Biped Gait Stability Classification Based on the Predicted Step Viability.

In this paper, we address the challenge of ensuring stability in bipedal walking robots and exoskeletons. We explore the feasibility of real-time implementation for the Predicted Step Viability algorithm (PSV), a complex multi-step optimization criterion for planning future steps in bipedal gait. To overcome the high computational cost of the PSV algorithm, we performed an analysis using 11 classification algorithms and a stacking strategy to predict if a step will be stable or not. We generated three datasets of increasing complexity through PSV simulations to evaluate the classification performance. Among the classifiers, k Nearest Neighbors, Support Vector Machine with Radial Basis Function Kernel, Decision Tree, and Random Forest exhibited superior performance. Multi-Layer Perceptron also consistently performed well, while linear-based algorithms showed lower performance. Importantly, the use of stacking did not significantly improve performance. Our results suggest that the feature vector applied with this approach is applicable across various robotic models and datasets, provided that training data is balanced and sufficient points are used. Notably, by leveraging classifiers, we achieved rapid computation of results in less than 1 ms, with minimal computational cost.

Design of Low-Cost Modular Bio-Inspired Electric-Pneumatic Actuator (EPA)-Driven Legged Robots.

Exploring the fundamental mechanisms of locomotion extends beyond mere simulation and modeling. It necessitates the utilization of physical test benches to validate hypotheses regarding real-world applications of locomotion. This study introduces cost-effective modular robotic platforms designed specifically for investigating the intricacies of locomotion and control strategies. Expanding upon our prior research in electric-pneumatic actuation (EPA), we present the mechanical and electrical designs of the latest developments in the EPA robot series. These include EPA Jumper, a human-sized segmented monoped robot, and its extension EPA Walker, a human-sized bipedal robot. Both replicate the human weight and inertia distributions, featuring co-actuation through electrical motors and pneumatic artificial muscles. These low-cost modular platforms, with considerations for degrees of freedom and redundant actuation, (1) provide opportunities to study different locomotor subfunctions-stance, swing, and balance; (2) help investigate the role of actuation schemes in tasks such as hopping and walking; and (3) allow testing hypotheses regarding biological locomotors in real-world physical test benches.

Towards Visual-Tactile Integration of Shoulder and Hand Using Immersive Virtual Reality.

Shoulder-controlled hand neuroprostheses are wearable devices designed to assist hand function in people with cervical spinal cord injury (SCI). They use preserved shoulder movements to control artificial actuators. Due to the concurrent afferent (i.e., shoulder proprioception) and visual (i.e., hand response) feedback, these wearables may affect the user's body somatosensory representation. To investigate this effect, we propose an experimental paradigm that uses immersive virtual reality (VR) environment to emulate the use of a shoulder-controlled hand neuroprostheses and an adapted version of a visual-tactile integration task (i.e., Crossmodal Congruency Task) as an assessment tool. Data from seven non-disabled participants validates the experimental setup, with preliminary statistical analysis revealing no significant difference across the means of VR and visual-tactile integration tasks. The results serve as a proof-of-concept for the proposed paradigm, paving the way for further research with improvements in the experimental design and a larger sample size.

Analysis of Motor Control and Learning in Human-Robot Interaction during Game Guided Movements.

Robotic devices can be used in upper limb rehabilitation in order to help the total or partial functional recovery. Robots can perform repetitive activities for a long period of time, which may be beneficial for rehabilitation processes. In this context, this study uses a bi-manual robotic device to investigate motor learning and control for the upper limbs among different game guided tasks, and inspect the user's grip force exerted in response to perturbations. The robotic device resembles a bicycle handlebar, instrumented with load cells to measure torques and grip forces. It is equipped with a DC motor to apply external torques to the guiding system. A game was developed containing in-game and physical perturbations to the natural movement of the handlebar. Tests were carried out with 16 healthy subjects that were instructed to move the handlebar guiding a character displayed on the screen with the objective of collecting tokens to get the higher score in the game. During the trials, corresponding data from the game and the load cells were collected and used to infer the learning process, the mean error in the trajectory and the variations in the force applied to the handles of the handlebar. Analyses showed that there was learning in the first repetitions, and the learning was retained further. The higher values of the grip force occurred when there was a physical perturbation to the handlebar's natural movement. The larger errors in the trajectories occurred immediately after the perturbations ended. In conclusion, there was a performance improvement, probably related to learning. The increase of the mean error at the transitions of the perturbations indicates the need for adaptation to the new conditions of the task.

