This study aims to compare brain activity differences under the motor imagery-brain-computer interface (MI-BCI), motor imagery (MI), and resting (REST) paradigms through EEG microstate and functional connectivity (FC) analysis, providing a theoretical basis for applying MI-BCI in the rehabilitation of children with cerebral palsy (CP). This study included 30 subjects aged 4-6 years with GMFCS II-III grade, diagnosed with CP and classified as spastic diplegia. They sequentially completed EEG signal acquisition under REST, MI, and MI-BCI conditions. Clustering analysis was used to analyze EEG microstates and extract EEG microstate temporal parameters. Additionally, the strength of brain FC in different frequency bands was analyzed to compare the differences under various conditions. Four microstate classes (A-D) were identified to best explain the datasets of three groups. Compared to REST, the average duration and coverage rate of microstate D under MI and MI-BCI significantly increased (P < 0.05), while their frequency and the coverage rate and frequency of microstate A decreased. Compared to MI, the average duration of microstate C under MI-BCI significantly decreased (P < 0.05), while the frequency of microstate B significantly increased (P < 0.05). Additionally, the transition probability results showed that other microstates under REST had a higher transition probability to microstate A, while under MI and MI-BCI, other microstates had a higher transition probability to microstate D. The brain network results revealed significant differences in brain network connectivity among REST, MI, and MI-BCI across different frequency bands. No FC differences were found between REST, MI, and MI-BCI in the α2 frequency band. In the δ and γ frequency bands, MI and MI-BCI both had greater inter-electrode connectivity strength than REST. In the θ frequency band, REST had greater inter-electrode connectivity strength than MI-BCI, while MI-BCI had greater inter-electrode connectivity strength than both REST and MI. In the α1 frequency band, MI-BCI had greater inter-electrode connectivity strength than REST, and in the β frequency band, MI-BCI had greater inter-electrode connectivity strength than MI. MI-BCI can significantly alter the brain activity patterns of children with CP, particularly by enhancing the activity intensity of EEG microstates related to attention, motor planning, and execution, as well as the brain FC strength in different frequency bands. It holds high application value in the lower limb motor rehabilitation of children with CP.

Access to early intervention and rehabilitation services among children with cerebral palsy (CP) remains limited in Bangladesh, which demands an innovative and sustainable service delivery model.

To evaluate the effectiveness of the Sustainable Model of Early Intervention and Telerehabilitation for Children With CP (SMART-CP) model compared with usual care in improving access to and utilization of early diagnosis, early intervention, and rehabilitation services in rural Bangladesh.

This was a 2-arm cluster randomized clinical trial, with 8 clusters (ie, subdistricts) randomly allocated to the intervention (SMART-CP model) or control arm. The setting was in Sirajganj, Bangladesh, and included children with CP 18 years or younger. Outcomes were measured at 0 and 12 months, and an intention-to-treat analysis was conducted. Data were analyzed from December 2023 to May 2024.

The SMART-CP model comprised (1) a rural referral network involving key informants and caregiver peer groups (called mPower or mothers' power), (2) subdistrict level SMART-CP centers, and (3) telerehabilitation services. Children in the intervention arm received weekly goal-directed therapy, mPower group meetings every 2 weeks, and monthly telerehabilitation sessions.

The primary outcome was whether a child with CP accessed any form of rehabilitation services, with secondary outcomes analyzed as hypothesis generating.

Overall, 968 children with CP (mean [SD] age, 7.9 [4.9] years; 581 male [60.0%]) were enrolled, with 500 in the intervention arm and 468 in the control arm. Between baseline and endline, rehabilitation services uptake significantly increased in the intervention arm (70.2% [351 of 500] vs 99.4% [497 of 500]), compared with the control arm (63.9% [299 of 468] vs 68.2% [319 of 468]; P <.001). Children in the intervention arm were 1.5 times more likely to access rehabilitation than the control arm. Secondary analyses suggested that the intervention arm also facilitated early CP diagnosis (mean [SD] diagnosis time, 2.0 [2.0] years vs 3.8 [3.3] years; Cohen d = -0.7) and initiation of rehabilitation (mean [SD] rehabilitation time, 1.8 [1.8] years vs 3.6 [2.4] years; Cohen d = -0.9). Additionally, higher therapy session counts (mean [SD] session counts, 23.4 [31.7] vs 4.3 [20.8]; Cohen d = 0.7), increased assistive device utilization (20.8% [104 of 500] vs 3.0% [14 of 468]; risk ratio, 0.82; 95% CI, 0.78-0.86; P < .001), and lower out-of-pocket expenditure per month (mean [SD] expenditure, $1.5 [$1.6] vs $2.9 [$5.1]; Cohen d = -0.4) were found in the intervention arm. No significant difference in clinical outcomes and mortality rates was observed between the intervention and control groups.

