ICIIBMS 2017, Track 2: Artificial Intelligence, Robotics and Human-Computer Interaction, Okinawa, Japan
Design of a Pressure Sensitive Ankle-Foot Orthosis
for FES-aided Gait Training
Eun-Young Lee Myeong-Hyeon Heo Dongho Kim
Department of ICMC Department of ICMC Tianjin Normal University
Convergence Technology, Convergence Technology, Tianjin, People's Republic of China
Program of Sports IT Convergence Program of Sports IT Convergence and
Soongsil University Soongsil University Soongsil University
Seoul, Republic of Korea Seoul, Republic of Korea Seoul, Republic of Korea
[email protected] [email protected] [email protected] Abstract— In this paper, we suggest a gait training system for
the foot drop patients. The intent of this system is to control II. SYSTEM DESIGN
electrical stimulation pattern by utilizing the plantar pressure
distributions as a source of the real-time gait event detection. The A. System Overview
design of a prototype of the ankle-foot orthosis to test this This FES-aided gait training system stimulates an ankle
concept is described. In the future, this gait assistive system could dorsiflexion at the appropriate time during the gait cycle. The
be an alternative rehabilitation device for the foot drop patients.. system stimulates dorsiflexor muscles at the appropriate time
during the gait cycle. During the swing phase of gait, the
Keywords— foot drop; ankle-foot orthosis; functional electrical
stimulation; Orthotic devices
system electrically stimulates the common peroneal nerve that
signals the ankle to dorsiflex. The system comprises a hinged
ankle-foot orthosis, a pair of pressure sensitive footplate, an
I. INTRODUCTION electrical stimulation device and a smartphone. Fig. 1
Foot-drop(or dorsiflexor weakness) results from stroke [1] illustrates the configuration of the system.
or other conditions and injuries such as brain trauma [2] or an First, each footplate sends the data from pressure sensors
incomplete paralysis [3]. It prevents the patients from lifting wirelessly to the smartphone. And then, the smartphone detects
their foot and toes properly during walking, affecting their the state of gait and extracts the gait feature based on received
general mobility. As a result, the patients are reluctant to walk data. When it is determined that muscle contraction is
a long distance and their quality of life and confidence can be necessary, the smartphone wirelessly transmits the stimulation
negatively affected. pattern reflecting the patient's gait feature to the electrical
Traditionally, foot-drop patients are treated with bracing stimulation device. Lastly, the electrical stimulation device
using an ankle foot orthosis (AFO). An AFO is usually made stimulates the nerve via surface electrodes. The stimulated
of plastic, which wraps behind the calf and under the foot to nerve activates the ankle dorsiflexion muscles to lift the foot at
passively assist ankle dorsiflexion. Rigid AFOs improve static the right time.
balance, while impaire dynamic balance during a gait test [4].
And one of the most effective treatments for foot-drop is
functional electrical stimulation (FES). FES uses electrical
impulse to the common peroneal nerve in order to elicit the
desired muscle contraction [5].
In recent years, commercially available sensor-based
wearable FES systems for foot drop have been released [6-8].
Because these systems just detect gait events using simple
pressure sensor or tilt sensor, they cannot provide a
personalized electrical stimulation for more effective treatment.
If the FES system can analyze the patient's gait pattern by
collecting data from not only the affected side but also the non-
affected side, personalized electrical stimulation can be
provided according to his or her own gait parameters.
The objective of this paper is to describe the design of an
ankle-foot orthosis for gait training, which assists the foot-drop
patient’s ankle dorsiflexion during gait through plantar
pressure-based electrical stimulation. Fig. 1. Configuration of the system.
978-1-5090-6664-3/17/$31.00 ©2017 IEEE
243
ICIIBMS 2017, Track 2: Artificial Intelligence, Robotics and Human-Computer Interaction, Okinawa, Japan
B. Ankle-Foot Orthosis For auto stimulation adjustment, we defined several
The orthotic tibial and foot piece components are custom- temporal gait parameters including the duration of gait cycle,
fabricated with light, flexible and firm materials and are the step count, step pace, stance time, swing time and average
structural elements of the system. A bi-channel adjustable pressure. To calculate temporal gait parameters, it is essential
ankle joint connects the foot piece to the tibial piece on the to detect gait events. The temporal gait parameters can be
medial aspect and is used to provide minimal support during easily calculated by gait event detection. The smartphone
swing phase to prevent foot-drop. Velcro straps secure the application adjusts the stimulation parameters in response to
AFO to the leg and foot so that the electrodes are located over the temporal gait parameters while stimulating the common
the motor point. The pressure sensitive footplate is easy to peroneal nerve during the gait.