Asymmetric effects of different training-testing mismatch types on myoelectric regression via deep learning.

This paper investigates how predictions of a convolutional neural network (CNN) suited for myoelectric simultaneous and proportional control (SPC) are affected when training and testing conditions differ. We used a dataset composed of electromyogram (EMG) signals and joint angular accelerations measured from volunteers drawing a star. This task was repeated multiple times using different combinations of motion amplitude and frequency. CNNs were trained with data from a given combination and tested under different combinations. Predictions were compared between situations in which training and testing conditions matched versus when there was a training-testing mismatch. Changes in predictions were assessed through three metrics: normalized root mean squared error (NRMSE), correlation, and slope of the linear regression between targets and predictions. We found that predictive performance declined differently depending on whether the confounding factors (amplitude and frequency) increased or decreased between training and testing. Correlations dropped as the factors decreased, whereas slopes deteriorated when factors increased. NRMSEs worsened when factors increased or decreased, with more accentuated deterioration for increasing factors. We argue that worse correlations could be related to differences in EMG signal-to-ratio (SNR) between training and testing, which affected the noise robustness of the CNNs' learned internal features. Slope deterioration could be a result of the networks' inability to predict accelerations outside the range seen during training. These two mechanisms may also asymmetrically increase NRMSE. Finally, our findings open further possibilities to develop strategies to mitigate the negative impact of confounding factor variability on myoelectric SPC devices.

Effect of Photobiomodulation With Different Wavelengths on Radiodermatitis Treatment.

Approximately 80% of patients submitted to radiotherapy develop radiodermatitis. Photobiomodulation based on light-emitted diode (LED) is one of the therapeutic strategies for treating inflammation. This study aimed to investigate the effect of the photobiomodulation with two wavelengths, in an acute radiodermatitis animal model. Methods: Twenty rats were submitted to one radiotherapy session. After 15 days, the rats that developed radiodermatitis were divided into control groups, LED-630 nm, LED-850 nm, and LED-630 + 850 nm. The treatment regimen was one session lasting 10 minutes on alternate days for 21 days. We analyzed macroscopy aspects (RTOG scale), vascular density, dermal appendages, VEGF-a, TNF-alpha, MMP-9, and MMP-9 genic expression level. Results: All LED groups revealed a two-point reduction on the radiodermatitis severity grade compared with the baseline classification. Dermal appendage and vascular analysis showed a higher counting in all LED groups compared to control. This study showed dermal appendages twice in the 630/850 nm group compared with the control group. The 630/850 nm group showed six times more arterioles than the control group. Regarding genic expression, this study showed a 10-fold decrease between LED-630 nm versus LED-630 + 850 nm (P = 0.02) interleukin-10 expression and a 12-fold decrease between control versus LED-630 nm (P = 0.006) and LED-850 nm (P = 0.002) in TNF-alpha. Conclusion: LED (630 nm, 850 nm, and 630 nm + 850 nm) showed benefit in the treatment of radiodermatitis, and the association of the 630 nm + 850 nm and 630 nm parameters demonstrated the best macroscopic and microscopic results.

Aging effects of haptic input on postural control under a dual-task paradigm.

Postural control relies on three principal sensory systems: vision, vestibular and proprioceptive; that are affected by aging. When performing a cognitive task concomitantly with a motor task, those sensory impairments lead to even greater deleterious effects on balance. We aimed to study the effects of a sensory aid (a light touch) on a dual task paradigm and sought to understand the different responses on balance due to aging. Fifty healthy and highly physical active women were divided in two groups: young (N = 25, 24.2 ± 4.0 years) and older adults (N = 25, 67.3 ± 4.2 years). In a random and balanced order, all participants performed five tasks: Stroop test while seated (Seated); Stroop test while standing quiet (ST); Standing quiet (BL); Standing quiet with a haptic input (LT); and Stroop test with a haptic input while standing quiet (SL). In the Stroop test, older women committed more errors (50 vs 11 errors, p < 0.001) and had higher reaction time (1.001 ± 0.191 vs 0.699 ± 0.081 s, p < 0.001). The haptic input (LT) reduced all body sway parameters, in both groups, regardless the condition. This means that postural control under a dual task paradigm (ST) deleterious effect can be mitigated by a haptic input.

Is Balance Control Affected by Sleep Deprivation? A Systematic Review of the Impact of Sleep on the Control of Balance.

