Journal of NeuroEngineering and Rehabilitation最新文献

Background: Impaired arm position sense is a common somatosensory impairment after stroke, which significantly impacts the performance of functional activities using the upper limb. However, few clinical interventions target loss of position sense after stroke. Our aim was to use interlimb force-coupling to augment position sense of the stroke-affected arm during a bilateral reaching task and investigate the impact of training with this feedback manipulation on measures of arm position matching ability and both bilateral and unilateral motor control.

Methods: Twenty-four participants with a history of stroke were randomized (N = 12/group) to perform mirrored bimanual aiming movements with either interlimb force-coupling (Augmented PF) or uncoupled symmetrical reaches with only visual feedback about movement position. Participants completed 11 sessions (295 bimanual reaches/session) using a Kinarm End-Point robot. Performance on measures of arm position sense (Arm Position Matching, APM), motor impairment (Fugl-Meyer Upper Limb, FM), motor function (Wolf Motor Function Test, WMFT), unilateral reach accuracy and speed (Visually Guided Reaching, VGR), and bilateral reach symmetry were collected before and after training to characterize changes in upper limb somatosensory and motor control performance.

Results: APM Task Scores improved for both groups. This improvement was specifically observed through reduced APM variability, but not accuracy. FM scores also improved for both groups. The group that did not practice with force-coupling between limbs improved on measures of bilateral movement symmetry on a mirrored reaching task and had faster VGR movement times in post-test.

Conclusion: Symmetrical reach training with or without augmented PF led to reduced motor impairment and benefited upper limb position matching ability by reducing APM variability. Augmenting position sense during reaching did not provide additional benefits for position matching accuracy. Advantages for unilateral movement speed and bilateral reach symmetry measures in the group that practiced without interlimb coupling may reflect specificity of practice effects due to similarity between test and training conditions for this group.

Background: Brain/neural hand exoskeletons (B/NHEs) can restore motor function after severe stroke, enabling bimanual tasks critical for various activities of daily living. Yet, reliable clinical tests for assessing bimanual function compatible with B/NHEs are lacking. Here, we introduce the Berlin Bimanual Test for Stroke (BeBiT-S), a 10-task assessment focused on everyday bimanual activities, and evaluate its psychometric properties as well as compatibility with assistive technologies such as B/NHEs.

Methods: BeBiT-S tasks were selected based on their relevance to daily activities, representation of various grasp types, and compatibility with current (neuro-)prosthetic devices. A scoring system was developed to assess key aspects of bimanual function-including reaching, grasping, stabilizing, manipulating, and lifting-based on video recordings of task performance. The BeBiT-S was administered without support of assistive technology (unassisted condition) to 24 stroke survivors (mean age = 56.5 years; 9 female) with upper-limb hemiparesis. We evaluated interrater reliability through the intraclass correlation coefficient (ICC) and construct validity through correlations with the Chedoke Arm and Hand Activity Inventory (CAHAI), and Stroke Impact Scale (SIS). A subgroup of 15 stroke survivors (mean age 50.3 years, 5 female) completed a second session supported by a B/NHE (B/NHE-assisted condition) to assess the BeBiT-S' sensitivity to change related to B/NHE-application.

Results: The BeBiT-S demonstrated high interrater reliability in both the unassisted (ICC = 0.985, P < .001) and B/NHE-assisted (ICC = 0.862, P < .001) conditions. Unassisted BeBiT-S scores correlated with the CAHAI-8 (r(22) = 0.95, P < .001) and the SIS subscales "strength" (r(20) = 0.53, P = .012) and "hand function" (r(20) = 0.50, P = .018), indicating construct validity. BeBiT-S scores improved significantly with B/NHE assistance (Mdn = 60, P < .05), compared to when no assistance was provided (Mdn = 38, P < .05), demonstrating the test's sensitivity to change following the application of a B/NHE.

Conclusions: The findings support that the BeBiT-S is a reliable and valid tool for evaluating bimanual task performance in stroke survivors and is compatible with the use of assistive technology such as B/NHEs. Trial registration NCT04440709, submitted June 18th, 2020.

Background: EMG-based hand-gesture recognition can enable home-based post-stroke rehabilitation, yet one-size-fits-all feature sets overlook differences across recovery stage METHODS: Thirteen post-stroke participants performed seven gestures while EMG was recorded from six forearm sensors. From 38 time- and frequency-domain features, we derived stage-specific subsets for Low (Brunnstrom 1-2, minimal movement), Medium (3-4, partial movement), and High (5-6, near-normal movement) using a wrapper approach Sequential Forward Selection (SFS). For reference, we included a filter comparison using minimum Redundancy-Maximum Relevance (mRMR). To provide fair baselines, we reproduced two literature feature sets within an identical Light Gradient Boosting Machine (LightGBM) pipeline: (i) a healthy-cohort feature set and (ii) a patient-cohort feature set that was not stage-stratified and did not focus on feature selection (we adopted the features as reported). Multiple classifiers-Linear Discriminant Analysis, Support Vector Machines, Random Forest, LightGBM, Logistic Regression, and K-Nearest Neighbors-were evaluated via group-wise cross-validation. Within-stage variability was quantified using pairwise Jaccard overlap of selected features.