Results of this cluster randomized clinical trial reveal that the SMART-CP model improved access to and utilization of early diagnosis and intervention services for children with CP in rural Bangladesh. This model holds promise for global scalability.

ANZCTR Trial Identifier: ACTRN12622000396729.

Mental fatigue during long-term motor imagery (MI) may affect intention recognition in MI applications. However, the current research lacks the monitoring of mental fatigue during MI and the definition of robust biomarkers. The present study aims to reveal the effects of mental fatigue on motor imagery recognition at the brain region level and explore biomarkers of mental fatigue. To achieve this, we recruited 10 healthy participants and asked them to complete a long-term motor imagery task involving both right- and left-handed movements. During the experiment, we recorded 32-channel EEG data and carried out a fatigue questionnaire for each participant. As a result, we found that mental fatigue significantly decreased the subjects' motor imagery recognition rate during MI. Additionally the theta power of frontal, central, parietal, and occipital clusters significantly increased after the presence of mental fatigue. Furthermore, the phase synchronization between the central cluster and the frontal and occipital lobes was significantly weakened. To summarize, the theta bands of frontal, central, and parieto-occipital clusters may serve as powerful biomarkers for monitoring mental fatigue during motor imagery. Additionally, changes in functional connectivity between the central cluster and the prefrontal and occipital lobes during motor imagery could be investigated as potential biomarkers.

A motor imagery brain-computer interface connects the human brain and computers via electroencephalography (EEG). However, individual differences in the frequency ranges of brain activity during motor imagery tasks pose a challenge, limiting the manual feature extraction for motor imagery classification. To extract features that match specific subjects, we proposed a novel motor imagery classification model using distinctive feature fusion with adaptive structural LASSO. Specifically, we extracted spatial domain features from overlapping and multi-scale sub-bands of EEG signals and mined discriminative features by fusing the task relevance of features with spatial information into the adaptive LASSO-based feature selection. We evaluated the proposed model on public motor imagery EEG datasets, demonstrating that the model has excellent performance. Meanwhile, ablation studies and feature selection visualization of the proposed model further verified the great potential of EEG analysis.

Motor-Imagery-based Brain-Computer Interface (MI-BCI) is a promising technology to assist communication, movement, and neurological rehabilitation for motor-impaired individuals. Electroencephalography (EEG) decoding techniques using deep learning (DL) possess noteworthy advantages due to automatic feature extraction and end-to-end learning. However, the DL-based EEG decoding models tend to show large variations due to intersubject variability of EEG, which results from inconsistencies of different subjects' optimal hyperparameters.

This study proposes a multi-branch multi-attention mechanism EEGNet model (MBMANet) for robust decoding. It applies the multi-branch EEGNet structure to achieve various feature extractions. Further, the different attention mechanisms introduced in each branch attain diverse adaptive weight adjustments. This combination of multi-branch and multi-attention mechanisms allows for multi-level feature fusion to provide robust decoding for different subjects.

The MBMANet model has a four-classification accuracy of 83.18% and kappa of 0.776 on the BCI Competition IV-2a dataset, which outperforms other eight CNN-based decoding models. This consistently satisfactory performance across all nine subjects indicates that the proposed model is robust.

The combine of multi-branch and multi-attention mechanisms empowers the DL-based models to adaptively learn different EEG features, which provides a feasible solution for dealing with data variability. It also gives the MBMANet model more accurate decoding of motion intentions and lower training costs, thus improving the MI-BCI's utility and robustness.

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.

Cerebral palsy (CP) describes a group of disorders affecting movement, balance, and posture. Disturbances in motor functions constitute the main body of CP symptoms. These symptoms surface in early childhood and patients are affected for the rest of their lives. Currently, treatment involves various pharmacotherapies for different types of CP, including antiepileptics for epilepsy and Botox A for focal spasticity. However, none of these methods can provide full symptom relief. This has prompted researchers to look for new treatment modalities, one of which is mesenchymal stem cell therapy (MSCT). Despite being a promising tool and offering a wide array of possibilities, mesenchymal stem cells (MSCs) still need to be investigated for their efficacy and safety.

To analyze the efficacy and safety of MSCT in CP patients.

