Search

Stroke recovery is multifaceted and complex. Machine learning approaches have potential to identify patterns of brain activity associated with clinical outcomes, providing new insights into recovery. We aim to use machine learning to characterise the contribution of and potential interaction between resting state functional connectivity networks in predicting touch discrimination outcomes in a well-phenotyped, but small, stroke cohort. We interrogated and compared a suite of automated machine learning approaches to identify patterns of brain activity associated with clinical outcomes. Using feature reduction, the identification of combined ‘golden features’, and five-fold cross-validation, two golden features patterns emerged. These golden features identified patterns of resting state connectivity involving interactive relationships: 1. The difference between right insula and right superior temporal lobe correlation and left cerebellum and vermis correlation; 2. The ratio between right inferior temporal lobe and left cerebellum correlation and left frontal inferior operculum and left supplementary motor area correlation. Our findings demonstrate evidence of the potential for automated machine learning to provide new insights into brain network patterns and their interactions associated with the prediction of quantitative touch discrimination outcomes, through the automated identification of robust associations and golden feature brain patterns, even in a small cohort of stroke survivors.

Background: Exposure to thousands of head and body impacts during a career in contact and collision sports may contribute to current or later life issues related to brain health. Wearable technology enables the measurement of impact exposure. The validation of impact detection is required for accurate exposure monitoring. In this study, we present a method of automatic identification (classification) of head and body impacts using an instrumented mouthguard, video-verified impacts, and machine-learning algorithms.Methods: Time series data were collected via the Nexus A9 mouthguard from 60 elite level men (mean age = 26.33; SD = 3.79) and four women (mean age = 25.50; SD = 5.91) from the Australian Rules Football players from eight clubs, participating in 119 games during the 2020 season. Ground truth data labeling on the captures used in this machine learning study was performed through the analysis of game footage by two expert video reviewers using SportCode and Catapult Vision. The visual labeling process occurred independently of the mouthguard time series data. True positive captures (captures where the reviewer directly observed contact between the mouthguard wearer and another player, the ball, or the ground) were defined as hits. Spectral and convolutional kernel based features were extracted from time series data. Performances of untuned classification algorithms from scikit-learn in addition to XGBoost were assessed to select the best performing baseline method for tuning.Results: Based on performance, XGBoost was selected as the classifier algorithm for tuning. A total of 13,712 video verified captures were collected and used to train and validate the classifier. True positive detection ranged from 94.67% in the Test set to 100% in the hold out set. True negatives ranged from 95.65 to 96.83% in the test and rest sets, respectively.Discussion and conclusion: This study suggests the potential for high performing impact classification models to be used for Australian Rules Football and highlights the importance of frequencies

Background and Purpose: Altered executive functions and resting-state functional connectivity (rsFC) are common following a minor stroke or transient ischemic attack (TIA). However, the long-term persistence of these abnormalities is not well-studied. We investigated whether there were cognitive and rsFC differences between (a) controls and minor cerebrovascular event (CVE) patients and (b) between CVE patients with and without an imaging confirmed infarct (i.e., minor stroke and TIA, respectively) at an average of 3.8 years following their event.Methods: Structural and resting-state imaging and cognitive assessments including the Montreal Cognitive Assessment, the Trail Making Task and the National Institute of Health (NIH) Cognition Toolbox were conducted on 42 patients (minor stroke = 17, TIA = 25) and 20 healthy controls (total N = 62).Results: Controls performed better than patients on two measures of executive functioning (both p < 0.046) and had reduced rsFC between the frontoparietal and default mode networks (FPN and DMN, respectively; p = 0.035). No cognitive differences were found between minor stroke and TIA patients, however, rsFC differences were found within the FPN and the DMN (both p < 0.013). Specifically, increased connectivity within the FPN was associated with faster performance in the minor stroke group but not the TIA group (p = 0.047).Conclusions: These findings suggest that transient or relatively minor cerebrovascular events are associated with persistent disruption of functional connectivity of neural networks and cognitive performance. These findings suggest a need for novel interventions beyond secondary prevention to reduce the risk of persistent cognitive deficits.

