Ongoing research on wearable rehabilitation robots explores challenges related to actuation and control in human-centric environments. To achieve precise force control and ensure smooth operation of rehabilitation devices, new actuation mechanisms and advanced control methods are continually being analyzed. The task becomes more challenging when the focus is on the human ankle joint. Conventional actuation methods are being replaced by hybrid actuation methods like series elastic actuators to take advantage of both passive and active elements. Selecting parameters for passive elements and choosing control methods remain challenging due to their behavior and limited operational range. This work has been done to investigate the effect of the passive element spring in series elastic actuators and improve their applicability in the rehabilitation domain. Response of the series elastic actuator for torque tracking when subjected to various standard input signals is thoroughly analysed employing conventional Proportional Integral Derivative (PID) control and Robust Integral of Sign of Error (RISE) based advanced control. The emphasis is on understanding how these controls perform when applied to active orthosis for rehabilitation purposes using RMSE and mean error metrics. The PID control stabilizes much more quickly within a time frame of similar to 0.02 seconds, whereas RISE control for the same input stabilises, taking similar to 0.48 seconds to start tracking with minimal error. RISE excels for ankle reference inputs, with extremely low RMSE and Mean Error, especially at higher stiffness values. Results obtained from the analysis will aid orthotic designers in designing robotic ankle foot orthosis and implementing control methodology for ankle rehabilitation robots.

The rapid advancement of neurofeedback interfaces assimilating electroencephalograms (EEG) with Virtual and Augmented Reality (VR/AR) technologies offers innovative rehabilitation opportunities. This review systematically examines electroencephalogram-based immersive computing approaches for upper extremity rehabilitation in neuromotor dysfunctions, aiming to restore motor function. Following PRISMA guidelines, we screened 3,490 articles from databases including CINAHL, Cochrane, Elsevier, IEEE Xplore, PubMed, Science Direct, Springer Link, Wiley Online Library, and Google Scholar (January 2011-May 2023), selecting 38 high-quality studies. Four central paradigms emerged: motor imagery, VR/AR-based activities, electroencephalogram acquisition, and immersive rehabilitation characteristics. The integration of VR/AR with electroencephalogram neurofeedback enhances nervous system recovery, surpassing conventional techniques by improving feedback and brain function restoration. This review highlights the efficacy of immersive computing and electroencephalogram-based rehabilitation, demonstrating their potential to advance rehabilitation methodologies, overcome conventional limitations, and improve clinical outcomes for neuromotor dysfunctions.

Neurodegenerative disorders particularly Freezing-of-Gait (FoG) in Parkinson’s disease (PD) accounts for movement inabilities. Disease assessment is based on subjective criterion leading to unreliability and bias. Therefore, an accurate detection system can aid in strengthening clinical decision-making. This research work aims to develop a machine-learning based feature fusion methodology for FoG detection. Two feature extraction schemes (statistical and spectral) have been adopted to derive signal features from three sensor locations (shank, trunk, and upper leg). In the initial phase, publicly-available DAPHNET dataset of 10PD patients has been used for developing our proposed methodology. Sensor optimization has been accomplished by performing detailed set of experiments, thus contributing to use of single shank acceleration sensor. Top-ranked statistical and spectral features selected by Mutual Information technique have been fused together to form hybrid feature-set of 19 features. Proposed Random Forest model is able to achieve an accuracy, precision, recall, and F1-score of 93 ± 1.4

This work reports the extraction and evaluation of clinically relevant spatiotemporal and statistical gait parameters from developed wireless foot sensor module as recommended by the Biomathics and Canadian Gait Consortium Initiative. Further, normalization of extracted spatiotemporal gait parameters reduces inter-subject physiological variations. To validate their performance towards gait analysis, a machine learning framework is implemented for personnel identification. The study results suggest a promising potential for utilizing the extracted feature-set for the automatic multiclass gait disorders classification. Developed module is a low cost, easy-to-use device, and has potential application for setups with limited access to state of art gait analysis laboratory.

