power towers, and simulated environments modelling a UAV during inspection of different power towers in different settings. Our findings reveal that the network trained on the hybrid dataset outperforms the networks trained with the synthetic and the physical-world image datasets. Most notably, the networks trained with the hybrid dataset demonstrates a superior performance on multiples evaluation metrics related to the image-segmentation task. This suggests that, the combination of synthetic and physical-world images represents the best trade-off to minimise the costs related to capturing and annotating physical-world images, and to maximise the task performances. Moreover, the results of our study demonstrate the potential of photogrammetry in creating effective training datasets to design networks to automate the precise movement of visually-guided UAVs.

Irrigation strengthens climate resilience: Long-term evidence from Mali using satellites and surveys.

Agriculture in the Sahel and much of sub-Saharan Africa remains to a large extent rainfed. At the same time, climate change is already causing less predictable rainfall patterns in the region, even as rising temperatures increase the amount of water needed for agricultural production. We assess to what extent irrigation can strengthen the climate resilience of farming communities. Our study sample consists of nearly 1,000 distinct locations in Mali in which small-scale, river-based irrigation was introduced over the past two decades, as weather conditions worsened and political upheaval erupted. Using the staggered roll-out of the irrigation and repeated observations over 20 years allows us to compare the pre- and postirrigation outcomes of locations while adjusting for confounding factors. We geospatially link data on irrigation interventions with agricultural conditions measured using satellite imagery and surveys, as well as child nutrition and health outcomes and conflict event data. Using a two-way fixed effects model to quasi-experimentally estimate counterfactual outcomes, we find that the introduction of irrigation led to substantial increases in agricultural production on supported fields, with these gains persisting even a decade later. Children in nearby communities are less likely to be stunted or wasted due to the irrigation, and conflict risks decrease in the closest communities. Some of these gains are offset by worsening conditions farther away from the newly installed irrigation. These findings suggest that, even with political conflicts in semi-arid areas already increasing, sustainable irrigation may offer a valuable tool to improve communities' long-term well-being and social cohesion.

Case Report: Phantom limb pain relief after cognitive multisensory rehabilitation.

Introduction: Relieving phantom limb pain (PLP) after amputation in patients resistant to conventional therapy remains a challenge. While the causes for PLP are unclear, one model suggests that maladaptive plasticity related to cortical remapping following amputation leads to altered mental body representations (MBR) and contributes to PLP. Cognitive Multisensory Rehabilitation (CMR) has led to reduced pain in other neurologic conditions by restoring MBR. This is the first study using CMR to relieve PLP. Methods: A 26-year-old woman experienced excruciating PLP after amputation of the third proximal part of the leg, performed after several unsuccessful treatments (i.e., epidural stimulator, surgeries, analgesics) for debilitating neuropathic pain in the left foot for six years with foot deformities resulting from herniated discs. The PLP was resistant to pain medication and mirror therapy. PLP rendered donning a prosthesis impossible. The patient received 35 CMR sessions (2×/day during weekdays, October-December 2012). CMR provides multisensory discrimination exercises on the healthy side and multisensory motor imagery exercises of present and past actions in both limbs to restore MBR and reduce PLP. Results: After CMR, PLP reduced from 6.5-9.5/10 to 0/10 for neuropathic pain with only 4-5.5/10 for muscular pain after exercising on the Numeric Pain Rating Scale. McGill Pain Questionnaire scores reduced from 39/78 to 5/78, and Identity (ID)-Pain scores reduced from 5/5 to 0/5. Her pain medication was reduced by at least 50% after discharge. At 10-month follow-up (9/2013), she no longer took Methadone or Fentanyl. After discharge, receiving CMR as outpatient, she learned to walk with a prosthesis, and gradually did not need crutches anymore to walk independently indoors and outdoors (9/2013). At present (3/2024), she no longer takes pain medication and walks independently with the prosthesis without assistive devices. PLP is under control. She addresses flare-ups with CMR exercises on her own, using multisensory motor imagery, bringing the pain down within 10-15 min. Conclusion: The case study seems to support the hypothesis that CMR restores MBR which may lead to long-term (12-year) PLP reduction. MBR restoration may be linked to restoring accurate multisensory motor imagery of the remaining and amputated limb regarding present and past actions.

Family dementia caregivers with suicidal ideation improve with mentalizing imagery therapy: Results from a pilot study.

