VA Maryland Health Care System
Researchers Use Robotics to Help Stroke Survivors
Researchers at the VA Maryland Health Care System have found that an adaptive approach to stroke treatment can help stroke survivors improve gait. Using a specialized ankle robot – Anklebot − researchers can train participants in various ways to improve control of their weakened leg, either by playing videogames from a seated position or during task-oriented locomotor training on a treadmill. This flexibility provides a versatile training tool that permits the researchers to customize therapy to meet the specific needs of participants, and that may offer exciting new options for improving gait and balance in stroke survivors. Initial results from the first participants in a two-year $199,900 VA Rehabilitation Research & Development Merit Pilot funded study suggest that the repetitive practice of walking on a treadmill with the Anklebot may also improve balance and propulsion by improving ankle control at key events during walking. Their preliminary research findings are forthcoming in the Proceedings of the Institute of Electrical and Electronics Engineers (IEEE) International Conference on Robotics and Automation (IEEE-ICRA). The IEEE-ICRA is the most prestigious of all robotics conferences with an average acceptance rate of just under 40 percent. The research team will present their work at this year’s IEEE-ICRA in Karlsruhe, Germany in May.
With more than 800,000 Americans suffering annually from strokes, it is the leading cause of long-term disability in this nation. And with the graying of America, the prevalence of stroke and its debilitating after-effects are expected to increase. The ability to work ranks high among the debilitating post-stroke effects, affecting individuals to various degrees from inefficient gait due to asymmetric step lengths and timing, poor propulsion from the affected side, reduced balance control, and causes a more inconsistent gaits overall. As part of a collection of studies conducted on stroke rehabilitation, VA Maryland Health Care System researchers are exploring robotics to better rehabilitate stroke survivors and to help them achieve functions that are as close to normal as possible.
"Drop foot is a common impairment after a stroke, caused by the weakness in the dorsiflexors and evertor muscles [muscles used in walking] that lift the foot," says Larry Forrester, PhD, director of the Baltimore VA Medical Center Human Motor Performance Laboratory and associate professor at the University of Maryland School of Medicine. "A major complication of drop foot is the increased likelihood of dragging the foot on the ground as the leg swings forward during stepping. Lack of control in the weakened ankle also can contribute to slapping of the foot after the heel strike, in addition to excessive inversion (rolling the ankle inward), which is a risk for injury. Using the Anklebot during gait training may help patients practice and learn how to control foot clearance, orientation at contact, and foot slap. This is possible because the device can independently provide supportive forces during the gait cycle to modulate stance, swing and specific sub-tasks within the gait cycle to better address the walking issues that appear post stroke."
"To address the diversity of walking deficits in those with stroke, we had to invent a novel systems-level control system; one that utilizes the "assist-as-needed" Anklebot in a deficit adjusted manner, to time robotic assistance to key functional deficits. This enables therapists to tailor robotic gait therapy to not only the severity, but also the type of gait deficit," said Anindo Roy, PhD, chief robotics engineer for the VA Maryland Health Care System’s Exercise & Robotics Center of Excellence and assistant professor at the University of Maryland School of Medicine, the latest paper’s lead author. "Our preliminary findings with this approach indicate that it can yield durable benefits in those with drop foot, weak propulsion and/or poor landing. To our knowledge, this is the first-of-its-kind approach in the field of contemporary lower extremity robotics that enables customizability of gait therapy across survivors with a wide range of deficit profiles," Roy says.
Just ask Vivian Elaine James. For James, the world changed April 2011 after she suffered a stroke. "I was totally shocked I had a stroke," says James, who, prior to the stroke, had owned a construction company that kept her physically active. "I did painting, renovating, dry wall installation, and other things," James says. The stroke, which afflicted her left side, stole her balance and capacity to move with ease, to function and complete many tasks she had taken for granted. "At the time, I had no idea what was happening to me. I thought I was having an anxiety attack, but apparently I had had some mini strokes and didn’t know it before the major one hit me."
Hope for James came from her participation in the study using the Anklebot, and this 52-year old credits improvements in her ability to walk directly to her experience with the device. "I’ve only participated in the study for three months, but those three months made a major difference in my life," James says. "I still need to be careful when I’m walking if the surface is not solid, but I don’t need assistance anymore. Using the Anklebot made a big difference in the strength of my left leg."
Roy, Forrester and the team are conducting an ongoing VA-funded study to compare outcomes of robotic treadmill training to the use of the Anklebot in both seated, computer-video aided interventions and treadmill walking. They are exploring the differences of the effect on ankle strength and walking since each approach targets specific deficiencies. During the seated approach, individuals use their weakened ankles to play a video game three times a week in one-hour long sessions. In the treadmill approach, individuals walk on the treadmill three times a week for up to 40 minutes per session.
"Our earlier work examined the effects of seated Anklebot in chronic stroke and found that 18 sessions over six weeks produced some measureable gains in floor walking speed and the underlying components of the gait cycle. We now need to compare the relative effectiveness of the seated versus the task-oriented approach on the treadmill, to see if they yield similar results. So far the treadmill locomotor approach seems very promising, but it is too early to draw sweeping conclusions," says Forrester, a principle investigator on the study.
During the treadmill approach, the team employs footswitches embedded in the individual’s shoes to detect the occurrence of key events during walking such as the feet rising and falling within the gait cycle, enabling the Anklebot to time assistance precisely to those events, and, over time, progressively wean robotic assistance so that the individual can move incrementally toward a more normal gait. "The Anklebot is unique in that, it is impedance-controlled allowing it to be programmed to apply more or less support to the effort, depending on the user’s performance and tolerance," Roy said. "It is also highly ‘backdrivable,’ meaning it can ‘get out of the way’ when appropriate, allowing patients to perform movements on their own."
For James, it's simpler: "Anyone who has had a stroke needs the Anklebot," she says.