Congenital hemiparesis resulting from cerebral palsy adversely affects the psychomotor functions of children. Virtual reality (VR)–based exergames represent a promising and advanced approach to psychomotor rehabilitation, as they deliver intensive, repetitive, and task-specific training in a motivating and engaging format. The present study aimed to evaluate the efficacy of VR-based exergames in improving motor and perceptual functions by analyzing performance scores and task completion times, thereby assessing the rehabilitation progress in children with hemiparesis. The exergames were individually designed based on standardized assessments, specifically the Bruininks-Oseretsky Test of Motor Proficiency (BOT-2) and the Developmental Test of Visual Perception (DTVP-3). The exergames incorporated biosensing technologies to enhance eye-hand coordination and visual perception, including hand motion sensors, inertial measurement units (IMUs), and touch-based interfaces. Eight children diagnosed with congenital hemiparesis participated in a three-month VR-based intervention targeting upper extremity rehabilitation. Statistical analysis of BOT-2 outcomes revealed a significant improvement in motor function, both in comparison with conventional therapy (F = 4.75, p = 0.01) and across varying levels of exergame difficulty (F = 10.62, p = 0.0006). The combined outcomes from BOT-2 and DTVP-3 further confirmed that the intervention effectively enhanced fine and gross motor skills, visual perception, and eye-hand coordination. These findings support the potential of biosensor-assimilated VR exergames as an effective rehabilitation tool for improving upper limb functionality in children with congenital hemiparesis due to cerebral palsy.

Wearable sensors allow mobility assessment required for better locomotion, neurological and musculoskeletal disorders, current limitations include unknown reliability and accuracy in real-life settings. This work determined the concurrent validity and repeatability of the proposed foot-worn gait evaluation system using objective gait features and recurrence quantification analysis from 52 participants. Its agreement with the commercially available OpenGo® system in the unrestricted outdoor environment is determined. Reported measures showed no significant differences (p > 0.05) between systems. Test-retest reliability showed that the mean of the second-third trial (T2-T3) is the most significant. Thus, an affordable system provides accurate measurement of gait ensuring its suitability even in small clinical-settings.

Decoding motor activity accurately and reliably from electroencephalography (EEG) signals is essential for several portable brain-computer interface (BCI) applications ranging from neural prosthetics to the control of industrial and mobile robots. Spiking neural networks (SNNs) is an emerging brain-inspired architecture that is well-suited for decoding EEG signals due to their built-in ability to integrate information at multiple timescales, leading to energy-efficient solutions for portable BCI. In practice, however, current SNN solutions suffer from i) an inefficient spike encoding of the EEG signals; ii) non-specialized network architectures that cannot capture EEG priors of spatiotemporal dependencies; and iii) the limited generalizability of the local learning rules commonly used to train the networks. These untapped challenges result in a performance gap between the current SNN approaches and the state-of-the-art deep neural network (DNN) methods. Moreover, the black-box nature of most current SNN solutions masks their correspondence with the underlying neurophysiology, further hindering their reliability for real-world applications. Here, we propose an SNN architecture with an input encoding and network design that exploits the priors of spatial and temporal dependencies in the EEG signal. To extract spatiotemporal features, the network comprised of spatial convolutional, temporal convolutional, and recurrent layers. The network weights and the neuron membrane parameters were trained jointly using gradient descent and our method was validated in classifying movement on two datasets: i) an in-house dataset comprising of complex components of movement, namely reaction time and directions, and ii) the publicly available eegmmidb dataset for motor imagery and movement. We deployed our SNN on Intel's Loihi neuromorphic processor, and show that our method consumed 95\% less energy per inference than the state-of-the-art DNN methods on NVIDIA Jeston TX2, while achieving similar levels of classification performance. Finally, we interpreted the SNN using a network perturbation study to identify the spectral bands and brain activity that correlated with the SNN outputs. The results were in agreement with the current neurophysiological knowledge implicating the activation patterns in the low-frequency oscillations over the motor cortex for hand movement and imagery tasks. Overall, our approach demonstrates the effectiveness of SNNs in accurately and reliably decoding EEG while availing the computational advantages offered by neuromorphic computing, and paves the way for employing neuromorphic methods in portable BCI systems.

Rate of Recovery during the rehabilitation procedure is not effectively evaluated due to existing limitations of measurement tools and large human variability. The conventional way to treat lower-limb injuries is by conducting a physician-guided rehabilitation process that could take several weeks or months. But the process of estimating the rate of recovery is entirely based on the expertise of a therapist. Moreover, it remains unclear to a large extent about the factors have contributed to the success of treatment, the most effective rehabilitation duration, and recovery progression rate. To address these limitations, a random subspace ensemble k-nearest neighbor (RSEnkNN) model is proposed to predict the rehabilitation duration of an individual after initial baseline assessment of gait trajectories. The model is trained using an open-source dataset comprising 211 healthy and 2084 patients with functional gait disorders. The proposed model is able to predict rehabilitation duration with an accuracy of 88-89 % in case of ankle, calcaneus, and hip injury while the success rate of 82 % is achieved for knee disorders. Features computed from ground reaction force, measures of postural steadiness, and bilateral symmetry turn out to be effective in predicting the early, moderate, or late recovery of an individual. The significantly contributing features for each anomaly class are reported using recursive feature elimination (RFE) technique to support clinicians. Thereafter, recovery rate is determined using a novel metric Rate of Gait Recovery (RoGR) index that could be effective for treatment planning and re-scheduling of therapy sessions (if required). This work determined that the most effective rehabilitation period is 2-6 weeks post-surgery for restoration to normal walk. Thus, a prediction tool for strengthening clinical decision-making is proposed to support early rehabilitation.

