IEEE ... International Conference on Rehabilitation Robotics : [proceedings]最新文献

Power wheelchair driving entails safety risks, including tips and collisions. Thus, driving assessments are important to safeguard both drivers and their surroundings. However, current driving assessments are deterministic, whereas driving itself is stochastic due to individual and environmental uncertainty. Moreover, disabilities magnify this uncertainty. In this study, a robotic wheelchair was used for a novel function: stochastic assessment of PWC driving. The Robotic Individualized Driving Evaluation (RIDE) is a stochastic assessment that contrasts with deterministic assessments by accounting for individual differences via assessment profiles. The robotic wheelchair acquired the information needed for the stochastic model, and a probabilistic risk score was formulated. The purpose of this preliminary study was to test the effect of assessment profiles within and between driving tasks. Within tasks, there were significant differences in RIDE risk between the profiles. Between tasks, there were significant differences in the new stochastic RIDE metrics but not in the conventional deterministic metrics. Results demonstrated potential for this novel use of robotic wheelchairs to support personalized assessment and training in power mobility rehabilitation.

The human-robot attachments (HRA) connecting humans to exoskeleton robots should facilitate user independence, efficient donning/doffing, maintain user-robotkinematic alignment using interfaces that are sufficiently rigid, comfortable to wear, and (with the aid of the exoskeleton) accommodate dexterous human motion. This report presents a proof-of-concept comparison of an innovative self-aligning, sizeadjustable HRA system designed for the upper arm. A novel discretely contracting orthosis adjusted by a dial ratcheting cable mechanism was fabricated from flexible and semi-rigid plastics to improve support and rigidity. Experiments comparing this new design to an existing rigid HRA design for the BLUE SABINO are conducted to characterize suitability in terms of ability to maintain kinematic alignment and to distribute interaction forces evenly on the user's arm. Piezoelectric force-sensitive-resistor sensors embedded in an elastic sleeve are used to measure human/orthosis interface forces at 8 Locations distributed on the circumference of the arm interface. Kinematic alignment was assessed for $5^{text {th }}$ percentile female to $95^{text {th }}$-percentile male sizes using 3D-printed mock arm cross sections. Laser alignment experiments found that the proposed device reduced arm center to orthosis center misalignment magnitude by up to 11.5 mm for $5^{text {th }}$ percentile male arm circumference vs. the existing design. Force measurements are inconclusive but indicate the potential to alter force distribution while being adequately adjustable and usable with repeatable settings.

Upper limb amputation leads to severe restrictions in daily activities and psychosocial difficulties, which can dramatically reduce the quality of life of affected people. Despite the significant scientific and engineering effort in building advanced robotic prostheses, their abandonment rates suggest a discrepancy between user needs and device performance. To address this gap, we present an initial step in identifying, harmonizing, and prioritizing the needs of people with limb loss. Using the PRISMA methodology, we analysed 73 papers on the topic. Using a bottom-up approach, we clustered the needs identified in different surveys into categories and macro-categories based on a taxonomy derived from the terms identified in the literature. We then prioritized the needs after a normalization of the major surveys. We believe that this work will provide both a high-level and low-level understanding of the needs of people with limb loss, thereby helping to guide the design of more user-friendly prostheses. In the future, combining these results with the definition of technical specifications will enable the identification of needs that can be satisfied through personalized design, particularly considering recent advances in manufacturing processes.

Rehabilitation robots and assistive devices that detect the motor intent of their users can provide more intuitive and effective control. Pupil dilation occurs when people perform motor activities, but its utility for detecting motor intent has not been explored previously. In this work, a human participant research study is conducted to determine if pupillometric data can be used to differentiate between a person's intent to pick up or observe an object. Thirty participants were recruited to perform 120 trials of picking up and observing objects while their pupil dilation was recorded by an eye tracking headset. Features were extracted from the time series data and used to train a neural network classifier. The classifier was tested using leave-one-out cross-validation. The classifier achieved an average accuracy of 59.4% and F1 score of 0.578 across the thirty test datasets. The performance varied significantly depending on the participant used for testing, suggesting that the pupillometric approach to intent detection may be better suited to some participants than others. Future work should determine whether intent detection can be improved with more advanced machine learning methods, such as convolutional neural networks (CNN), and whether intent detection can be performed in real time.

The somatosensory function is crucial for motor control, but is often deemed secondary in the design of assistive technologies for individuals with neuromotor impairments. This study investigates osseoperception-auditory and vibrotactile sensations evoked through bone stimulation-at the wrist pisiform bone (PB) and metacarpal head (MCH) of the index finger in 12 participants. Vibratory stimuli (100-6000 Hz) were applied during four psychophysical experiments: sensation discrimination, perception thresholds, sensory mapping, and loudness evaluation. Tactile sensations occurred below 940 Hz at the PB and 1000 Hz at the MCH, while auditory sensations predominated above these thresholds. Stimulation at 400 Hz expanded tactile zones, with PB sensations extending to the forearm and MCH sensations to carpal and interphalangeal joints. MCH stimulation had lower perception thresholds (0.14 N), while PB stimulation exhibited frequency-dependent loudness variations. These findings suggest the MCH and wrist as promising sites to elicit osseoperception, paving the way for the development of sensory feedback strategies to complement the effect of assistive technologies, especially those interfaced through osseointegration.