Background: Family caregivers of persons living with dementia often experience increased depression and suicidal ideation (SI). However, the feasibility and impact of therapies on caregiver SI has remained largely unexplored. Mentalizing imagery therapy (MIT) helps reduce psychological symptoms through mindfulness and guided imagery. This pilot study examined the feasibility of participation by caregivers with SI in a randomized controlled trial (RCT) of MIT versus a psychosocial support group (SG), and the respective impact of group on SI, depression, and secondary outcomes. Methods: A secondary analysis of data from an RCT of 4-week MIT or SG for caregivers (n = 46) was performed, identifying SI (n = 23) and non-SI (n = 23) cohorts. Group attendance and home practice were compared between cohorts. In the SI cohort (total n = 23, MIT n = 11, SG n = 12), group differences in SI, depression, and secondary outcomes were evaluated post-group and at 4-month follow-up. Results: Attendance in both groups and home practice in MIT were similar between SI and non-SI cohorts. In the SI cohort, MIT evinced greater improvements relative to SG in SI (p=.02) and depression (p=.02) post-group, and other secondary outcomes at follow-up. Limitations: Limitations include small sample size and single-item assessments of SI from validated depression rating scales. Conclusions: Participation in an RCT was feasible for caregivers with SI. MIT resulted in important benefits for SI and depression, while SG showed no acute SI benefit. The role of MIT in improving SI should be confirmed with adequately powered trials, as effective therapies to address caregiver SI are critical.

Greater utilization of executive functions networks when listening to stories with visual stimulation is related to lower reading abilities in children.

Narrative comprehension relies on basic sensory processing abilities, such as visual and auditory processing, with recent evidence for utilizing executive functions (EF), which are also engaged during reading. EF was previously related to the "supporter" of engaging the auditory and visual modalities in different cognitive tasks, with evidence of lower efficiency in this process among those with reading difficulties in the absence of a visual stimulus (i.e. while listening to stories). The current study aims to fill out the gap related to the level of reliance on these neural circuits while visual aids (pictures) are involved during story listening in relation to reading skills. Functional MRI data were collected from 44 Hebrew-speaking children aged 8-12 years while listening to stories with vs without visual stimuli (i.e., pictures). Functional connectivity of networks supporting reading was defined in each condition and compared between the conditions against behavioral reading measures. Lower reading skills were related to greater functional connectivity values between EF networks (default mode and memory networks), and between the auditory and memory networks for the stories with vs without the visual stimulation. A greater difference in functional connectivity between the conditions was related to lower reading scores. We conclude that lower reading skills in children may be related to a need for greater scaffolding, i.e., visual stimulation such as pictures describing the narratives when listening to stories, which may guide future intervention approaches.

Attention-based convolutional neural network with multi-modal temporal information fusion for motor imagery EEG decoding.

Convolutional neural network (CNN) has been widely applied in motor imagery (MI)-based brain computer interface (BCI) to decode electroencephalography (EEG) signals. However, due to the limited perceptual field of convolutional kernel, CNN only extracts features from local region without considering long-term dependencies for EEG decoding. Apart from long-term dependencies, multi-modal temporal information is equally important for EEG decoding because it can offer a more comprehensive understanding of the temporal dynamics of neural processes. In this paper, we propose a novel deep learning network that combines CNN with self-attention mechanism to encapsulate multi-modal temporal information and global dependencies. The network first extracts multi-modal temporal information from two distinct perspectives: average and variance. A shared self-attention module is then designed to capture global dependencies along these two feature dimensions. We further design a convolutional encoder to explore the relationship between average-pooled and variance-pooled features and fuse them into more discriminative features. Moreover, a data augmentation method called signal segmentation and recombination is proposed to improve the generalization capability of the proposed network. The experimental results on the BCI Competition IV-2a (BCIC-IV-2a) and BCI Competition IV-2b (BCIC-IV-2b) datasets show that our proposed method outperforms the state-of-the-art methods and achieves 4-class average accuracy of 85.03% on the BCIC-IV-2a dataset. The proposed method implies the effectiveness of multi-modal temporal information fusion in attention-based deep learning networks and provides a new perspective for MI-EEG decoding. The code is available at https://github.com/Ma-Xinzhi/EEG-TransNet.

Authentication with a one-dimensional CNN model using EEG-based brain-computer interface.

