Stroke patients employ robotics in gait research


By Roman Petrowski, Office of Communications

Dr. Fangshi Zhu - Exoskeleton-assisted rehabilitation paper
Fangshi Zhu, PhD

Recent research from Fangshi Zhu, PhD, postdoctoral research fellow in the Department of Physical Medicine and Rehabilitation, on the effects of exoskeleton-assisted rehabilitation for stroke survivors was published in the June edition of the Journal of Neural Engineering.

With their research, the authors found that the Ekso 1.1TM exoskeleton showed the potential to correct impaired gait patterns on the paretic leg and increase the motor coordination complexity to facilitate a normative gait pattern. However, the assistive torque from the Ekso could be aggressive sometimes to correct the patients’ limb to normal joint position from healthy subjects, which is outside of the realm that patients are used to. This results in increased paretic quadriceps activation as well as non-paretic hamstring activation,  causing patients to resist the assistance and hinder guidance.

“We found that while 15 sessions of Ekso-assisted gait training did not significantly improve the stroke subjects’ motor coordination, assistance from direct robot usage had a significant immediate impact on the stroke subjects’ lower-limb muscle synergy pattern,” the authors said. “This demonstrated the possibility of using a powered exoskeleton to augment impaired gait and the potential for helping stroke survivors regain their normal walking ability in a long-term exoskeleton-assisted rehabilitation program.”

Zhu’s paper is titled “Effects of an exoskeleton-assisted gait training on post-stroke lower-limb muscle coordination.” Contributing authors from the Department of Physical Medicine and Rehabilitation include Taimoor Afzal, PhD, postdoctoral research fellow; Shuo-Hsiu Chang, PT, PhD, assistant professor; and Gerard E. Francisco, MD, The Wulfe Family Chair of Physical Medicine and Rehabilitation and chief medical officer at TIRR Memorial Hermann. Contributors from the Center for Wearable Exoskeletons, NeuroRecovery Research Center at TIRR Memorial Hermann are Marcie Kern, PT; and Erin Fowkes, PT. Jose-Luis Contreras-Vidal, PhD, from the Department of Electrical and Computer Engineering at the University of Houston also contributed.

“Stroke is one of the leading causes of death in the United States and is a major cause of serious, long-term disability worldwide,” the authors said. “Post-stroke survivors suffer from neurological deficits and impairments that may cause several disabilities, e.g. diminished mobility and basic activities of daily living.”

A critical step in post-stroke rehabilitation is the recovery of the gait-related motor functions, caused by hemiplegia. Hemiplegia can be any number of abnormal features including asymmetric step times and step length, slowed gait velocity, impaired joint and posture control, muscle weakness, abnormal muscle tone, and abnormal muscle activation patterns, mostly affecting the paretic side of the patient and greatly affecting mobility and activities of daily living.

However, the current accepted methods of therapy can be taxing on both the patient, and the therapists. According to Zhu, conventional gait rehabilitation therapies can be very labor intensive, and involve two to three therapists to guide affected limbs.

“Recent advances in robot-assisted rehabilitation allow precise and automated training,” the authors said. “Compared to conventional therapy, the use of a robot enables longer training time, more precisely controlled forces delivery in repetitive exercises, and kinematics and kinetics monitoring during training, which therefore makes it increasingly popular in post-stroke gait rehabilitation.”

In the first stage of the research, Zhu and his team collected data from 11 able-bodied subjects (healthy group) and 10 persons with chronic post-stroke hemiparesis (stroke group) who participated in a single-visit treadmill walking experiment. Five of the stroke subjects (including three from the first stage) then participated in a longitudinal exoskeleton-assisted gait training.

In stage one, each subject walked continuously for 3-5 minutes on a  Biodex Gait Trainer treadmill at a self-selected, comfortable speed determined in a practice walk prior to data collection. During the walk, subjects wore gravity-compensating harnesses connected to an over-head body weight suspension frame to ensure safety and eliminate a potential loss of balance.

The second stage consisted of 10-15 sessions of exoskeleton-assisted walk training over a course of three to four weeks. In each session, patients walked over-ground with an Ekso 1.1TM exoskeleton for up to 50 minutes while guided by a physical therapist. The subjects were encouraged to walk continuously as much as possible, but short breaks were allowed.

“This study gave us some critical insight into how a powered exoskeleton affects the stroke subjects’ neuromuscular coordination during gait and demonstrated the potential to use muscle synergy as a method to evaluate the effect of the exoskeleton training,” the authors said.

“Acute and chronic hemiplegic and hemiparetic subjects who suffer from severe asymmetric gait coordination, serious foot drop, and reduced joint range of movement could benefit from the Ekso assistance,” the authors said. “Active assistance from the powered knee and hip actuators and passive assistance from the footplate could compensate desired task demand and constrain the stroke subjects’ impairment to minimize abnormal muscle activation and reinforce a normative motor coordination pattern.”

Zhu’s findings could lead to improved treatments for stroke patients while using the exoskeleton-assisted gait rehabilitation, while also providing a way to evaluate its effects, which could facilitate the development of future rehabilitation devices.