The metatarsophalangeal (MTP) joints play critical roles in human locomotion. Functional restriction or loss of MTP joints will lead to lower walking speed, poorer walking balance, and more consumed metabolic energy cost compared with normal walking. However, existing foot orthoses are focused on maintaining the movement of the ankle joint, without assisting the MTP joints. In this paper, in order to improve the walking performance of people with lower limb impairments, a wearable powered foot orthosis (WPFO) which has actuated MTP joint is designed and constructed. Preliminary experiments on three nondisabled subjects demonstrated functionality and capabilities of the WPFO to provide correctly timed dorsiflexion and plantar flexion assistance at the MTP joint during walking. These results also suggest that the WPFO could offer promise in certain rehabilitation applications and clinical treatment.
Ankle-foot orthoses and prostheses have been studied widely over the past few decades for patients with walking disabilities caused by injuries or neurological and muscular pathologies such as trauma, stroke, cerebral palsy, spinal cord injuries, and muscular dystrophies. They are intended to enhance the basic walking function of patients by maintaining the range of motion of the ankle and foot [
One important issue existing in the research of powered ankle-foot orthoses and prostheses is that active assistance is applied only to the ankle joint. The metatarsophalangeal (MTP) joints are usually incorporated into the devices passively using an arc-shaped carbon-composite forefoot or a passive rotary joint [
Therefore, in this paper, we propose a wearable powered foot orthosis (WPFO) that consists of an actuated MTP joint. The powered MTP joint is intended to provide assistance in dorsiflexion and plantar flexion when the foot pushes off the ground during walking. In order to verify the functionality and performance of the WPFO, preliminary experiments on three nondisabled subjects are carried out.
The rest of this paper is organized as follows. Section
Human walking is the repetition of consecutive gait cycles in which the body is alternately supported by the right and left leg. A normal gait cycle presented in Figure
A gait cycle of human walking.
During walking, the movement of MTP joints mainly occurs in the sagittal plane. In the terminal stance subphase, the heel lifts up from the ground and the body rotates relative to the toes around the MTP joints; in other words, the MTP joints are dorsiflexed. After reaching the maximum dorsiflexion angle of about 30° in the middle of preswing subphase, the MTP joints are rapidly plantar flexed towards the neural position, as having been shown in Figure
The virtual model and prototype of the WPFO device are shown in Figures
The virtual model of the WPFO.
The prototype of the WPFO.
The control system hardware of the WPFO is composed of a controller, a servo drive, and two contact switches. For the sake of simplicity, a personal computer instead of a single-board microcontroller is adopted as the controller. One of the two contact switches is installed at the back of the heel plate, and the other one is installed on the forefoot plate near the pivot shaft. The signals of the contact switches are fed to the servo drive. The controller receives the feedback from the servo drive and sends commands to the drive based on the sensor status. The controller communicates with the drive via RS-232 protocol.
The control scheme of the WPFO is based on finite-state machine control, as illustrated in Figure
Working mechanism of the control system of the WPFO.
Reference trajectory of the MTP joint of the WPFO.
Three nondisabled subjects (mean ± standard deviation age 27 ± 3 yr, height 172 ± 8 cm, and weight 65 ± 12 kg) were recruited from Northwestern University to participate in the experiments. These subjects were experienced treadmill walkers and had no walking-related injuries such as muscle strains, joint sprains, or back injuries at the time of testing. And they all wore size 8-9 shoes in order to fit with the WPFO. All the subjects received written and verbal information about the experiment procedure and gave written informed consent prior to participation. The experiments were carried out in the Legs and Walking Lab of Shirley Ryan AbilityLab after the approval was obtained from the Institutional Review Board of Northwestern University.
In order to verify the functionality of the WPFO, subjects walked on a split-belt treadmill with embedded force plates (Fully Instrumented Treadmill, Bertec Corporation, Columbus, OH, USA) wearing the WPFO on the right foot. All subjects were instructed to walk at their self-selected speed under two conditions, that is, (1) without assistance torque and (2) with assistance torque. The order of experimental conditions was randomized across subjects. Under each condition, subjects completed 3 walking trials and walked for 2 minutes in each trial with a 2 min standing break inserted between two trials to avoid fatigue effect. Before the start of formal experiments, multiple practice trials were performed for subjects to acclimate to the devices.