Osteoarthritis is a widespread health issue that causes pain, limits physical function, lowers quality of life, and makes individuals disabled. To improve impairments, virtual reality-based rehabilitation treatments are becoming more acceptable because of effectiveness in recent decades. This pre-post study examines rehabilitative training program that trains different motor areas. To check feasibility and benefits of a developed system for postoperative rehabilitation in individuals with osteoarthritis-undergone primary Total knee Arthroplasty to improve their performance. Eleven-individuals suffering from various motor impairments aged 50-85 years with post-operative duration of >= 4weeks. The experimental group followed, virtual reality-based rehabilitation training, whereas control group received standard therapies for 4 weeks, 5 sessions/week of 40-45 min. To evaluate baseline and post-assessment with Visual Analogue scale, Womac Osteoarthritis Index, Activities-specific Balance Confidence Scale, Knee Range of motion, Lower limb muscle strength. Results of system were evaluated using system usability scale for performance and user-friendliness. Statistically significant differences were found between experimental and control group: pain, range of motion, muscle strength, and balance confidence (p < 0.05). No significant improvement in lower limb muscle strength was recorded. Mean SUS score was 72 % (SD = 08.55). The findings suggest that virtual reality-based rehabilitation treatment is an effective therapy, resulting in motivation and functional recovery with lower extremity in these individuals.

Cognitive enhancement is becoming increasingly crucial for adolescents with neurodevelopmental impairments. This study highlights the evaluation and treatment of intellectual disabilities in adolescents who have difficulty with daily living activities despite having conceptual-based skills. To assess cognitive skills in these adolescents, the study utilized Virtual Reality (VR) approaches, integrating various modalities in the field of rehabilitation. The research aimed to develop virtual reality therapeutic modules that target two critical aspects of cognitive skills, using behavioral assessment scales tailored for Indian adolescents with intellectual disabilities, such as the Vocational Assessment and Programming System for Persons with Mental Retardation and the Functional Assessment Checklist for Programming. The results of this investigation showed the feasibility of implementing the virtual reality approach with intellectually disabled adolescents, with evaluations based on a single trial and the average time taken to complete the modules. The system was user-friendly, adaptable, and did not require any external hardware. A significant advantage of the developed therapeutic modules is their availability in multiple languages, tailored to the needs of intellectually disabled adolescents. Over time, these virtual reality therapeutic modules can provide a comprehensive rehabilitation program for these adolescents.

Decades of research determined electromyography (EMG) and electroencephalography (EEG), individually, as paramount controls aimed at rehabilitation. However, correlating commands from the central nervous system to lower-limb movements are highly complex raising challenges for anthropomorphic lower-limb prosthesis control. This work establishes a hierarchical relationship between 12-channel EEG and 4-channel surface EMG using a hybrid model (LRG-2L-LSTM) for lower-limb ankle movement recognition by estimating muscular activity from cortical brain signals. The proposed model achieved R-2 = 0.742 & PLUSMN; 0.03, RMSE = 0.067 & PLUSMN; 0.002 for EMG estimation and averaged recognition accuracy of 84.86 & PLUSMN; 0.27% for ankle movements using estimated EMG, thereby, establishing lower-limb prosthesis and exoskeleton control for amputees with little to no muscular strength.