Our sample consists of four CP patients who cannot stand or walk without external support. All of these cases received allogeneic MSCT six times as 1 × 106/kg intrathecally, intravenously, and intramuscularly using umbilical cord-derived MSCs (UC-MSC). We monitored and assessed the patients pre- and post-treatment using the Wee Functional Independence Measure (WeeFIM), Gross Motor Function Classification System (GMFCS), and Manual Ability Classification Scale (MACS) instruments. We utilized the Modified Ashworth Scale (MAS) to measure spasticity.

We found significant improvements in MAS scores after the intervention on both sides. Two months: Right χ2 = 4000, P = 0.046, left χ2 = 4000, P = 0.046; four months: Right χ2 = 4000, P = 0.046, left χ2 = 4000, P = 0.046; 12 months: Right χ2 = 4000, P = 0.046, left χ2 = 4000, P = 0.046. However, there was no significant difference in motor functions based on WeeFIM results (P > 0.05). GMFCS and MACS scores differed significantly at 12 months after the intervention (P = 0.046, P = 0.046). Finally, there was no significant change in cognitive functions (P > 0.05).

In light of our findings, we believe that UC-MSC therapy has a positive effect on spasticity, and it partially improves motor functions.

Motor Imagery (MI) is a cognitive process consisting in mental simulation of body movements without executing physical actions: its clinical use has been investigated prevalently in adults with neurological disorders.

Review of the best-available evidence on the use and efficacy of MI interventions for neurorehabilitation purposes in common and rare childhood neurological disorders.

systematic literature search conducted according to PRISMA by using the Scopus, PsycArticles, Cinahl, PUBMED, Web of Science (Clarivate), EMBASE, PsychINFO, and COCHRANE databases, with levels of evidence scored by OCEBM and PEDro Scales.

Twenty-two original studies were retrieved and included for the analysis; MI was the unique or complementary rehabilitative treatment in 476 individuals (aged 5 to 18 years) with 10 different neurological conditions including, cerebral palsies, stroke, coordination disorders, intellectual disabilities, brain and/or spinal cord injuries, autism, pain syndromes, and hyperactivity. The sample size ranged from single case reports to cohorts and control groups. Treatment lasted 2 days to 6 months with 1 to 24 sessions. MI tasks were conventional, graded or ad-hoc. MI measurement tools included movement assessment batteries, mental chronometry tests, scales, and questionnaires, EEG, and EMG. Overall, the use of MI was stated as effective in 19/22, and uncertain in the remnant studies.

MI could be a reliable supportive/add-on (home-based) rehabilitative tool for pediatric neurorehabilitation; its clinical use, in children, is highly dependent on the complexity of MI mechanisms, which are related to the underlying neurodevelopmental disorder.

Currently, there is no effective treatment for pediatric patients with complete spinal cord injury. Motor imagery has been proposed as an alternative to physical training for patients who are unable to move voluntarily. Our aim was to reveal the potential mechanism of motor imagery in the rehabilitation of pediatric complete spinal cord injury.

Twenty-six pediatric patients with complete spinal cord injury and 26 age- and sex-matched healthy children as healthy controls were recruited. All participants underwent the motor imagery task-related fMRI scans, and additional motor execution scans were performed only on healthy controls. First, we compared the brain-activation patterns between motor imagery and motor execution in healthy controls. Then, we compared the brain activation of motor imagery between the 2 groups and compared the brain activation of motor imagery in pediatric patients with complete spinal cord injury and that of motor execution in healthy controls.

In healthy controls, compared with motor execution, motor imagery showed increased activation in the left inferior parietal lobule and decreased activation in the left supplementary motor area, paracentral lobule, middle cingulate cortex, and right insula. In addition, our results revealed that the 2 groups both activated the bilateral supplementary motor area, middle cingulate cortex and left inferior parietal lobule, and supramarginal gyrus during motor imagery. Compared with healthy controls, higher activation in the bilateral paracentral lobule, supplementary motor area, putamen, and cerebellar lobules III-V was detected in pediatric complete spinal cord injury during motor imagery, and the activation of these regions was even higher than that of healthy controls during motor execution.

Our study demonstrated that part of the motor imagery network was functionally preserved in pediatric complete spinal cord injury and could be activated through motor imagery. In addition, higher-level activation in sensorimotor-related regions was also found in pediatric complete spinal cord injury during motor imagery. Our findings may provide a theoretic basis for the application of motor imagery training in pediatric complete spinal cord injury.