During the walking trials, ground reaction forces (GRF) on the right foot were collected by the force plates at 1000 Hz. And three-dimensional motion of the lower limbs was captured by an 8-camera motion capture system (NaturalPoint Inc., Corvallis, OR, USA) at 100 Hz. To make them available for analysis and comparison, the collected data were processed using custom programs written in MATLAB (The Mathworks, Natick, MA, USA) in the following method. GRF and marker position data were first low-pass filtered using a fourth-order Butterworth filter with cutoff frequencies of 20 Hz and 6 Hz, respectively [
Based on the introduction in Section
Figure
Averaged vertical GRF and anterior/posterior GRF during treadmill walking under the two conditions.
The support impulse and propulsion impulse incorporating both the magnitude and duration of vertical GRF and A/P GRF were analyzed and presented in Figure
The support impulse and propulsion impulse during treadmill walking under the two conditions.
Figures
The right ankle moment over one complete stride for walking under the two conditions.
The right ankle power over one complete stride for walking under the two conditions.
In the above figures from Figures
The objective of this study was to design a novel wearable foot orthosis which had an actuated MTP joint to help patients to achieve dorsiflexion and plantar flexion of the MTP joint. From the experiments, we found that sufficient torque could be provided by a 200-watt electric servo motor that transmitted the power through a reducing mechanism. Pilot data collected from nondisabled subjects were used to demonstrate the function of the WPFO device. It was verified that the WPFO was able to provide both dorsiflexion and plantar flexion assistance during walking. The assistive capabilities of the proposed WPFO were most clearly illustrated by the vertical GRF and A/P GRF data during the walking trials with assistance torque. The increased second peak value of the vertical GRF and the increased peak of positive A/P GRF were indicative of larger push-off force that was provided by the WPFO for forward propulsion during the late stance phase. What is more, the timing of the GRF data and the moment and power of ankle joint showed small differences between the durations of stance phase in the two experimental cases, indicating that the WPFO could provide quick response and appropriately timed functional assistance.
Nowadays, plenty of people are suffering from lower limb disabilities caused by stroke, cerebral palsy, spinal cord injuries, or muscular dystrophies. The disabilities have severely affected their quality of life. Consequently, there is a growing demand for technological advances in orthotic and prosthetic systems, especially portable and powered devices. On the basis of our previous researches on the effects of the MTP joints on human walking performances, we proposed and constructed the novel WPFO that has a powered MTP joint. The WPFO provided a potentially new modality for applications in in-home assistive walking training and clinical rehabilitation or treatment. And hopefully, it may contribute to the improvement of the functional outcomes of daily training and treatment.
Although encouraging results are obtained from this study, there are still some important limitations to consider. Firstly, the WPFO device is a little bit heavy. To reduce the weight, the actuator and control system can be relocated on the leg or waist. Metal materials can also be replaced by composite materials with high strength and rigidity but low density. Secondly, the actuator adopted in this study is rigid. It will be better if compliant actuators like series elastic actuators or pneumatic artificial muscles are used, because inherent compliance is essential for human-machine interaction. Thirdly, the two contact switches can only detect the contact status between the ground and the two specific points on the sole. More force-sensitive resistors distributed beneath the sole can offer more reliable gait event detection during walking. Last but not the least, we only tested the WPFO on nondisabled subjects. More experiments have to be performed on patients with impairments to the MTP joints to further demonstrate the functionality of the device.
In this paper, the design and evaluation of a novel WPFO device with an actuated MTP joint are presented. The WPFO is proposed to help improve walking performance of patients with MTP joint impairments. The pilot data from three nondisabled subjects demonstrated the capabilities of the WPFO to provide functional assistance at the MTP joint during walking, which suggests that the WPFO could be potentially utilized in certain rehabilitation applications and clinical treatment. In the future, our research will be focused on the improvements to hardware structure and control scheme of the WPFO device.
The data used to support the findings of this study are available from the corresponding author upon request.
The authors declare that they have no conflicts of interest.
This research is funded and supported by the National Institutes of Health (NIH) under Grant R01HD082216 to author M. Wu, the China Scholarship Council (CSC) under Grant 201606120094 to author Y. Liu, and the National Key Research and Development Plan of China under Grant 2017YFC0806501 to author X. Zang.