Background: Sleep is a complex physiological function that should be addressed from different perspectives and consider the circadian rhythm. Sleep deprivation, either acute or chronic, negatively affects several functions, including motor control. Balance control is essential in several daily life activities and balance problems are related to falls. Research Question: This review focuses on how sleep conditions impact balance control. Methods: Systematic literature review according to PRISMA guidelines. Results: The literature provided strong evidence that acute sleep deprivation impairs postural control. Chronic sleep deprivation as well as low sleep quality had similar effects, although there is a lower number of works addressing this issue. Furthermore, time awake worsens postural controls and it can be used to detect sleepiness and fatigue. The sleep deprivation showed a stronger negative effect on postural control when removing the visual information (eyes closed) than when reducing proprioceptive feedback (soft surface). There is scarce literature about the effects of chronotype, circadian patterns and chronic sleep deprivation, a frequent problem, on balance control; however they consistently indicate that there is an relationship between them. Most of the studies only consider one-night (acute) sleep deprivation without monitoring prior sleep conditions and the circadian rhythm phase of the participants. However, a few studies indicated that these factors must be considered. Significance: These results suggest that the sleep conditions of a subject should be considered for several days prior to balance control tests. Therefore, we propose a revision of current postural measurement protocols to include sleep assessment, such as sleep quality questionnaires or actimetry, and to consider the circadian rhythm of the participants to plan the hour of the tests.

Soft robotics and functional electrical stimulation advances for restoring hand function in people with SCI: a narrative review, clinical guidelines and future directions.

BACKGROUND: Recovery of hand function is crucial for the independence of people with spinal cord injury (SCI). Wearable devices based on soft robotics (SR) or functional electrical stimulation (FES) have been employed to assist the recovery of hand function both during activities of daily living (ADLs) and during therapy. However, the implementation of these wearable devices has not been compiled in a review focusing on the functional outcomes they can activate/elicit/stimulate/potentiate. This narrative review aims at providing a guide both for engineers to help in the development of new technologies and for clinicians to serve as clinical guidelines based on the available technology in order to assist and/or recover hand function in people with SCI. METHODS: A literature search was performed in Scopus, Pubmed and IEEE Xplore for articles involving SR devices or FES systems designed for hand therapy or assistance, published since 2010. Only studies that reported functional outcomes from individuals with SCI were selected. The final collections of both groups (SR and FES) were analysed based on the technical aspects and reported functional outcomes. RESULTS: A total of 37 out of 1101 articles were selected, 12 regarding SR and 25 involving FES devices. Most studies were limited to research prototypes, designed either for assistance or therapy. From an engineering perspective, technological improvements for home-based use such as portability, donning/doffing and the time spent with calibration were identified. From the clinician point of view, the most suitable technical features (e.g., user intent detection) and assessment tools should be determined according to the particular patient condition. A wide range of functional assessment tests were adopted, moreover, most studies used non-standardized tests. CONCLUSION: SR and FES wearable devices are promising technologies to support hand function recovery in subjects with SCI. Technical improvements in aspects such as the user intent detection, portability or calibration as well as consistent assessment of functional outcomes were the main identified limitations. These limitations seem to be be preventing the translation into clinical practice of these technological devices created in the laboratory.

Shoulder muscle activity and perceived comfort of industry workers using a commercial upper limb exoskeleton for simulated tasks.

We compared the effects of using a commercial exoskeleton on shoulder muscle activity, task completion time, perceived effort and comfort while performing four tasks in different shoulder positions. Fourteen automotive industry workers performed four simulated tasks with shoulder at A≈0°, B ≈ 45°, C ≈ 90° and D ≈ 115° flexion. The electromyographic activity of the Medial Deltoid (MD) and the Anterior Deltoid (AD) decreased when wearing the exoskeleton. The effect sizes (ES) were, for MD: ES = 0, ES = -0.2, ES = -0.6, ES = -0.3; and for AD: ES = 0.3, ES = -0.6, ES = -0.8, ES = -0.6; for tasks A, B, C and D, respectively. We also found increased Anterior Deltoid/Triceps Brachii co-contraction, a typical joint stabilization mechanism. Wearing the exoskeleton increased the completion time of task B and reduced the perceived effort of tasks A and C, improving overall comfort. These findings are useful to organize the logistics of the workstations that use upper limb exoskeletons to improve the effectiveness of this equipment.