One in two survivors experience impairment in touch sensation after stroke. The nature of this impairment is likely associated with changes associated with the functional somatosensory network of the brain; however few studies have examined this. In particular, the impact of lesioned hemisphere has not been investigated. We examined resting state functional connectivity in 28 stroke survivors, 14 with left hemisphere and 14 with right hemisphere lesion, and 14 healthy controls. Contra-lesional hands showed significantly decreased touch discrimination. Whole brain functional connectivity (FC) data was extracted from four seed regions, i.e. primary (S1) and secondary (S2) somatosensory cortices in both hemispheres. Whole brain FC maps and Laterality Indices (LI) were calculated for subgroups. Inter-hemispheric FC was greater in healthy controls compared to the combined stroke cohort from the left S1 seed and bilateral S2 seeds. The left lesion subgroup showed decreased FC, relative to controls, from left ipsi-lesional S1 to contra-lesional S1 and to distributed temporal, occipital and parietal regions. In comparison, the right lesion group showed decreased connectivity from contra-lesional left S1 and bilateral S2 to ipsi-lesional parietal operculum (S2), and to occipital and temporal regions. The right lesion group also showed increased intra-hemispheric FC from ipsi-lesional right S1 to inferior parietal regions compared to controls. In comparison to the left lesion group, those with right lesion showed greater intra-hemispheric connectivity from left S1 to left parietal and occipital regions and from right S1 to right angular and parietal regions. Laterality Indices were significantly greater for stroke subgroups relative to matched controls for contra-lesional S1 (left lesion group) and contra-lesional S2 (both groups). We provide evidence of altered functional connectivity within the somatosensory network, across both hemispheres, and to other networks in stroke survivors with impaired touch sensation. Hemisphere of lesion was associated with different patterns of altered functional connectivity within the somatosensory network and with related function was associated with different patterns of altered functional connectivity within the somatosensory network and with related functional networks. Keywords: Brain imaging, Resting state fMRI, Somatosensory, Tactile, Stroke, Laterality, Functional connectivity

Accumulating evidence shows that brain functional deficits may be impacted by damage to remote brain regions. Recent advances in neuroimaging suggest that stroke impairment can be better predicted based on disruption to brain networks rather than from lesion locations or volumes only. Our aim was to explore the feasibility of predicting post-stroke somatosensory function from brain functional connectivity through the application of machine learning techniques. Somatosensory impairment was measured using the Tactile Discrimination Test. Functional connectivity was employed to model the global brain function. Behavioral measures and MRI were collected at the same timepoint. Two machine learning models (linear regression and support vector regression) were chosen to predict somatosensory impairment from disrupted networks. Along with two feature pools (i.e., low-order and high-order functional connectivity, or low-order functional connectivity only) engineered, four predictive models were built and evaluated in the present study. Forty-three chronic stroke survivors participated this study. Results showed that the regression model employing both low-order and high-order functional connectivity can predict outcomes based on correlation coefficient of r = 0.54 (p = 0.0002). A machine learning predictive approach, involving high- and low-order modelling, is feasible for the prediction of residual somatosensory function in stroke patients using functional brain networks.

People with depression have shown alterations in processing emotional information and working memory functionality. There is some evidence that emotional content may interact with working memory update processes, however neurological correlates are current unknown. In this preliminary study we utilized a novel version of the emotional variant of the n-back working memory task in fMRI. We examined BOLD response of 14 healthy controls and 13 depressed participants in response to happy, sad, and neutral displays of facial affect. No accuracy or reaction time differences were found between the two groups. The depressed group showed significantly decreased BOLD response to happy faces compared to the control group areas of the dorsal striatum and anterior cingulate. Significant, moderate, positive associations were found between right caudate activation with anxiety score and anterior cingulate activation with depression score in those with depression. Our novel task was able to elicit group level differences in emotional processing during working memory update. These results suggest those with depression fail to differentiate between positive emotional stimuli and stimuli with no emotional content.

Background: Brain regions involved in processing somatosensory information have been well documented through lesion, post-mortem, animal, and more recently, structural and functional neuroimaging studies. Functional neuroimaging studies characterize brain activation related to somatosensory processing; yet a meta-analysis synthesis of these findings is currently lacking and in-depth knowledge of the regions involved in somatosensory-related tasks may also be confounded by motor influences.Objectives: Our Activation Likelihood Estimate (ALE) meta-analysis sought to quantify brain regions that are involved in the tactile processing of the right (RH) and left hands (LH) separately, with the exclusion of motor related activity.Methods: The majority of studies (n = 41) measured activation associated with RH tactile stimulation. RH activation studies were grouped into those which conducted whole-brain analyses (n = 29) and those which examined specific regions of interest (ROI; n = 12). Few studies examined LH activation, though all were whole-brain studies (N = 7).Results: Meta-analysis of brain activation associated with RH tactile stimulation (whole-brain studies) revealed large clusters of activation in the left primary somatosensory cortex (S1) and bilaterally in the secondary somatosensory cortex (S2; including parietal operculum) and supramarginal gyrus (SMG), as well as the left anterior cingulate. Comparison between findings from RH whole-brain and ROI studies revealed activation as expected, but restricted primarily to S1 and S2 regions. Further, preliminary analyses of LH stimulation studies only, revealed two small clusters within the right S1 and S2 regions, likely limited due to the small number of studies. Contrast analyses revealed the one area of overlap for RH and LH, was right secondary somatosensory region.Conclusions: Findings from the whole-brain meta-analysis of right hand tactile stimulation emphasize the importance of taking into consideration bilateral activation, particularly in secondary somatosensory cortex. Further, the right parietal operculum/S2 region was commonly activated for right and left hand tactile stimulation, suggesting a lateralized pattern of somatosensory activation in right secondary somatosensory region. Implications for further research and for possible differences in right and left hemispheric stroke lesions are discussed.