Brain-computer interface (BCI) technology uses electroencephalogram (EEG) signals to create a direct interaction between the human body and its surroundings. Motor imagery (MI) classification using EEG signals is an important application that can help a rehabilitated or motor-impaired stroke patient perform certain tasks. Robust classification of these signals is an important step toward making the use of EEG more practical in many applications and less dependent on trained professionals. Deep learning methods have produced impressive results in BCI in recent years, especially with the availability of large electroencephalography (EEG) data sets. Dealing with EEG-MI signals is difficult because noise and other signal sources can interfere with the electrical amplitude of the brain, and its generalization ability is limited, so it is difficult to improve EEG classifiers. To address these issues, this paper presents a methodology based on one-dimensional convolutional neural networks (1-D CNN) for motor imagery (MI) recognition for the right hand, left hand, feet, and sedentary task. The proposed model is a lightweight model with fewer parameters and has an accuracy of 91.75%. Then, in an innovative exploitation of the four output classes, there is an idea that allows people with disabilities who are deprived of security measures, such as entering a secret code, to use the output classification, such as password codes. It is also an idea for a unique authentication system that is more secure and less vulnerable to theft or the like for a healthy person at the same time.

Multi-source deep domain adaptation ensemble framework for cross-dataset motor imagery EEG transfer learning.

OBJECTIVE: Electroencephalography (EEG) is an important kind of bioelectric signal for measuring physiological activities of the brain, and motor imagery (MI) EEG has significant clinical application prospects. Convolutional neural network (CNN) has become a mainstream algorithm for MI EEG classification, however lack of subject-specific data considerably restricts its decoding accuracy and generalization performance. To address this challenge, a novel transfer learning (TL) framework using auxiliary dataset to improve the MI EEG classification performance of target subject is proposed in this paper. APPROACH: We developed a multi-source deep domain adaptation ensemble framework (MSDDAEF) for cross-dataset MI EEG decoding. The proposed MSDDAEF comprises three main components: model pre-training, deep domain adaptation, and multi-source ensemble. Moreover, for each component, different designs were examined to verify the robustness of MSDDAEF. MAIN RESULTS: Bidirectional validation experiments were performed on two large public MI EEG datasets (openBMI and GIST). The highest average classification accuracy of MSDDAEF reaches 74.28% when openBMI serves as target dataset and GIST serves as source dataset. While the highest average classification accuracy of MSDDAEF is 69.85% when GIST serves as target dataset and openBMI serves as source dataset. In addition, the classification performance of MSDDAEF surpasses several well-established studies and state-of-the-art algorithms. SIGNIFICANCE: The results of this study show that cross-dataset TL is feasible for left/right-hand MI EEG decoding, and further indicate that MSDDAEF is a promising solution for addressing MI EEG cross-dataset variability.

Continuous tracking using deep learning-based decoding for noninvasive brain-computer interface.

Brain-computer interfaces (BCI) using electroencephalography provide a noninvasive method for users to interact with external devices without the need for muscle activation. While noninvasive BCIs have the potential to improve the quality of lives of healthy and motor-impaired individuals, they currently have limited applications due to inconsistent performance and low degrees of freedom. In this study, we use deep learning (DL)-based decoders for online continuous pursuit (CP), a complex BCI task requiring the user to track an object in 2D space. We developed a labeling system to use CP data for supervised learning, trained DL-based decoders based on two architectures, including a newly proposed adaptation of the PointNet architecture, and evaluated the performance over several online sessions. We rigorously evaluated the DL-based decoders in a total of 28 human participants, and found that the DL-based models improved throughout the sessions as more training data became available and significantly outperformed a traditional BCI decoder by the last session. We also performed additional experiments to test an implementation of transfer learning by pretraining models on data from other subjects, and midsession training to reduce intersession variability. The results from these experiments showed that pretraining did not significantly improve performance, but updating the models' midsession may have some benefit. Overall, these findings support the use of DL-based decoders for improving BCI performance in complex tasks like CP, which can expand the potential applications of BCI devices and help to improve the quality of lives of healthy and motor-impaired individuals.

Virtual Reality Action Observation and Motor Imagery to Enhance Neuroplastic Capacity in the Human Motor Cortex: A Pilot Double-blind, Randomized Cross-over Trial.

