Beta-band oscillations as a biomarker of gait recovery in spinal cord injury patients: A quantitative electroencephalography analysis

Gait Training-Based Motor Imagery and EEG Neurofeedback in Lokomat: A Clinical Intervention With Complete Spinal Cord Injury Individuals

Robotic systems, such as Lokomat® have shown promising results in people with severe motor impairments, who suffered a stroke or other neurological damage. Robotic devices have also been used by people with more challenging damages, such as Spinal Cord Injury (SCI), using feedback strategies that provide information about the brain activity in real-time. This study proposes a novel Motor Imagery (MI)-based Electroencephalogram (EEG) Visual Neurofeedback (VNFB) system for Lokomat® to teach individuals how to modulate their own μ (8-12 Hz) and β (15-20 Hz) rhythms during passive walking. Two individuals with complete SCI tested our VNFB system completing a total of 12 sessions, each on different days. For evaluation, clinical outcomes before and after the intervention and brain connectivity were analyzed. As findings, the sensitivity related to light touch and painful discrimination increased for both individuals. Furthermore, an improvement in neurogenic bladder and bowel functions was observed according to the American Spinal Injury Association Impairment Scale, Neurogenic Bladder Symptom Score, and Gastrointestinal Symptom Rating Scale. Moreover, brain connectivity between different EEG locations significantly ( [Formula: see text]) increased, mainly in the motor cortex. As other highlight, both SCI individuals enhanced their μ rhythm, suggesting motor learning. These results indicate that our gait training approach may have substantial clinical benefits in complete SCI individuals.

Aided diagnosis of cervical spondylotic myelopathy using deep learning methods based on electroencephalography

Cervical spondylotic myelopathy (CSM) is the most severe type of cervical spondylosis. It is challenging to achieve early diagnosis with current clinical diagnostic tools. In this paper, we propose an end-to-end deep learning approach for early diagnosis of CSM. Electroencephalography (EEG) experiments were conducted with patients having spinal cord cervical spondylosis and age-matched normal subjects. A Convolutional Neural Network with Long Short-Term Memory Networks (CNN-LSTM) model was employed for the classification of patients versus normal individuals. In contrast, a Convolutional Neural Network with Bidirectional Long Short-Term Memory Networks and attention mechanism (CNN-BiLSTM-attention) model was used to classify regular, mild, and severe patients. The models were trained using focal Loss instead of traditional cross-entropy Loss, and cross-validation was performed. Our method achieved a classification accuracy of 92.5 % for the two-class classification among 40 subjects and 72.2 % for the three-class classification among 36 subjects. Furthermore, we observed that the proposed model outperformed traditional EEG decoding models. This paper presents an effective computer-aided diagnosis method that eliminates the need for manual extraction of EEG features and holds potential for future auxiliary diagnosis of spinal cord-type cervical spondylosis.

Neurophysiological biomarkers of motor improvement from Constraint-Induced Movement Therapy and Robot-Assisted Therapy in participants with stroke

Background

The mechanism of stroke recovery is related to the reorganization of cerebral activity that can be enhanced by rehabilitation therapy. Two well established treatments are Robot-Assisted Therapy (RT) and Constraint-Induced Movement Therapy (CIMT), however, it is unknown whether there is a difference in the neuroplastic changes induced by these therapies, and if the modifications are related to motor improvement. Therefore, this study aims to identify neurophysiological biomarkers related to motor improvement of participants with chronic stroke that received RT or CIMT, and to test whether there is a difference in neuronal changes induced by these two therapies.

Methods

This study included participants with chronic stroke that took part in a pilot experiment to compare CIMT vs. RT. Neurophysiological evaluations were performed with electroencephalography (EEG) and transcranial magnetic stimulation (TMS), pre and post rehabilitation therapy. Motor function was measured by the Wolf Motor Function Test (WMFT) and Fugl-Meyer Assessment Upper Limb (FMA-UL).

Results

Twenty-seven participants with chronic stroke completed the present study [mean age of 58.8 years (SD ± 13.6), mean time since stroke of 18.2 months (SD ± 9.6)]. We found that changes in motor threshold (MT) and motor evoked potential (MEP) in the lesioned hemisphere have a positive and negative correlation with WMFT improvement, respectively. The absolute change in alpha peak in the unlesioned hemisphere and the absolute change of the alpha ratio (unlesioned/lesioned hemisphere) is negatively correlated with WMFT improvement. The decrease of EEG power ratio (increase in the lesioned hemisphere and decrease in the unlesioned hemisphere) for high alpha bandwidths is correlated with better improvement in WMFT. The variable "type of treatment (RT or CIMT)" was not significant in the models.