Background. One in three survivors of stroke experience poststroke depression (PSD). PSD has been linked with poorer recovery of function and cognition, yet our understanding of potential mechanisms is currently limited. Alterations in resting-state functional MRI have been investigated to a limited extent. Fluctuations in low frequency signal are reported, but it is unknown if interactions are present between the level of depressive symptom score and intrinsic brain activity in varying brain regions. Objective. To investigate potential interaction effects between whole-brain resting-state activity and depressive symptoms in stroke survivors with low and high levels of depressive symptoms. Methods. A cross-sectional analysis of 63 stroke survivors who were assessed at 3 months poststroke for depression, using the Montgomery–Åsberg Depression Rating Scale (MÅDRS-SIGMA), and for brain activity using fMRI. A MÅDRS-SIGMA score of >8 was classified as high depressive symptoms. Fractional amplitude of frequency fluctuations (fALFF) data across three frequency bands (broadband, i.e., ~0.01–0.08; subbands, i.e., slow-5: ~0.01–0.027 Hz, slow-4: 0.027–0.07) was examined. Results. Of the 63 stroke survivors, 38 were classified as “low-depressive symptoms” and 25 as “high depressive symptoms.” Six had a past history of depression. We found interaction effects across frequency bands in several brain regions that differentiated the two groups. The broadband analysis revealed interaction effects in the left insula and the left superior temporal lobe. The subband analysis showed contrasting fALFF response between the two groups in the left thalamus, right caudate, and left cerebellum. Across the three frequency bands, we found contrasting fALFF response in areas within the fronto-limbic-thalamic network and cerebellum. Conclusions. We provide evidence that fALFF is sensitive to changes in poststroke depressive symptom severity and implicates frontostriatal and cerebellar regions, consistent with previous studies. The use of multiband analysis could be an effective method to examine neural correlates of depression after stroke. The START-PrePARE trial is registered with the Australian New Zealand Clinical Trial Registry, number ACTRN12610000987066.

Aim. Neural plastic changes are experience and learning dependent, yet exploiting this knowledge to enhance clinical outcomes after stroke is in its infancy. Our aim was to search the available evidence for the core concepts of neuroplasticity, stroke recovery, and learning; identify links between these concepts; and identify and review the themes that best characterise the intersection of these three concepts. Methods. We developed a novel approach to identify the common research topics among the three areas: neuroplasticity, stroke recovery, and learning. A concept map was created a priori, and separate searches were conducted for each concept. The methodology involved three main phases: data collection and filtering, development of a clinical vocabulary, and the development of an automatic clinical text processing engine to aid the process and identify the unique and common topics. The common themes from the intersection of the three concepts were identified. These were then reviewed, with particular reference to the top 30 articles identified as intersecting these concepts. Results. The search of the three concepts separately yielded 405,636 publications. Publications were filtered to include only human studies, generating 263,751 publications related to the concepts of neuroplasticity (n=6,498), stroke recovery (n=79,060), and learning (n=178,193). A cluster concept map (network graph) was generated from the results; indicating the concept nodes, strength of link between nodes, and the intersection between all three concepts. We identified 23 common themes (topics) and the top 30 articles that best represent the intersecting themes. A time-linked pattern emerged. Discussion and Conclusions. Our novel approach developed for this review allowed the identification of the common themes/topics that intersect the concepts of neuroplasticity, stroke recovery, and learning. These may be synthesised to advance a neuroscience-informed approach to stroke rehabilitation. We also identified gaps in available literature using this approach. These may help guide future targeted research.

Binaural beats are an auditory phenomenon that has been suggested to alter physiological and cognitive processes including vigilance and brainwave entrainment. Some personality traits measured by the NEO Five Factor Model have been found to alter entrainment using pulsing light stimuli, but as yet no studies have examined if this occurs using steady state presentation of binaural beats for a relatively short presentation of two minutes. This study aimed to examine if binaural beat stimulation altered vigilance or cortical frequencies and if personality traits were involved. Thirty-one participants were played binaural beat stimuli designed to elicit a response at either the Theta (7 Hz) or Beta (16 Hz) frequency bands while undertaking a zero-back vigilance task. EEG was recorded from a high-density electrode cap. No significant differences were found in vigilance or cortical frequency power during binaural beat stimulation compared to a white noise control period. Furthermore, no significant relationships were detected between the above and the Big Five personality traits. This suggests a short presentation of steady state binaural beats are not sufficient to alter vigilance or entrain cortical frequencies at the two bands examined and that certain personality traits were not more susceptible than others.