With Robots, a New Way to Understand Strokes

By JAMES WARREN

James Warren writes a column for The Chicago News Cooperative.

Consider the 1,001 little movements we take for granted, like reaching a few inches for a salt shaker or trying on new shoes.

Most involve simple reflexes and signaling from the brain to our arms, hands and fingers. But for many people — 750,000 have strokes each year in this country, and an estimated 60 percent of them suffer injuries that could include a partly paralyzed arm — those acts are nearly impossible.

Dr. Julius Dewald is trying to meld medicine, science and engineering in a path-breaking way to better understand such impairment and how robotic therapy might help people who have had strokes reach for a hamburger or pull on a fancy boot. It’s a coincidence that he works a few floors above the Salvatore Ferragamo store and T.G.I. Friday’s on Michigan Avenue.

Dr. Dewald is chairman of the department of physical therapy and human movement sciences at Northwestern University’s Feinberg School of Medicine. He was born in the Netherlands, raised in Mexico City and educated in Belgium and the United States.

He is overseeing a project that is changing the way people think about strokes. It is partly financed by the National Institutes of Health, the Department of Education and the American Heart Association. He and his 25-person team are trying to determine if electromechanical devices can more precisely measure impairment and accelerate what is now a belated, long and expensive rehabilitation.

We’ve all seen the stroke victim with an arm hanging down and slightly bent. It often tends to be in a fixed position, across the chest at a more than 90-degree angle.

Most stroke rehabilitation focuses on the legs and getting patients ambulatory as soon as possible so they can leave a hospital within two weeks, Dr. Dewald said. The emphasis is on walking, even though impaired arms tend to be in worse shape than impaired legs.

Much also remains unclear about arm impairment. That’s why Dr. Dewald and his team are trying to learn how independent control of joints is linked to one of the two brain hemispheres affected by a stroke.

If the damage is on the left side of the brain, the right arm is affected, and with it the ability to control a joint at a time. Even if patients can move the arm somewhat, they won’t have the independent control of joints in the elbow, wrist and fingers.

In Dr. Dewald’s laboratory this week, a man sat in a chair wearing what looked like a skull cap dotted with electrodes. Researchers attached 160 electrodes to assess brain activity in both hemispheres, enabling them to see brain signals with improved spatial resolution.

In another room, I participated in another part of their research, as I was hooked to a robot by wrapping my left arm in a mechanical sleeve. The device would change the weight of my left limb, moving it upward to give the sensation of being weightless, which requires far fewer brain signals to move the arm. Later the researchers added weight and created resistance to reaching, as if my arm were in a tub of molasses.

“I am essentially bringing you back from space to planet Earth,” Dr. Dewald said. “If the arm is weightless, you don’t need to use as much brain tissue. And without needing as many brain pathways, you can better control the shoulder and elbow joints, allowing them to move the arm and shoulder much closer to how you and I do.”

For a person who has had a stroke, just using shoulder muscles can deplete limited neural resources and make it impossible to extend an elbow or move fingers.

Over time, the project team hopes the robot can help patients better extend their arm — not completely, but more than they can now. Then the robot can add weight, making the limb heavier, so the patient can mirror a situation akin to living without the robot.

But the therapy must add more than the actual weight of the limb so it can be made strong enough for real-life tasks, like holding a gallon of milk or pushing open a door.

Dr. Dewald’s team just concluded a pilot study with 10 patients and is starting a second with 20. A comprehensive clinical trial would follow. The goal is to produce robotic devices that cost less than $15,000, affordable for small clinics and some individuals.

It all makes you hope the research goes well — especially if you’re still lucky enough to be able to hold a pencil or move your fingers one at a time to type a few words about the stricken.