Conclusion

Our results suggest that distinct treatments (RT and CIMT) have similar neuroplastic mechanisms of recovery. Moreover, motor improvements in participants with chronic stroke are related to decreases of cortical excitability in the lesioned hemisphere measured with TMS. Furthermore, the balance of both EEG power and EEG alpha peak frequency in the lesioned hemisphere is related to motor improvement.

Neuroplasticity changes in knee osteoarthritis (KOA) indexed by event-related desynchronization/synchronization during a motor inhibition task

PURPOSE

Event-related desynchronisation (ERD) and event-related synchronisation (ERS) reflect pain perception and integration of the nociceptive sensory inputs. This may contribute to the understanding of how neurophysiological markers of Knee Osteoarthritis (KOA) patients can differ from control individuals, which would improve aspects such as prediction and prognosis. We performed a cross-sectional analysis of our cohort study (DEFINE cohort), KOA arm, with 71 patients, compared with 65 control participants. The study aimed to examine possible differences between ERD and ERS in control participants compared to Knee Osteoarthritis (KOA) patients when adjusting for important covariates.

MATERIALS AND METHODS

We performed independent multivariate regression models considering as dependent variables the power value related to ERD and ERS for four different sensorimotor tasks (Motor Execution, Motor Imagery, Active Observation and Passive Observation) and four sensorimotor oscillations (Alpha, Beta, Low Beta, and High Beta), each model, controlled by age and sex.

RESULTS

We demonstrate that the differences between KOA and healthy subjects are frequency specific, as most differences are in the beta bandwidth range. Also, we observed that subjects in the KOA group had significantly higher ERD and ERS. This may be correlated to the amount of lack of brain organisation and a subsequent attempt at compensation induced by KOA.

CONCLUSIONS

Our findings strengthen the notion that subjects with KOA have a higher degree of brain plasticity changes that are also likely correlated to the degree of compensation and behavioural dysfunction.

EEG theta and beta bands as brain oscillations for different knee osteoarthritis phenotypes according to disease severity

This study aims to investigate the multivariate relationship between different sociodemographic, clinical, and neurophysiological variables with resting-state, high-definition, EEG spectral power in subjects with chronic knee osteoarthritis (OA) pain. This was a cross-sectional study. Sociodemographic and clinical data were collected from 66 knee OA subjects. To identify associated factors, we performed independent univariate and multivariate regression models by frequency bands (delta, theta, alpha, beta, low-beta, and high-beta) and by pre-defined regions (frontal, central, and parietal). From adjusted multivariate models, we found that: (1) increased frontocentral high-beta power and reduced central theta activity are positively correlated with pain intensity (β = 0.012, 95% CI 0.004-0.020; and β = - 0.008; 95% CI 0.014 to - 0.003; respectively); (2) delta and alpha oscillations have a direct relationship with higher cortical inhibition; (3) diffuse increased power at low frequencies (delta and theta) are associated with poor cognition, aging, and depressive symptoms; and (4) higher alpha and beta power over sensorimotor areas seem to be a maladaptive compensatory mechanism to poor motor function and severe joint degeneration. Subjects with higher pain intensity and higher OA severity (likely subjects with maladaptive compensatory mechanisms to severe OA) have higher frontocentral beta power and lower theta activity. On the other hand, subjects with less OA severity and less pain have higher theta oscillations power. These associations showed the potential role of brain oscillations as a marker of pain intensity and clinical phenotypes in chronic knee OA patients. Besides, they suggest a potential compensatory mechanism of these two brain oscillators according to OA severity.

Insights and Future Directions on the Combined Effects of Mind-Body Therapies with Transcranial Direct Current Stimulation: An Evidence-based Review

Mind-body therapies (MBTs) use mental abilities to modify electrical neural activity across brain networks. Transcranial direct current stimulation (tDCS) is a non-invasive brain stimulation technique that modulates neuronal membrane potentials to enhance neuroplasticity. A combination of these treatment strategies may generate synergistic or additive effects, and thus has been more commonly tested in clinical trials, fostering a novel yet promising field of research. We conducted a literature search in four different databases including only randomized clinical trials (RCTs) that tested the combination of MBTs with tDCS. Ten studies (n=461) were included. Combined protocols included meditation/mindfulness (8/10), biofeedback (1/10), and hypnosis (1/10). The RCTs were heterogeneous with regards to population, design, and types of outcomes. Based on the findings of this search, we provide here a content description, methodological and practical insights, and future directions for the field. We hope this review will provide future authors with information to facilitate the development of trials with improved protocols.