Results: Stage-tailored subsets achieved compact yet accurate models: High = 81.5% (14 features, LightGBM), Medium = 80.2% (9 features, LightGBM), Low = 65.0% (7 features, Random Forest). SFS exceeded the mRMR filter comparison and outperformed both literature baselines under the same LightGBM pipeline (paired tests across CV folds, [Formula: see text]). Relative to the healthy-cohort baseline, gains were +6.5% (High), +6.2% (Medium), and +12.0% (Low); relative to the non-stage-stratified patient baseline, gains were +9.5%, +10.2%, and +21.0%, respectively. Time-domain metrics-particularly Difference Absolute Standard Deviation Value and Sample Entropy were most frequently selected. Jaccard analyses indicated within-stage heterogeneity alongside convergence on a small set of core discriminative features.

Conclusions: Brunnstrom stage-specific feature engineering substantially improves EMG gesture-classification accuracy over both healthy-derived and non-stage-stratified patient baselines while reducing computational load. These findings support adaptive, stage-aware designs for wearable rehabilitation systems and motivate larger Low-stage cohorts and models robust to sparse or low-SNR signals.

Background: Gait instability is a common and disabling symptom of Parkinson's disease (PD), contributing to frequent falls and reduced quality of life. While clinical balance tests and spatiotemporal gait measures can predict fall risk, they do not fully explain the underlying control mechanisms. In healthy individuals, foot placement is actively adjusted based on an estimate of the Center of Mass (CoM) state to maintain gait stability, known as foot placement control. This estimation relies on the integration of multisensory information, which has been shown to be impaired in PD, potentially disrupting the control of gait stability through foot placement. This study aimed to investigate whether foot placement coordination during overground walking is impaired in people with PD.

Methods: Fifty people with PD and 51 healthy older adults walked overground for 10 min at self-selected walking speed. Foot placement errors were quantified as the deviation between the actual foot placement and the predicted placement derived from the CoM kinematic state during the preceding swing phase.

Results: Foot placement errors were significantly higher in people with PD than in healthy older adults in both mediolateral (p < 0.05) and anteroposterior directions (p < 0.0001), at both mid-swing and terminal swing. Relative explained variance in mediolateral direction was significantly higher in people with PD compared to healthy older adults (p < 0.005).

Conclusion: We provide first evidence of impaired coordination between the CoM and foot placement in PD. Future work should investigate a causal relationship between impaired foot placement control, sensorimotor integration and gait instability.

Background: Precise prognostication of functional outcomes in individuals with prolonged disorders of consciousness (PDOC) is crucial for clinical decision-making. We systematically reviewed and meta-analyzed prognostic models for functional outcomes, as well as promising prognostic factors in PDOC.

Methods: We conducted a TRIPOD-SRMA-compliant systematic review, searching four databases through July 15, 2024. Risk of bias (ROB) was assessed by Prediction Model Risk of Bias Assessment Tool. Pooled sensitivity, specificity and the area under the summary receiver operating characteristic curve (SROC-AUC) were calculated based on outcome classifications (consciousness recovery, consciousness improvement, and favorable prognosis) and modality types (neurophysiological and demographic & clinical). Candidate predictors were ranked across studies, and promising factors were identified from models with both sensitivity and specificity exceeding 80%.

Results: 38 studies comprising 54 prognostic models were included in the systematic review, and 20 models from 19 studies were included in the meta-analysis. None of the included studies were assessed as having a low ROB. The overall pooled sensitivity was 82% (95% CI: 75-87%) and specificity was 85% (95% CI: 80-89%). The pooled SROC-AUC was 0.875 (95% CI: 0.842-0.911). Neurophysiological models demonstrated higher sensitivity. Both sensitivity and specificity exceeded 80% in predicting consciousness improvement but not in consciousness recovery or favorable prognosis. Male sex, minimally conscious state, and traumatic etiology indicated better outcomes, whereas no specific neurophysiological factor could be confirmed due to heterogeneity across studies. Most studies reported limited model performance metrics, especially clinical utility.

Conclusions: Although neurophysiological factors improve prognostic sensitivity, models' clinical impact remains limited without external validation. Future work must enhance clinical translation through more rigorous model development and validation.