From Our Print Archives

Man vs. Machine

Robotic devices can improve stroke patients' upper-extremity mobility

Vol. 24 • Issue 17 • Page 30

Stroke and Robotics

It may sound like a 10-year-old's dream - working with robots - but for the therapists at the Burke Rehabilitation Center and Ohio State University's Wexner Medical Center, it's all in a day's work.

The core research at Burke, located in White Plains, N.Y., is the physical rehabilitation of the neurological patient. "We're purely a research institute, and that positions us to put a lot of effort into that," explained Dylan Edwards, PT, PhD, director of the non-invasive brain stimulation and robotics lab at Burke. The lab is one of the few programs at Burke Medical Research Institute to use human subjects, and complements a large pre-clinical research program.

At the non-invasive brain stimulation and robotics lab, researchers use interactive motion technology robots designed by MIT engineers. There are two: the planar robot -which includes grasp and release - and the wrist robot.

Since 2009, Edwards has led an NIH trial that is funded through 2016. Because of the strict guidelines of clinical trials, participation is limited to first-time, ischemic stroke patients with right side weakness who do not have a history of seizures or a pacemaker.

Trial participants get 20 minutes of brain stimulation followed by 1 hour of upper-limb exercises while hooked up to the robots. They repeat the process three times per week for 12 weeks. Goals include improving their accuracy and upper-extremity range of motion.

The nature of repetition - over 1000 repetitive motions per session - forces their brains to re-wire the connections. Subjects must perform many repetitions of directed, voluntary movement, which they simply can't do with a traditional therapist.

"These robots deliver more therapy," said Edwards. "Robotic therapy of the upper limbs reduces impairment."

Through the trial, researchers like Edwards are testing whether brain stimulation primes the brain prior to movement. Preliminary data, which he termed "promising so far," showed that even a single stimulation session with robotic wrist practice can lead to improvement in motion.

According to Edwards, studies have shown that brain stimulation alone can be useful tool. Yet as an emerging therapy not yet approved by the FDA, patients must enroll in a clinical trial to reap the benefits. The devices, however, are already endorsed by the American Heart Association and approved overseas by the European equivalent of the FDA.

Researchers at Burke are still trying to understand the specifics of how brain stimulation combined with robotics improves function in stroke patients. They aim to better understand which patients respond best and why, refine therapy techniques to deliver more useful treatments, and give feedback to the robotics industry on what matters to physical therapists, so ultimately engineers can design better machines.

Turning Research into Rehab

The strict guidelines for enrollment in the NIH-funded study mean that the majority of interested subjects miss out on the benefits. Avrielle Rykman, MA, OTR/L, clinical robotics research coordinator at Burke, had an idea for a fee-for-service based program for patients ineligible for Edwards' study, but who could be helped by robotic therapy.

"Our goal is to bring the research we do into rehab," she said.

The Restorative Neurology Clinic at Burke hosts a six-week robotics therapy program for patients with neurological issues that inhibit function. A significant difference from the controlled-study is the workshops do not use brain stimulation, only robotics.

But the planar robot and the wrist robot function in the same way as the NIH study. Therapists record patients' movements with the video games when they are engaged in therapy. Then the patients repeat the movements with the robots assisting if necessary.

There is an alternating robot protocol. One session, the therapy focuses on the shoulder and elbow; the next session, it focuses on the wrist, for 18 total sessions.

Patients complete flexion, extension and rotation of the shoulder and elbow while affixed to the appropriate machine. The planar robot initially moves the paralyzed limb, but as function improves, the patient starts the movement.

The wrist robot also strengthens wrist and forearm rotation. If the patient cannot move a joy-stick to control the video game, the robot guides their motion. As they progress through therapy, the games and required movements get progressively harder.

Archive ImageA

Therapists receive feedback on flexor synergy patterns, motion and strength. Rykman and colleagues pull up measures and graphs to show patients how they've progressed. Rykman discourages patients from participating in more sessions on the same day, as the sessions at Burke are tiring. Yet many of the patients have been phased out of traditional rehab due to insurance caps, and are still seeking improvements.