Transcranial direct current stimulation combined with robotic training in incomplete spinal cord injury: a randomized, sham-controlled clinical trial

STUDY DESIGN

A randomized, sham-controlled clinical trial.

OBJECTIVE

To test the effects of tDCS, combined with robotic training, on gait disability in SCI. Our hypothesis was that participants who received active tDCS would experience greater walking gains, as indexed by the WISCI-II, than those who received sham tDCS.

SETTING

University of São Paulo, Brazil.

METHODS

This randomized, double-blind study comprised 43 participants with incomplete SCI who underwent 30 sessions of active (n = 21) or sham (n = 22) tDCS (20 min, 2 mA) before every Lokomat session of 30 min (3 times a week over 12 weeks or 5 times a week over 6 weeks). The main outcome was the improvement in WISCI-II. Participants were assessed at baseline, after 15 and 30 sessions of Lokomat, and after three months of treatment.

RESULTS

There was a significant difference in the percentage of participants that improved in WISCI-II at the 30-session, compared with baseline: 33.3% in the sham group and 70.0% in the active group (p = 0.046; OR: 3.7; 95% CI: 1.0-13.5). At the follow-up, the improvement compared with baseline in the sham group was 35.0% vs. 68.4% for the active group (p = 0.046; OR: 3.7; 95% CI: 1.0-13.5). There was no significant difference at the 15-session.

CONCLUSION

Thirty sessions of active tDCS is associated with a significant improvement in walking, compared to sham. Moreover, 15 sessions had no significant effect. The improvement in WISCI-II can be related to different aspects of motor learning, including motor recovery and compensation.

Deficit of Inhibition as a Marker of Neuroplasticity (DEFINE Study) in Rehabilitation: A Longitudinal Cohort Study Protocol

Background: Brain plasticity is an intrinsic property of the nervous system, which is modified during its lifetime. This is one mechanism of recuperation after injuries with an important role in rehabilitation. Evidence suggests that injuries in the nervous system disturb the stability between inhibition and excitability essential for the recuperation process of neuroplasticity. However, the mechanisms involved in this balance are not completely understood and, besides the advancement in the field, the knowledge has had a low impact on the rehabilitation practice. Therefore, the understanding of the relationship between biomarkers and functional disability may help to optimize and individualize treatments and build consistent studies in the future.

Methods: This cohort study, the deficit of inhibition as a marker of neuroplasticity study, will follow four groups (stroke, spinal cord injury, limb amputation, and osteoarthritis) to understand the neuroplasticity mechanisms involved in motor rehabilitation. We will recruit 500 subjects (including 100 age- and sex-matched controls). A battery of neurophysiological assessments, transcranial magnetic stimulation, electroencephalography, functional near-infrared spectroscopy, and magnetic resonance imaging, is going to be used to assess plasticity on the motor cortex before and after rehabilitation. One of the main hypotheses in this cohort is that the level of intracortical inhibition is related to functional deficits. We expect to develop a better understanding of the neuroplasticity mechanisms involved in the rehabilitation, and we expect to build neurophysiological “transdiagnostic” biomarkers, especially the markers of inhibition, which will have great relevance in the scientific and therapeutic improvement in rehabilitation. The relationship between neurophysiological and clinical outcomes will be analyzed using linear and logistic regression models.

Discussion: By evaluating the reliability of electroencephalography, functional near-infrared spectroscopy, transcranial magnetic stimulation, and magnetic resonance imaging measures as possible biomarkers for neurologic rehabilitation in different neurologic disorders, this study will aid in the understanding of brain plasticity mechanisms in rehabilitation, allowing more effective approaches and screening methods to take place.

Electroencephalography as a Biomarker for Functional Recovery in Spinal Cord Injury Patients

BACKGROUND

Functional changes after spinal cord injury (SCI) are related to changes in cortical plasticity. These changes can be measured with electroencephalography (EEG) and has potential to be used as a clinical biomarker.

METHOD

In this longitudinal study participants underwent a total of 30 sessions of robotic-assisted gait training (RAGT) over a course of 6 weeks. The duration of each session was 30 min. Resting state EEG was recorded before and after 30-session rehabilitation therapy. To measure gait, we used the Walking Index for Spinal Cord Injury Scale, 10-Meter- Walking Test, Timed-Up-and-Go, and 6-Min-Walking Test. Balance was measured using Berg Balance Scale.