Neuroplasticity is important for learning, development and recovery from injury. Therapies that can upregulate neuroplasticity are therefore of interest across a range of fields. We developed a novel virtual reality action observation and motor imagery (VR-AOMI) intervention and evaluated whether it could enhance the efficacy of mechanisms of neuroplasticity in the human motor cortex of healthy adults. A secondary question was to explore predictors of the change in neuroplasticity following VR-AOMI. A pre-registered, pilot randomized controlled cross-over trial was performed. Twenty right-handed adults (13 females; mean age: 23.0 ±â€¯4.53 years) completed two experimental conditions in separate sessions; VR-AOMI and control. We used intermittent theta burst stimulation (iTBS) to induce long term potentiation-like plasticity in the motor cortex and recorded motor evoked potentials at multiple timepoints as a measure of corticospinal excitability. The VR-AOMI task did not significantly increase the change in MEP amplitude following iTBS when compared to the control task (Group × Timepoint interaction p = 0.17). However, regression analysis identified the change in iTBS response following VR-AOMI was significantly predicted by the baseline iTBS response in the control task. Specifically, participants that did not exhibit the expected increase in MEP amplitude following iTBS in the control condition appear to have greater excitability following iTBS in the VR-AOMI condition (r = -0.72, p < 0.001). Engaging in VR-AOMI might enhance capacity for neuroplasticity in some people who typically do not respond to iTBS. VR-AOMI may prime the brain for enhanced neuroplasticity in this sub-group.

Nature's beauty versus urban bustle: Chinese folk music influences food choices by inducing mental imagery of different scenes.

Previous research has demonstrated that music can impact people's food choices by triggering emotional states. We reported two virtual reality (VR) experiments designed to examine how Chinese folk music influences people's food choices by inducing mental imagery of different scenes. In both experiments, young healthy Chinese participants were asked to select three dishes from an assortment of two meat and two vegetable dishes while listening to Chinese folk music that could elicit mental imagery of nature or urban scenes. The results of Experiment 1 revealed that they chose vegetable-forward meals more frequently while listening to Chinese folk music eliciting mental imagery of nature versus urban scenes. In Experiment 2, the participants were randomly divided into three groups, in which the prevalence of their mental imagery was enhanced, moderately suppressed, or strongly suppressed by performing different tasks while listening to the music pieces. We replicated the results of Experiment 1 when the participants' mental imagery was enhanced, whereas no such effect was observed when the participants' mental imagery was moderately or strongly suppressed. Collectively, these findings suggest that music may influence the food choices people make in virtual food choice tasks by inducing mental imagery, which provides insights into utilizing environmental cues to promote healthier food choices.

Trait food craving predicts functional connectivity between dopaminergic midbrain and the fusiform food area during eating imagery.

Neurofunctional coupling between the dopaminergic midbrain (i.e., ventral tegmental area, VTA) and higher-order visual regions may contribute to food craving, leading to the onset or maintenance of obesity. We recently showed that the VTA resting-state functional connectivity with the occipitotemporal cortex, at the level of the fusiform gyrus (FFG), was specifically associated with trait food craving and the implicit bias for food images, suggesting that VTA-FFG connectivity may reflect the association between the visual representations of food and its motivational properties. To further test this hypothesis, this time we studied task-based functional connectivity in twenty-eight healthy-weight participants while imagining eating their most liked high-calorie (HC) or least liked low-calorie food (LC) or drinking water (control condition). Trait food craving scores were used to predict changes in task-based functional connectivity of the VTA during imagery of HC compared to LC foods (relative to the control condition). Trait food craving was positively associated with the functional connectivity of the VTA with the left FFG: people with higher trait food craving scores show stronger VTA-FFG connectivity, specifically for the imagery of the liked HC foods. This association was not linked to the quality of imagery nor to state measures of craving, appetite, or thirst. These findings emphasize the contribution of the functional coupling between dopaminergic midbrain and higher-order visual regions to food craving, suggesting a neurofunctional mechanism by which the mental representations of the HC food we like can become much more salient if not irresistible.

Interventions Targeting Negative Mental Imagery in Social Anxiety: A Systematic Review and Meta-Analysis of Characteristics and Outcomes.

Psychological treatment for social anxiety disorder (SAD) has been found to be less effective than for other anxiety disorders. Targeting the vivid and distressing negative mental images typically experienced by individuals with social anxiety could possibly enhance treatment effectiveness. To provide both clinicians and researchers with an overview of current applications, this systematic review and meta-analysis aimed to evaluate the possibilities and effects of imagery-based interventions that explicitly target negative images in (sub)clinical social anxiety. Based on a prespecified literature search, we included 21 studies, of which 12 studies included individuals with a clinical diagnosis of SAD. Imagery interventions (k = 28 intervention groups; only in adults) generally lasted one or two sessions and mostly used imagery rescripting with negative memories. Others used eye movement desensitization and reprocessing and imagery exposure with diverse intrusive images. Noncontrolled effects on social anxiety, imagery distress and imagery vividness were mostly large or medium. Meta-analyses with studies with control groups resulted in significant medium controlled effects on social anxiety (d = -0.50, k = 10) and imagery distress (d = -0.64, k = 8) and a nonsignificant effect on imagery vividness. Significant controlled effects were most evident in individuals with clinically diagnosed versus subclinical social anxiety. Overall, findings suggest promising effects of sessions targeting negative mental images. Limitations of the included studies and the analyses need to be considered. Future research should examine the addition to current SAD treatments and determine the relevance of specific imagery interventions. Studies involving children and adolescents are warranted.

