Beyond aphasia: Altered EEG connectivity in Broca's patients during working memory task

Integrating EEG and Machine Learning to Analyze Brain Changes during the Rehabilitation of Broca's Aphasia

The fusion of electroencephalography (EEG) with machine learning is transforming rehabilitation. Our study introduces a neural network model proficient in distinguishing pre- and post-rehabilitation states in patients with Broca's aphasia, based on brain connectivity metrics derived from EEG recordings during verbal and spatial working memory tasks. The Granger causality (GC), phase-locking value (PLV), weighted phase-lag index (wPLI), mutual information (MI), and complex Pearson correlation coefficient (CPCC) across the delta, theta, and low- and high-gamma bands were used (excluding GC, which spanned the entire frequency spectrum). Across eight participants, employing leave-one-out validation for each, we evaluated the intersubject prediction accuracy across all connectivity methods and frequency bands. GC, MI theta, and PLV low-gamma emerged as the top performers, achieving 89.4%, 85.8%, and 82.7% accuracy in classifying verbal working memory task data. Intriguingly, measures designed to eliminate volume conduction exhibited the poorest performance in predicting rehabilitation-induced brain changes. This observation, coupled with variations in model performance across frequency bands, implies that different connectivity measures capture distinct brain processes involved in rehabilitation. The results of this paper contribute to current knowledge by presenting a clear strategy of utilizing limited data to achieve valid and meaningful results of machine learning on post-stroke rehabilitation EEG data, and they show that the differences in classification accuracy likely reflect distinct brain processes underlying rehabilitation after stroke.

Brain Analysis with a Complex Network Approach in Stroke Patients Based on Electroencephalography: A Systematic Review and Meta-Analysis

BACKGROUND AND PURPOSE

Brain function can be networked, and these networks typically present drastic changes after having suffered a stroke. The objective of this systematic review was to compare EEG-related outcomes in adults with stroke and healthy individuals with a complex network approach.

METHODS

The literature search was performed in the electronic databases PubMed, Cochrane and ScienceDirect from their inception until October 2021.

RESULTS

Ten studies were selected, nine of which were cohort studies. Five of them were of good quality, whereas four were of fair quality. Six studies showed a low risk of bias, whereas the other three studies presented a moderate risk of bias. In the network analysis, different parameters such as the path length, cluster coefficient, small-world index, cohesion and functional connection were used. The effect size was small and not significant in favor of the group of healthy subjects (Hedges'g = 0.189 [-0.714, 1.093], Z = 0.582, p = 0.592).

CONCLUSIONS

The systematic review found that there are structural differences between the brain network of post-stroke patients and healthy individuals as well as similarities. However, there was no specific distribution network to allows us to differentiate them and, therefore, more specialized and integrated studies are needed.

Functional connectivity drives stroke recovery: shifting the paradigm from correlation to causation

Stroke is a leading cause of disability, with deficits encompassing multiple functional domains. The heterogeneity underlying stroke poses significant challenges in the prediction of post-stroke recovery, prompting the development of neuroimaging-based biomarkers. Structural neuroimaging measurements, particularly those reflecting corticospinal tract injury, are well-documented in the literature as potential biomarker candidates of post-stroke motor recovery. Consistent with the view of stroke as a 'circuitopathy', functional neuroimaging measures probing functional connectivity may also prove informative in post-stroke recovery. An important step in the development of biomarkers based on functional neural network connectivity is the establishment of causality between connectivity and post-stroke recovery. Current evidence predominantly involves statistical correlations between connectivity measures and post-stroke behavioral status, either cross-sectionally or serially over time. However, the advancement of functional connectivity application in stroke depends on devising experiments that infer causality. In 1965, Sir Austin Bradford Hill introduced nine viewpoints to consider when determining the causality of an association: [1] Strength, [2] Consistency [3] Specificity, [4] Temporality, [5] Biological gradient, [6] Plausibility, [7] Coherence, [8] Experiment, and [9] Analogy. Collectively referred to as the Bradford Hill Criteria, these points have been widely adopted in epidemiology. In this review, we assert the value of implementing Bradford Hill's framework to stroke rehabilitation and neuroimaging. We focus on the role of neural network connectivity measurements acquired from task-oriented and resting-state functional magnetic resonance imaging, electroencephalography, magnetoencephalography, and functional near-infrared spectroscopy in describing and predicting post-stroke behavioral status and recovery. We also identify research opportunities within each Bradford Hill tenet to shift the experimental paradigm from correlation to causation.

Simultaneously stimulating both brain hemispheres by rTMS in patients with unilateral brain lesions decreases interhemispheric asymmetry

BACKGROUND

Interhemispheric asymmetry caused by brain lesions is an adverse factor in the recovery of patients with neurological deficits. Repetitive transcranial magnetic stimulation (rTMS) has been shown to modulate cortical oscillation and proposed as an approach to rebalance the symmetry, which has not been documented well.

OBJECTIVE

In this study, we investigated the influence of repetitive transcranial magnetic stimulation (rTMS) on EEG power in patients with unilateral brain lesions by simultaneously stimulating both brain hemispheres and to elucidate asymmetrical changes in rTMS-induced neurophysiological activity.

METHODS

Fourteen patients with unilateral brain lesions were treated with one active and one sham session of 10 Hz rTMS over the vertex (Cz position). Resting-state EEGs were recorded before and immediately after rTMS. The brain symmetry index (BSI), calculated from a fast Fourier transform, was employed to quantify the power asymmetry in both hemispheres and paired channels over the entire range and five frequency bands (delta, theta, alpha, beta and gamma bands).