RESULTS

Fifteen participants with incomplete SCI who had AIS C or D injuries based on American Spinal Cord Injury Association Impairment Scale classification were included in this study. Mean age was 35.7 years (range 17-51) and the mean time since injury was 17.08 (range 4-37) months. All participants showed clinical improvement with the rehabilitation program. EEG data revealed that high beta EEG activity in the central area had a negative correlation with gait (p = 0.049; β coefficient: -0.351; and adj-R 2: 0.23) and balance (p = 0.043; β coefficient: -0.158; and adj-R 2:0.24) measured at baseline, in a way that greater high beta EEG power was related to worse clinical function at baseline. Moreover, improvement in gait and balance had negative correlations with the change in alpha/theta ratio in the parietal area (Gait: p = 0.049; β coefficient: -0.351; adj-R 2: 0.23; Balance: p = 0.043; β coefficient: -0.158; and adj-R 2: 0.24).

CONCLUSION

In SCI, functional impairment and subsequent improvement following rehabilitation therapy with RAGT correlated with the change in cortical activity measured by EEG. Our results suggest that EEG alpha/theta ratio may be a potential surrogate marker of functional improvement during rehabilitation. Future studies are necessary to improve and validate these findings as a neurophysiological biomarker for SCI rehabilitation.

Toward improving functional recovery in spinal cord injury using robotics: a pilot study focusing on ankle rehabilitation

Background: Conventional physical therapy interventions are strongly recommended to improve ambulation potential and upright mobility in persons with incomplete spinal cord injury (iSCI). Ankle rehabilitation plays a significant role, as it aims to stem drop foot consequences.Research question: This pilot study aimed to assess the neurophysiological underpinnings of robot-aided ankle rehabilitation (using a platform robot) compared to conventional physiotherapy and its efficacy in improving gait performance and balance in persons with iSCI.Methods: Ten individuals with subacute/chronic iSCI (six males and four females, 39 ± 13 years, time since injury 8 ± 4 months, ASIA impairment scale grade C-D) were provided with one-month intensive training for robot-aided ankle rehabilitation (24 sessions, 1 h daily, six times a week). Clinical (10-Meter Walk Test (10MWT), 6-Minute Walk Test (6MWT), and Timed Up and Go test (TUG)), and electrophysiological aftereffects (surface-EMG from tibialis anterior and medial gastrocnemius muscles to estimate muscle activation patterns; and corticomuscular coherence-CMC-to assess functional synchronization between sensorimotor cortex and muscles, i.e. the functional integrity of corticospinal output) were assessed at baseline (PRE) and after the trial completion (POST). The experimental group (EG) data were compared with those coming from a retrospective control group (CG; n = 10) matched for clinical-demographic characteristics, who previously underwent conventional ankle rehabilitation.Results: the EG achieved a greater improvement in balance and gait as compared to the CG (TUG EG from 70 ± 18 to 45 ± 15 s, p = 0.002; CG from 68 ± 21 to 48 ± 18 s, p = 0.01; group-comparison p = 0.001; 10MWT EG from 0.43 ± 0.11 to 0.51 ± 0.09 m/s, p = 0.006; CG from 0.4 ± 0.13 to 0.45 ± 0.12, p = 0.01; group-comparison p = 0.006; 6 MWT EG from 231 ± 13 to 274 ± 15 m, p < 0.001; CG from 236 ± 13 to 262 ± 15 m, p = 0.003; group-comparison p = 0.01). Furthermore, the EG showed a retraining of muscle activation (an increase within proper movements, with a reduction of co-contractions) and CMC (beta frequency increase within proper movements, i.e. in a framework of preserved motor coordination). The improvements in CMC, gait, balance, and muscle activation were not correlated with each other.Conclusions: Robot-aided ankle rehabilitation improved gait performance by selectively ameliorating CMC, muscle activation patterns, and, lastly, gait balance and speed. Despite CMC, gait, balance, and muscle activation were not correlated, this pilot study suggests that robot-aided ankle rehabilitation may favor a better communication between above-SCI and below-SCI structures. This communication improvement may depend on a more synchronized corticospinal output (as per CMC increase) and a better responsiveness of below-SCI motorneurons to corticospinal output (as per specific and ankle movement focused muscle activation increases at the surface EMG), thus favoring greater recruitment of spinal motor units and, ultimately, improving muscle activation pattern and strength.Significance: Adopting robot-aided ankle rehabilitation protocols for persons with iSCI in the subacute/chronic phase may allow achieving a clinically significant improvement in gait performance.