Self-supervised motor imagery EEG recognition model based on 1-D MTCNN-LSTM network.

Objective.Aiming for the research on the brain-computer interface (BCI), it is crucial to design a MI-EEG recognition model, possessing a high classification accuracy and strong generalization ability, and not relying on a large number of labeled training samples.Approach.In this paper, we propose a self-supervised MI-EEG recognition method based on self-supervised learning with one-dimensional multi-task convolutional neural networks and long short-term memory (1-D MTCNN-LSTM). The model is divided into two stages: signal transform identification stage and pattern recognition stage. In the signal transform recognition phase, the signal transform dataset is recognized by the upstream 1-D MTCNN-LSTM network model. Subsequently, the backbone network from the signal transform identification phase is transferred to the pattern recognition phase. Then, it is fine-tuned using a trace amount of labeled data to finally obtain the motion recognition model.Main results.The upstream stage of this study achieves more than 95% recognition accuracy for EEG signal transforms, up to 100%. For MI-EEG pattern recognition, the model obtained recognition accuracies of 82.04% and 87.14% with F1 scores of 0.7856 and 0.839 on the datasets of BCIC-IV-2b and BCIC-IV-2a.Significance.The improved accuracy proves the superiority of the proposed method. It is prospected to be a method for accurate classification of MI-EEG in the BCI system.

Classification of motor imagery using chaotic entropy based on sub-band EEG source localization.

Objective.Electroencephalography (EEG) has been widely used in motor imagery (MI) research by virtue of its high temporal resolution and low cost, but its low spatial resolution is still a major criticism. The EEG source localization (ESL) algorithm effectively improves the spatial resolution of the signal by inverting the scalp EEG to extrapolate the cortical source signal, thus enhancing the classification accuracy.Approach.To address the problem of poor spatial resolution of EEG signals, this paper proposed a sub-band source chaotic entropy feature extraction method based on sub-band ESL. Firstly, the preprocessed EEG signals were filtered into 8 sub-bands. Each sub-band signal was source localized respectively to reveal the activation patterns of specific frequency bands of the EEG signals and the activities of specific brain regions in the MI task. Then, approximate entropy, fuzzy entropy and permutation entropy were extracted from the source signal as features to quantify the complexity and randomness of the signal. Finally, the classification of different MI tasks was achieved using support vector machine.Main result.The proposed method was validated on two MI public datasets (brain-computer interface (BCI) competition III IVa, BCI competition IV 2a) and the results showed that the classification accuracies were higher than the existing methods.Significance.The spatial resolution of the signal was improved by sub-band EEG localization in the paper, which provided a new idea for EEG MI research.

EEG motor imagery decoding: a framework for comparative analysis with channel attention mechanisms.

Objective.The objective of this study is to investigate the application of various channel attention mechanisms within the domain of brain-computer interface (BCI) for motor imagery decoding. Channel attention mechanisms can be seen as a powerful evolution of spatial filters traditionally used for motor imagery decoding. This study systematically compares such mechanisms by integrating them into a lightweight architecture framework to evaluate their impact.Approach.We carefully construct a straightforward and lightweight baseline architecture designed to seamlessly integrate different channel attention mechanisms. This approach is contrary to previous works which only investigate one attention mechanism and usually build a very complex, sometimes nested architecture. Our framework allows us to evaluate and compare the impact of different attention mechanisms under the same circumstances. The easy integration of different channel attention mechanisms as well as the low computational complexity enables us to conduct a wide range of experiments on four datasets to thoroughly assess the effectiveness of the baseline model and the attention mechanisms.Results.Our experiments demonstrate the strength and generalizability of our architecture framework as well as how channel attention mechanisms can improve the performance while maintaining the small memory footprint and low computational complexity of our baseline architecture.Significance.Our architecture emphasizes simplicity, offering easy integration of channel attention mechanisms, while maintaining a high degree of generalizability across datasets, making it a versatile and efficient solution for electroencephalogram motor imagery decoding within BCIs.