In the first workshop, patients showed improvement in arm function, regardless of how long ago their injury occurred. Anecdotally, therapists got reports from the patients and their families on how their arm function increased. Each patient in the initial workshop has re-enrolled for more treatment at the Restorative Neurology Clinic.

"We try to foster a social environment for patients and their families," noted Rykman.

In the future, the clinic hopes to offer aid to patients who can't afford workshops, which cost $5400 per non-NIH study patient for each six-week session, and it is seeking foundation support.

Game Changer

Despite common misconceptions, therapists do not need big budgets to use robotics to make a difference in stroke patients' mobility.

"The problem with conventional robotic protocols platforms is that they are large, costly, and often require specialized expertise," explained Stephen Page, PhD, MS, MOT, OTR/L, FAHA, associate professor at Ohio State University Wexner Medical Center's Division of Occupational Therapy in Columbus.

Additionally, many patients don't have access to the large research hospitals that have the budgets and the manpower for traditional robotics.

Page and his colleagues set out to find how can therapists and patients can meaningfully integrate robotics into activities of daily living. They studied 16 hemorrhagic and ischemic stroke patients with limited upper-extremity mobility, an average of six years post-stroke.

The subjects performed repetitive tasks such as pushing, eating and lifting 30 minutes a day, three days a week, for an eight-week session. Half performed the activities without any external aid, and half performed them while wearing a portable robotic device. Just as with the study at Burke, this was the only rehab the patients received for the duration of the session.

In this therapy, electrodes on the bicep and triceps of the stroke-affected arm are connected to a portable robotic band worn over the forearm and elbow. The patient flexes their bicep to bring a cup to their mouth - a motion some stroke patients can't fully complete. The robot amplifies the signals of extension and flexion, helping the patient complete the motion. Therapists can program the robot via an app to determine how much assistance it supplies each patient.

"This particular robot is a game changer," said Page. It costs a few thousand dollars and an orthotist and prosthetist can custom-build it to the specifications of each patient's arm.

"One of the exciting things about this study is that robotic therapy works just as well as conventional therapy in people who are years post-stroke, and without excessive cost," explained Page. "Plus, the robot lets people perform valued activities as opposed to being strapped into a device and remaining seated, as is the case with most platforms."

Researchers at Wexner Medical Center administered The Fugl-Meyer, Canadian Occupational Performance Measure, and Stroke Impact Scales before intervention, as a baseline and again after the eight-week study. Both the robotics group and the non-robotics group had their Fugl-Meyer score increased by nearly the same amount.

The group that used the robotic arm showed greater improvements in all but one measure of the Canadian Occupational Performance Measure and Stroke Impact Scale subscales, including a 12.5-point increase on the Stroke Impact Scale recovery subscale.

In addition to the numeric data, Page observed, "Study participants who trained with the robotic arm also reported feeling stronger and more positive about the rehabilitation process."

"We know that perception can affect willingness to do rehab," he added. Data suggests the mere presence of the bionic arm can convince patients to work harder on their recovery. The study suggested that a tough-to-treat population, such as one many years post-stroke with moderate arm impairment, can benefit from robotics, when traditional therapies fail to make an improvement.

Page's team has received approval and funding for an NIH study that will look at three groups - one that only uses the bionic arm, one that does not use the bionic arm, and one that receives both conventional and robotic therapies. They will use kinematics to quantify the movement changes that can't be seen with the naked eye.

"As a field, we need to embrace new technologies because they will enhance therapy," said Edwards. "We need to learn how to add these to practice." 

Danielle Bullen is on staff at ADVANCE. Contact:


Email: *

Email, first name, comment and security code are required fields; all other fields are optional. With the exception of email, any information you provide will be displayed with your comment.

First * Last
Title Field Facility
City State

Comments: *
To prevent comment spam, please type the code you see below into the code field before submitting your comment. If you cannot read the numbers in the below image, reload the page to generate a new one.

Enter the security code below: *

Fields marked with an * are required.

View the Latest from ADVANCE


Search Jobs


Back to Top

© 2017 Merion Matters

660 American Avenue Suite 300, King of Prussia PA 19406