Rehabilitation using external cues has shown positive impact on Freezing-of-Gait (FoG) severity in Parkinson's disease (PD) patients. Research studies on FoG detection and prediction are widely accessible, however little is known about transitions from "FoG" to "normal-gait" phases and vice-versa. Especially, for the patient to recover completely from the freeze event as well as to avoid undesirable distractions due to long-term cue persistence, post-FoG periods must be optimally chosen for managing automatic cue deactivation. Therefore, our study attempts to device a machine-learning based approach for labelling of accelerometer data with state-transitions from "normal-gait" to "FoG" (i.e., FoG-prediction) as well as from "FoG" to "normal-gait" (i.e., unfreeze-prediction). Our prediction approach distinguishes itself by innovating in sensor utilization, feature-extraction, and latency reduction. We introduce a novel four-class labeling scheme based on state transitions, diverging from fixed-time periods utilized in prior studies. Model training involves regression and classification with a streamlined set of 23 hybrid features, minimizing latency with reduced window segments and operating on 4-second and 3-second pre- and post-FoG time-frames, respectively, exhibiting minimal 0.5-second latency. Our approach simply required one shank sensor providing minimum-instrumentation for the user's comfort. Four-class (No-FoG, pre-FoG, FoG, and post-FoG) data have been fed to train random-forest classifier. Our FoG prediction model achieved exceptional metrics, particularly for the "post-FoG" class, with 99.0 % accuracy, 97.4 % precision, 99.1 % sensitivity, and 98.1 % f1-score, along with 99.0 % cohen's-kappa, 98.3 % jaccard-index, and 99.0 % Matthews-correlation-coefficient. Therefore, an intelligent-strategy for timely activation and deactivation of cues is the major contribution of our work.

Purpose The aim of this study was to determine the ideal forearm position that allows maximum upper-limb function. In certain congenital/acquired upper-limb disorders, the management boils down to sacrificing rotatory movements of the forearm. The position of fusion that facilitates maximum upper-limb function is a topic of debate and is decided upon by personal preferences and assumptions. Although the literature has many level five evidence reports, there is a lack of well-designed research to answer the same question and we intended to study it both in dominant and non-dominant limbs. Methods 15 healthy adolescent volunteers were fitted with a custom adjustable brace that simulated forearm arthrodesis in five rotatory positions. They were asked to carry out a series of activities as per Sollerman’s hand function test, and each activity was scored using the standardized scoring system. The test was carried out with the brace fitted first in the dominant side, followed by the non-dominant side, and finally in both the upper limbs together. Results We found that the mid-prone position allowed for the best function overall in both dominant and non-dominant upper limbs, and if both upper limbs required simultaneous fusion, our results suggest that fixing the dominant side in mid-prone and non-dominant side in 45° supination would be ideal. Conclusions For unilateral forearm arthrodesis, the ideal position of fusion is the same irrespective of the dominance of the limb, whereas, for bilateral arthrodesis, limb dominance is to be taken into consideration. Level of Evidence Level III quasi-experimental study.

Estimation and control of zero-moment point (ZMP) is a widely used concept for planning the locomotion of bipedal robots and is commonly measured using integrated joint angle encoders and foot force sensors. Contemporary methods for ZMP measurement involve built-in contact sensors such as joint encoders or instrumented foot force sensors. This paper presents a novel approach for computing ZMP for a humanoid robot using inertial sensor-based wireless foot sensor modules (WFSMs). The developed WFSMs, strapped at different limb segments of a bipedal robot, measure lower limb joint angles in real time. The joint angle trajectories, further transformed into Cartesian position coordinates, are used for estimating the ZMP positions of humanoid robots using the planar biped model. The whole framework is presented through experimental studies for different real-life walking scenarios. Since the modules work based on the limb motion and inclination, any ground unevenness would be automatically reflected in the module output. Hence, this measurement process can be a convenient method for applications requiring humanoid control on uneven surfaces/outdoor terrains. To compare the performance of the proposed model, ZMP is simultaneously measured from inbuilt foot force sensors and joint encoders of the robot. Statistical tests exhibit a high linear correlation between the proposed method with integrated encoders and foot force sensors (Pearson’s coefficient, [Formula: see text]). Results indicate that ZMP estimated by WFSM is a viable method to monitor the dynamic gait balance of a humanoid robot and has potential application in outdoor and uneven terrains.

This work presents a novel method for the estimation of stride length (SL) from inertial sensor-based wireless foot sensor module (WFSM). SL variability is a widely used and clinically relevant spatial gait parameter. This work proposes and validates a novel method for computing SL by measuring the foot inclination angle and single-axis acceleration measured from WFSM. The foot inclination angle is used for detecting gait events as well as for gravity compensation to estimate the real acceleration of the foot. This approach thus requires minimum input information, i.e. foot angle trajectory and acceleration in only sagittal plane. The results are validated on a cohort of healthy subjects against a standard gait analysis platform. A set of trial data on 10 healthy volunteers were also analysed for outdoor walking to evaluate the performance against previously reported works. The proposed approach has the potential for applications involving real-time gait measurements.