Patients with complete spinal cord injury have a complete loss of motor and sensory functions below the injury plane, leading to a complete loss of function of the nerve pathway in the injured area. Improving the microenvironment in the injured area of patients with spinal cord injury, promoting axon regeneration of the nerve cells is challenging research fields. The brain-computer interface rehabilitation system is different from the other rehabilitation techniques. It can exert bidirectional stimulation on the spinal cord injury area, and can make positively rehabilitation effects of the patient with complete spinal cord injury. A dynamic model was constructed for the patient with spinal cord injury under-stimulation therapy, and the mechanism of the brain-computer interface in rehabilitation training was explored. The effects of the three current rehabilitation treatment methods on the microenvironment in a microscopic nonlinear model were innovatively unified and a complex system mapping relationship from the microscopic axon growth to macroscopic motor functions was constructed. The basic structure of the model was determined by simulating and fitting the data of the open rat experiments. A clinical rehabilitation experiment of spinal cord injury based on brain-computer interface was built, recruiting a patient with complete spinal cord injury, and the rehabilitation training and follow-up were conducted. The changes in the motor function of the patient was simulated and predicted through the constructed model, and the trend in the motor function improvement was successfully predicted over time. This proposed model explores the mechanism of brain-computer interface in rehabilitating patients with complete spinal cord injury, and it is also an application of complex system theory in rehabilitation medicine.

The online version contains supplementary material available at 10.1007/s11571-022-09804-3.

The clinical therapy of schizophrenia (SCZ) replies on the corresponding accurate and reliable recognition. Although efforts have been paid, the diagnosis of SCZ is still roughly subjective, it is thus urgent to search for related objective physiological parameters. Motivated by the great potential of resting-state networks in underling the brain deficits among different SCZ groups, in this study, we then developed a multi-class feature extraction approach that could effectively extract the spatial network topology and facilitate the recognition of the SCZ, by combining a network structure based supervised learning with an ensemble co-decision strategy. The results demonstrated that the multi-class spatial pattern of the network (MSPN) features outperformed the other conventional electrophysiological features, such as relative power spectrums and network properties, and achieved the highest classification accuracy of 71.58% in the alpha band. These findings did validate that the resting-state MSPN is a promising tool for the clinical assessment of the SCZ.

Investigation of the temporal variability of resting-state brain networks informs our understanding of how neural connectivity aggregates and disassociates over time, further shedding light on the aberrant neural interactions that underlie symptomatology and psychosis development. In the current work, an electroencephalogram-based sliding window analysis was utilized for the first time to measure the nonlinear complexity of dynamic resting-state brain networks of schizophrenia (SZ) patients by applying fuzzy entropy. The results of this study demonstrated the attenuated temporal variability among multiple electrodes that were distributed in the frontal and right parietal lobes for SZ patients when compared with healthy controls (HCs). Meanwhile, a concomitant strengthening of the posterior and peripheral flexible connections that may be attributed to the excessive alertness or sensitivity of SZ patients to the external environment was also revealed. These temporal fluctuation distortions combined reflect an abnormality in the coordination of functional network switching in SZ, which is further the source of worse task performance (i.e., P300 amplitude) and the negative relationship between individual complexity metrics and P300 amplitude. Notably, when using the network metrics as features, multiple linear regressions of P300 amplitudes were also exactly achieved for both the SZ and HC groups. These findings shed light on the pathophysiological mechanisms of SZ from a temporal variability perspective and provide potential biomarkers for quantifying SZ's progressive neurophysiological deterioration.

Hemiplegia is a common dysfunction caused by the brain stroke and leads to movement disability. Although the lateralization of movement-related potential, the event-related desynchronization, and more complicated inter-regional information coupling have been investigated, seldom studies have focused on investigating the dynamic information exchanging among multiple brain regions during motor execution for post-stroke hemiplegic patients. With high temporal-resolution electroencephalogram (EEG), the time-varying network is able to reflect the dynamical complex network modalities corresponding to the movements at a millisecond level. In our present study, the wrist extension experiment was designed, along with related EEG datasets being collected. Thereafter, the corresponding time-varying networks underlying the wrist extension were accordingly constructed by adopting the adaptive directed transfer function and then statistically explored, to further uncover the dynamic network deficits (i.e., motor dysfunction) in post-stroke hemiplegic patients. Results of this study found the effective connectivity between the stroked motor area and other areas decreased in patients when compared to healthy controls; on the contrary, the enhanced connectivity between non-stroked motor areas and other areas, especially the frontal and parietal-occipital lobes, were further identified for patients during their accomplishing the designed wrist extension, which might dynamically compensate for the deficited patients' motor behaviors. These findings not only helped deepen our knowledge of the mechanism underlying the patients' motor behaviors, but also facilitated the real-time strategies for clinical therapy of brain stroke, as well as providing a reliable biomarker to predict the future rehabilitation.

The online version contains supplementary material available at 10.1007/s11571-021-09738-2.