A Pneumatic Artificial Muscle and Spring Combination System that Assists Ankle Rocker and Transforms Energy into Push-Off Support: A Feasibility Study in Heathy Participants.

Insufficient push-off is a common problem for stroke hemiplegia patients. Assistive systems using extension spring to store energy during stance phase of gait to provide push-off assistance have been developed. However, patients could also suffer poor ankle rocker function; that is, poor dorsiflexion movement in stance phase. In such case, the spring could reversely become a burden for ankle movement. In this research, we proposed a system that combines a pneumatic artificial muscle and a tension spring. The artificial muscle mimics a human's tibialis anterior muscle, while the spring mimicking the Achilles tendon. Upon foot flat event of gait, the artificial muscle contracted to assist ankle rocker function and stretched the spring to store energy simultaneously. After heel off, the artificial muscle extended and the spring was released to provide push-off assistance. A feasibility study in seven healthy participants was conducted to verify assistance effects on their ankle rocker function and push-off movements. The results show significant increase in ankle rocker angle, push-off angle, and push-off torque compared with those during normal walk when the participants were assisted by our system. Therefore, we believe that the proposed system has great potential to assist stroke survivors with problems of poor ankle rocker function and push-off movements.

Investigation of optimal gait speed for motor learning of walking using the vibro-tactile biofeedback system.

In stroke patients, sensory loss often reduces the sensation of ground contact, which impairs motor learning during rehabilitation. In our previous study, we proposed a vibro-tactile biofeedback system (which we called the perception-empathy biofeedback system) for gait rehabilitation. The results of our 9-week pilot clinical test suggested that patients who had reached the autonomous phase in gait learning had difficulty noticing the external vibratory feedback provided by the biofeedback system, leading to ineffective intervention. We considered the possibility that slower walking speed might return the patient to the association phase and allow patients to improve their gait according to the sensory feedback provided. Thus, in this research, a method based on reducing walking speed to guide patients' attention was derived. A pilot clinical trial shows that there is a statistically significant increase of ankle dorsiflexion in the initial contact phase and increase of ankle plantarflexion in the push-off phase after vibro-tactile biofeedback system intervention with speed reduction, compared to intervention without speed reduction. The results suggest that, by reducing their walking speed during intervention, patients return to the association phase and recognize external vibratory feedback, which may result in better intervention effects.Clinical Relevance-This study provides knowledge about the optimal walking speed when using vibro-tactile biofeedback for motor learning in stroke patients.

Effects of Assisted Dorsiflexion Timing on Voluntary Efforts and Compensatory Movements: A Feasibility Study in Healthy Participants.

In previous research, we found that modulating the assistance timing of dorsiflexion may affect a user's voluntary efforts. This could constitute a focus area based on assistive strategies that could be developed to foster patients' voluntary efforts. In this present study, we conducted an experiment to verify the effects of ankle dorsiflexion assistance under different timings using a high-dorsiflexion assistive system. Nine healthy and young participants wore a dorsiflexion-restrictive device that enabled them to use circumduction or steppage gaits. On the basis of the transition from the stance to the swing phase of the gait, the assistance timings of the high-dorsiflexion assistive system were set to have delays, which ranged from 0 to 300 ms. The index results from eight out of nine participants evaluated compensatory movements and revealed positive strong/moderate correlations with assistance delay times (r = 0.627-0.965, p <.001), whereas the other participants also performed compensatory movement when dorsiflexion assistance timing was late. Meanwhile, the results from tibialis anterior surface electromyography from six out of nine participants showed positive strong/moderate correlations with dorsiflexion assistance delay times (r = 0.598-0.922, p <.001), indicating that tuning the assistance timing did foster these participants' voluntary dorsiflexion movements. This result indicates that there should be a trade-off between ensuring voluntary dorsiflexion movements and preventing incorrect gait patterns at different assistance timings. The findings of this feasibility study indicate the potential of developing an adaptive control method to ensure voluntary efforts during robot-assisted gait rehabilitation based on assistance timing modification. A new assistance mechanism should also be required to stimulate and motivate a patient's voluntary efforts and should reinforce the effects of active gait rehabilitation.

A Haptic-Based Perception-Empathy Biofeedback System with Vibration Transition: Verifying the Attention Amount.

In this paper, a perception-empathy biofeedback (PEBF) system is proposed that supplements the foot pressure status of a paralyzed foot with a wearable vibrotactile biofeedback (BF) vest to the back. Improvements in the ankle dorsiflexion and push-off movement in the swing phase and pre-swing phase, respectively, can be expected after using the proposed system. However, the results of the 3 week pilot clinical tests suggest that significant improvement is only observed for the push-off movement. It is assumed that the attention required to recognize the BF was beyond the ability of the patients. In this paper, a dual task (40 s walking and performing mental arithmetic at the same time) was conducted with the following conditions: no vibrations and providing BF to the lower back and the entire back. According to the results, the ankle joint angle of the paralyzed side at push-off under the entire back condition is statistically significant (p = 0.0780); however, there are no significant changes under the lower back condition (p = 0.4998). Moreover, the ankle joint angle of the paralyzed side at the initial contact is statistically significant with respect to the lower back condition (p = 0.0233) and shows a significant trend for the entire back condition (p = 0.0730). The results suggest that the limited attention capacity of hemiplegic patients fails to improve both dorsiflexion and push-off movements; moreover, ankle motion can be promoted if attention is concentrated on recognizing focalized vibratory feedback patterns.

Identification of Spring Coefficient for Heel Rocker Function Support Based on Estimated Dorsiflexion Torque.

In previous research, we have developed a high-dorsiflexion assistive robotic technology aiming for gait rehabilitation targeting on ankle dorsiflexion movement. A McKibben-type artificial muscle is applied to provide large dorsiflexion force while adding little weight to the device. This ensures the foot clearance before initial stance phase in gait. Meanwhile, a tension spring is deployed in series with the artificial muscle to support heel rocker function in loading response phase. Suitable spring coefficient for each individual differs according to ankle's dorsiflexion torque in loading response. An unsuitable spring would lead to knee deviation in this phase. In this study, we derived an identification equation to determine a suitable spring coefficient for individuals based on estimation of dorsiflexion torque required to support. An evaluation test on healthy objects was conducted, which shows no negative effects on participants' knee angles with the identified spring coefficient.