RESULTS

Comparison between active and sham sessions demonstrated rTMS-induced EEG after-effects. rTMS in the active session significantly reduced the BSI in patients with unilateral brain lesions over the entire frequency range (t = 2.767, P = 0.016). Among the five frequency bands, rTMS only induced a noticeable decrease in the BSI in the delta band (t = 2.254, P = 0.042). Furthermore, analysis of different brain regions showed that significant changes in the BSI of the alpha band were only demonstrated in the posterior parietal lobe. In addition, EEG topographic mapping showed a decreased power of delta oscillations in the ipsilesional hemisphere, whereas distinct cortical oscillations were observed in the alpha band around the parietal-occipital lobe in the contralesional hemisphere.

CONCLUSIONS

When both brain hemispheres were simultaneously activated, rTMS decreased interhemispheric asymmetry primarily via reducing the delta band in the lesioned hemisphere.

Spectral Resting-State EEG (rsEEG) in Chronic Aphasia Is Reliable, Sensitive, and Correlates With Functional Behavior

We investigated spectral resting-state EEG in persons with chronic stroke-induced aphasia to determine its reliability, sensitivity, and relationship to functional behaviors. Resting-state EEG has not yet been characterized in this population and was selected given the demonstrated potential of resting-state investigations using other neuroimaging techniques to guide clinical decision-making. Controls and persons with chronic stroke-induced aphasia completed two EEG recording sessions, separated by approximately 1 month, as well as behavioral assessments of language, sensorimotor, and cognitive domains. Power in the classic frequency bands (delta, theta, alpha, and beta) was examined via spectral analysis of resting-state EEG data. Results suggest that power in the theta, alpha, and beta bands is reliable for use as a repeated measure. Significantly greater theta and lower beta power was observed in persons with aphasia (PWAs) than controls. Finally, in PWAs theta power negatively correlated with performance on a discourse informativeness measure, while alpha and beta power positively correlated with performance on the same measure. This indicates that spectral rsEEG slowing observed in PWAs in the chronic stage is pathological and suggests a possible avenue for directly altering brain activation to improve behavioral function. Taken together, these results suggest that spectral resting-state EEG holds promise for sensitive measurement of functioning and change in persons with chronic aphasia. Future studies investigating the utility of these measures as biomarkers of frank or latent aphasic deficits and treatment response in chronic stroke-induced aphasia are warranted.

Post-Stroke Working Memory Dysfunction: A Meta-Analysis and Systematic Review

This review investigates the severity and nature of post-stroke working memory deficits with reference to the multi-component model of working memory. We conducted a systematic search in PubMed up to March 2019 with search terms for stroke and memory. Studies on adult stroke patients, that included a control group, and assessed working memory function, were selected. Effect sizes (Hedges' g) were extracted from 50 studies (in total 3,084 stroke patients) based on the sample size, mean and standard deviation of patients and controls. Performance of stroke patients was compared to healthy controls on low-load (i.e. capacity) and high-load (executively demanding) working memory tasks, grouped by modality (verbal, non-verbal). A separate analysis compared patients in the sub-acute and the chronic stage. Longitudinal studies and effects of lesion location were systematically reviewed. Stroke patients demonstrated significant deficits in working memory with a moderate effect size for both low-load (Hedges' g = -.58 [-.82 to -.43]) and high-load (Hedges' g = -.59 [-.73 to -.45]) tasks. The effect sizes were comparable for verbal and non-verbal material. Systematically reviewing the literature showed that working memory deficits remain prominent in the chronic stage of stroke. Lesions in a widespread fronto-parietal network are associated with working memory deficits. Stroke patients show decrements of moderate magnitude in all subsystems of working memory. This review clearly demonstrates the global nature of the impairment in working memory post-stroke.

Associations of Alpha and Beta Interhemispheric EEG Coherences with Indices of Attentional Control and Academic Performance

Introduction. Heretofore, research on optimizing academic performance has suffered from an inability to translate what is known about an individual's learning behaviors to how effectively they are able to use the critical nodes and hubs in their cerebral cortex for learning. A previous study from our laboratory suggests that lower theta-beta ratios (TBRs) measured by EEG may be associated with higher academic performance in a medical school curriculum.

METHODS

In this study, we tested the hypothesis that TBR and academic performance may be correlated with EEG coherence, a measure of brain connectivity. We analyzed the interhemispheric coherences of the subjects involved in our prior study. TBR and coherence measurements were made at 19 scalp electrode recording sites and 171 electrode combinations with eyes open and closed (EO, EC). Control data were acquired during a session of acclimation to the research protocol 3 d before an initial examination in anatomy-physiology (control exam) and were repeated five weeks later, 3 d before a second exam covering different anatomy-physiology topics (comparison exam).

RESULTS

Between the control and comparison exams, beta coherences increased significantly at the frontal pole, frontal, parietal, midtemporal, posterior temporal, and occipital recording sites under the EO condition and at the inferior frontal, central, midtemporal, and posterior temporal sites under the EC condition. Alpha coherences increased significantly at the same sites and under the same EO/EC conditions as found for the beta coherences. The beta coherences were negatively correlated with the TBR and were positively correlated with the comparison exam score at the midfrontal electrode site (F3-F4) but only under the EO condition. Beta and alpha coherences at the midfrontal, inferior frontal midtemporal, posterior temporal, and occipital sites were also negatively correlated with the average TBR under the EO condition.

CONCLUSIONS

Lower TBR, an indicator of attentional control, was associated with higher alpha and beta interhemispheric coherences measured with eyes open at sites overlying the frontal, temporal, and occipital cortices. Changes in EEG coherences and TBRs might be useful as neurophysiological measures of neuroplasticity and the efficacy of strategies for preventing academic underachievement and treatments for improving academic performance.