As spasticity is a major cause of illness in patients that suffer from spinal cord injury (SCI), there are several limitations in the present robotic-based therapy for the treatment of acute SCI patients. A large literature is present for the use of full body exoskeletons, but due to their limitations such as uneasiness due to repetitive motion, does not meet the anthropometric range. As manual therapy is tiresome, a need for ankle-foot therapy exoskeleton was felt and Ankle-Foot Orthosis was developed. This paper presents an ankle-foot therapy exoskeleton device for spasticity treatment of acute SCI patients to reduce the manual efforts of physiotherapists. The proposed device has one Degree of Freedom (DOF) in the sagittal plane in ankle joint. This paper analyses the control design of the prototype and clinical trial results on SCI patients, which show that the proposed prototype is viable in improving the range of ankle motion alleviating the spasticity with p = 0.000365.

Healthy aging is crucial for socioeconomic independence and quality living. Nevertheless, age-related gait variations result in mobility issues influencing human walk. Hence, studying lower-limb functionality with aging aids in sustaining good mobility. This work analyzed collected gait and ankle joint data of 80 healthy participants (21–60 years) using a developed G-Eva (system for gait evaluation) system during dynamic walking. Thereafter, data are preprocessed followed by feature extraction and selection using statistical, correlation, and relative importance analysis. With aging, gait variability is observed in walking speed, thus, a weighted majority-voting ensemble model (MVens) is proposed for age and gender prediction of middle-aged individuals to overcome imbalance influence. MVens model is able to achieve an accuracy of 90% for predicting age-related gait deterioration and 96.25% for gender estimation. It has also been demonstrated that among middle-aged adults, spatiotemporal parameters show significant changes followed by symmetry and variability analysis with aging as opposed to older adults. However, other significantly contributing features include differences in spatiotemporal and gait variability parameters followed by symmetry for gender prediction. Thus, the developed system with the proposed MVens model can be used for early diagnosis of mobility and joint-health disorders for initiation of timely intervention and rehabilitation.

The effective decoding of movement from non-invasive electroencephalography (EEG) is essential for informing several therapeutic interventions, from neurorehabilitation robots to neural prosthetics. Deep neural networks are most suitable for decoding real-time data but their use in EEG is hindered by the gross classes of motor tasks in the currently available datasets, which are solvable even with network architectures that do not require specialized design considerations. Moreover, the weak association with the underlying neurophysiology limits the generalizability of modern networks for EEG inference. Here, we present a neurophysiologically interpretable 3-dimensional convolutional neural network (3D-CNN) that captured the spatiotemporal dependencies in brain areas that get co-activated during movement. The 3D-CNN received topography-preserving EEG inputs, and predicted complex components of hand movements performed on a plane using a back-drivable rehabilitation robot, namely (a) the reaction time (RT) for responding to stimulus (slow or fast), (b) the mode of movement (active or passive, depending on whether there was an assistive force provided by the apparatus), and (c) the orthogonal directions of the movement (left, right, up, or down). We validated the 3D-CNN on a new dataset that we acquired from an in-house motor experiment, where it achieved average leave-one-subject-out test accuracies of 79.81%, 81.23%, and 82.00% for RT, active vs. passive, and direction classifications, respectively. Our proposed method outperformed the modern 2D-CNN architecture by a range of 1.1% to 6.74% depending on the classification task. Further, we identified the EEG sensors and time segments crucial to the classification decisions of the network, which aligned well with the current neurophysiological knowledge on brain activity in motor planning and execution tasks. Our results demonstrate the importance of biological relevance in networks for an accurate decoding of EEG, suggesting that the real-time classification of other complex brain activities may now be within our reach.

Parkinson's disease (PD) severity assessment in clinical settings largely depends on expertise level of clinicians which have inherent limitations and non-uniformity. Instrumented gait analysis plays a significant role in disease diagnosis and management. However, these are agonized from data dispersion due to demography, anthropometry, and self-selected walking speed of an individual. This research work aims to develop computationally efficient hybrid strategy using normalized features for PD severity evaluation. The relevance of each considered gait feature in demonstrating the outcomes is explained through feature importance and partial dependence plot (PDP) to build substantial insight for clinical needs. Gait, a biomarker, is used to access human mobility by utilizing vertical ground reaction force (VGRF) data of 72 healthy and 93 Parkinson's individuals. A multi-variate normalization approach identifies gait differences between PD and non-PD. The proposed hybrid model used is able to detect PD with accuracy of 99.39% and 99.9%, and its severity assessment based on MDS-UPDRS-III shows coefficient of determination (R2) as 97% and 98.7% using leave-one-out cross-validation (CV) and tenfold CV respectively. The significant features suitable for clinical implications are reported. Moreover, normalized gait parameters supplement capability to compare individuals with diverse physical properties, resulting in assistive system for evaluation of PD severity.