install and remove from the orthosis. The orthosis is easy to
combine and separate with the affected side footplate. III. CONCLUSIONS
We designed a pressure sensitive AFO system that controls
C. Pressure Sensitive Footplate electrical stimulation on dorsiflexor based on gait events to
prevent foot-drag during walking. This system could provide a
The pressure sensitive footplate is a portable pair of insole stabler and faster gait to the foot-drop patient without fear of
which consists of a plantar pressure pad and a mini electronic falling.
module. The commercial systems to date have tended to utilize the
A plantar pressure pad consists of a thin, planar, flexible data from only one or two sensor(s) to control the stimulator. In
and dielectric material that arrays thirty-sensors at strategic contrast, the system suggested in this article uses plantar
anatomical positions and is used to obtain the high-solution pressure data from thirty-sensors for personalized stimulation.
plantar pressure map from feet during the gait. A mini However, without verification experiments, the system must be
electronic module transmits stored sensor data converted from improved, as it might not be acceptable for clinical situations.
pressure voltage signals into resistance (ohm) respectively to Further studies will be needed to evaluate the utility of this
the smartphone via Bluetooth. newly-developed AFO system with various groups, including
healthy subjects and foot-drop patients.
D. Electrical Stimulation Device
The dual-channel FES device used in this system delivers ACKNOWLEDGMENT
electrical pulses during gait. Two electrodes and a simulation
unit are integrated on the velcro strap for AFO's tibial piece. This work was supported by the National Research
The electrodes are self-adhesive surface electrodes and located Foundation of Korea (NRF) grant funded by the Korea
over the common peroneal nerve and the tibialis anterior government (Ministry of Education, MOE) (No.
muscle to provide ankle dorsiflexion. 2017R1D1A1B04035651).
Additionally, a Bluetooth-enabled FES device is required to
connect with a smartphone. And the stimulation unit allows the
REFERENCES
application to adjust the stimulation parameter: pulse width,
pulse frequency, stimulus intensity, ramp-up time, ramp-down [1] P. T. Cheng, C. L. Chen, C. M. Wang and W. H. Hong, "Leg muscle
activation patterns of sit-to-stand movement in stroke patients," Am J
time and inter-pulse interval. Phys Med Rehabil, Vol. 83, pp.10-16, Jan 2004.
[2] W.C. Walker and T.C. Pickett, "Motor impairment after severe traumatic
E. Smartphone application brain injury: A longitudinal multicenter study," J Rehabil Res Dev, Vol.
Through the ergonomic GUI-based appllication, the users 44, pp.975-982, 2007.
(clinicians, therapists and patients) can operate the mobile [3] M. Wirz, H. J. van Hedel, R. Rupp, A. Curt and V. Dietz, "Muscle force
and gait performance: relationships after spinal cord injury," Arch Phys
application without difficulty. The application allows the user Med Rehabil, Vol. 87, pp.1218-1222, Sep 2006.
to: (ϸ) set the pressure threshold grades and reference impulse
[4] D. Cattaneo, F. Marazzini, A. Crippa and R. Cardini, "Do static or
before the training; (Ϲ) real-time display the gait status; (Ϻ) dynamic AFOs improve balance?," Clin Rehabil, Vol. 16, pp.894-899,
real-time observe and adjust the stimulation parameters. 2002.
[5] D.N. Rushton, "Functional electrical stimulation," Physiol Meas, Vol. 18,
We classified the thirty-sensors of the footplate into nine pp.241-275, Nov 1997.
regions: hallux, toes, medial metatarsal phalangeal joint (MTP),
[6] WalkAide & Foot Drop. (n.d.). Retrieved July 28, 2017, from
central MTP, lateral MTP, medial midfoot, lateral midfoot, http://www.walkaide.com/patients/Pages/default.aspx
medial heel and lateral heel for gait event detection. In this [7] What is the L300 Foot Drop System? (n.d.). Retrieved July 28, 2017,
system, a gait cycle is divided into eight phases: initial contact, from https://www.bioness.com/Products/L300_for_Foot_Drop/What_is
loading response, mid stance, terminal stance, pre-swing, initial _It.php
swing, mid swing and terminal swing from the terminology [8] Functional Electrical Stimulation (FES): MyGait — Ottobock. (n.d.).
developed by the Rancho Los Amigos Medical Center [9]. The Retrieved July 28, 2017, from http://www.ottobock.co.uk/orthopaedic-
gait phase detection algorithm judges whether the current rehabilitation/solution-overview/mygait/
pressure grades of each region satisfies a unique pressure [9] C.A. Downey, " Observational gait analysis handbook," Professional
Staff Association of Rancho Los Amigos Medical Center, pp. 252-256,
pattern of each normal gait phase. 1989.
244