This study investigates feature analysis and feature fusion from different sensor modalities for the task of identifying movement errors in physiotherapeutic exercises, using squats as a case study. Incorrectly performed exercises can lead to injuries, underscoring the need for accurate monitoring tools. In an experiment, ten participants performed squats in three variations: correct execution, forward lean, and lateral tilt. To identify movement patterns, we evaluated muscle activation through electromyography (EMG), kinematic data through Motion Capture (MoCap) and joint angles from video footage through MediaPipe Pose. Distinct movement patterns were identified for the erroneous variations: forward lean altered hip and knee angles, while lateral tilt caused asymmetries in posture. In the EMG signal, deviations in the activity of distinct muscles correlated clearly with specific erroneous movements. Activation in the Gluteus Maximus was higher for the forward lean, while activity in the Quadriceps was lower. For the lateral tilt, a clear difference between left and right muscle activation was visible. Signal processing techniques extracted key features, such as muscle activation peaks and joint angle deviations, that we used to discern between the different squat types with a decision tree model. MoCap-based features offered the highest precision when used on their own, but fusing different sensor modalities achieved the best results. Although the video-based classifications were less accurate, its cost-effectiveness and ease-of-use suggest potential for home rehabilitation. Future research should enhance marker-less technologies and enable real-time feedback for broader applications.

The goal of this study was to determine the feasibility of applying targeted force perturbations paired with spatiotemporal transcutaneous spinal cord stimulation during walking for improving balance in individuals post-stroke. Five individuals with chronic stroke ($>6$ months) were recruited and completed 6 sessions (3 times/week) of balance training paired with spatiotemporal spinal cord stimulation. Balance was assessed pre, post 2 weeks of training, and follow-up, i.e., 2 weeks after the end of training. Results indicated that individuals post-stroke could tolerate the force perturbations applied to the pelvis paired with spatiotemporal spinal cord stimulation during walking. Further, individual post-stroke showed improvements in balance, assessed using the Dynamic Gait Index and Berg Balance Scale scores, after targeted force perturbation balance training paired with spatiotemporal spinal cord stimulation, although this was not significant due to the small sample size. Results from this study suggest that it was feasible to improve balance in individuals poststroke using targeted pelvis force perturbations paired with spatiotemporal spinal cord stimulation during walking.

Individuals with grasping impairments face significant challenges in performing daily tasks due to reduced hand function. Existing supernumerary robotic fingers (SRFs) often rely on electronic components that limit their usability through added weight and power constraints. This work introduces the first body-powered wrist-driven SRF (bpSRF), a design that addresses these limitations by removing all electronic components. The proposed bpSRF weighs 52 g and is mostly made of 3D printed parts. Using body-powered actuation principles, the extra finger is driven by wrist movements, offering a practical, affordable, and lightweight solution for hand augmentation. The bpSRF has two degrees of freedom, enabling users to perform enhanced grasping techniques with minimal learning curve. Experimental validation with five healthy participants showed high success rates and rapid learning of novel grasping patterns for tasks involving large or multiple objects that typically require two hands. The design offers a workspace volume approximately three times larger than a human thumb, potentially expanding users' manipulation capabilities. This research contributes to a new paradigm in assistive technology, presenting a lightweight, cost-effective, and open source SRF that can enhance grasping abilities for individuals with motor impairments while also offering augmentation possibilities for healthy users.

Partial body weight support during gait training and other tasks is a common practice. Approaches using robotic devices allow for a wide range of functions but at great cost and complexity. Based on constant force springs, we have created a low-cost body weight support system from off-the-shelf parts, with no tools required for assembly or maintenance, and it is easily integrated into any overhead support system. The design is presented as open source for future improvements. We evaluated the hysteresis using two different types of constant force springs, comprising clinically relevant levels of weight support, 61.2 N and 131 N. Force constancy during walking was examined. Hysteresis had noticeable effects on weight support, illustrating a potential challenge for future designs with body weight support through constant force springs. By providing the constant weight support, the most common type of body weight support, this device represents an accessible alternative to robotic dynamic body weight support tools.

Current research on physical human-robot interactions (pHRI) in wearable assistive robots, such as lower-limb exoskeletons, primarily focuses on improving net force estimates at each interface to improve robot controller performance. Consequently, estimating force distribution along physical interfaces of wearable robots, crucial for user safety and comfort, has been largely overlooked. We propose a novel computational model that uses interface geometry and strapping tension as inputs, and predicts the static pressure field generated during the user donning process by treating the supporting surface as an elastic foundation. Accuracy of the proposed computational method was validated by comparing the estimated static pressure field of a commercially available interface to experimental data. While measured pressure magnitudes were significantly lower than model prediction, likely due to a combination of assumptions and limitations associated with model design, similar loading patterns were observed. Identifying regions of high pressure from simulation and similar patterns allow for reliable scaling to reduce inaccuracies, and may be used to inform design. Further refinements of the proposed model will provide a valuable tool for developing more comfortable and safer interfaces for wearable robots.