Researchers Pursue Truly Functional Prosthetic Arm An international team of engineers and scientists is hoping to revolutionize prosthetic-arm technology. The ultimate goal is to build a strong, lightweight arm activated by neural impulses, which means it is controlled by an amputee's thoughts.
NPR logo

Researchers Pursue Truly Functional Prosthetic Arm

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
Researchers Pursue Truly Functional Prosthetic Arm

Researchers Pursue Truly Functional Prosthetic Arm

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript


This is ALL THINGS CONSIDERED from NPR News. I'm Rebecca Roberts.


And I'm Robert Siegel.

Throughout military history, advances in technology - from internal combustion engines to airplanes to robotics - have changed war strategy and affected war outcomes. In Iraq and Afghanistan, advances in body armor and battlefield medical treatment mean more soldiers are surviving their wounds.

ROBERTS: But more of those survivors are amputees, which have highlighted a somewhat lagging technology, prosthetic limbs, especially arms.

Historically, artificial legs have advanced more rapidly, in part because there's a bigger commercial market for legs. But also, the human arm is an incredible design challenge.

Now, funded by the U.S. Defense Department, a huge international team of engineers and scientists is hoping to revolutionize prosthetic arms technology. The ultimate goal is to build a strong lightweight arm that's actually controlled by the amputee's thoughts. The project's home base is at Johns Hopkins University.

Mr. BOB ARBITER(ph) (Johns Hopkins University): And this is literally what the patient's able to do through their EMG control. They can actuate all the different degrees of freedom, wrist flexion/extension, wrist rotation. And as actually for this hand, there's two different types of grasps.

ROBERTS: Bob Arbiter works with the computer program that reads the patient's nerve impulses as the amputee imagines moving his or her arm.

Mr. ARBITER: Just ask the patient to do what feels natural, literally move their phantom limb. And we're just going to take sort of a snapshot of what the signals look like. It's called pattern recognition. Literally, the computer is learning what the patient is trying to do, so it can recognize those. When it sees the signals again, it can command the arm to perform that way.

ROBERTS: As sites for revolutions go, the applied physics lab at Johns Hopkins is an unlikely one. In a nondescript gray office building between D.C. and Baltimore, the lab is littered with plastic body parts and machine shop spec sheets. It's an unusual collaboration of more than 30 labs, universities and private companies, authorized to the tune of $30 million so far by DARPA, the usually secretive Defense Advance Research Projects Agency.

Matt Kozlowski is one of the lead engineers.

Mr. MATT KOZLOWSKI (Lead Engineer, DARPA Program): It is so darn neat, the technologies that are enabling this thing and the work that's being pulled in from all these collaborators all over the world. These people are the finest in their field and we're having the opportunity to work with them and take some of the things that they've been developing and actually put them to good use.

ROBERTS: And they're making a huge step forward. One way, they're sending signals from the amputee's brain to the prosthetic arm through neuroconnectivity, involves a special surgery. First, the nerves from the patient's shoulder that used to run down the arm have to be rerouted to the patient's chest, once again connecting those nerves with healthy muscle.

The arm itself is non-invasive. Electrodes can be placed on the amputee's chest to pick up electrical impulses from the tiny contractions of the muscle and move the prosthetic arm. Stuart Harshbarger, who heads the project for Hopkins, showed us a video of a double amputee named Jesse Sullivan using the lab's first prototype.

Mr. STUART HARSHBARGER (Project Manager, Applied Physics Lab, Johns Hopkins University): It's one of my favorite clips of Jesse, stacking checkers. So, you know, even though he got this - first came, you know, pretty close. You know, it wasn't perfect so that will give him a little bit of a hard time about it. But then, you know, as you watch he very precisely placing it the second time around, he's not wanting to give up. He's pretty happy with himself and like, you know, and rightly so that's pretty precise movement and also, you know, pretty smooth movement actually in terms of moving that from there to there.

ROBERTS: Jesse Sullivan was a civilian electrical lineman who lost both arms in a work-related accident. Amazingly, the system not only sends Jesse's thoughts to the robotic arm, the arm sends sensation back. Jesse could actually get a sense of grip.

Mr. HARSHBARGER: We have a sensor in the prosthetic limbs system and as he grips an object, that amount of force is actually communicated to the little black device we call a tactor and it provides little pressure sensation in this area that he already perceives naturally as is coming from the ball of his hand. So he actually perceives that as if it's coming from the - his hand itself.

ROBERTS: He can tell how tightly to hold a plastic cup, for instance, to keep from crushing or spilling it. The lab began this project by studying the limits of currently available prosthetics. The most cutting-edge one available now is called the myoelectric arm. It responds to muscle movements in the amputee's stump. If the patient moves a shoulder muscle a certain way, the elbow bends for instance. Another muscle contraction makes the hand move.

And the myoelectric arm is heavy, around 14 pounds, so patients find it hard to learn and cumbersome. In fact, when this lab wanted a myoelectric arm to study, they found one on eBay, put up for sale by someone who had lost patience with the technology. Researchers talked to several amputees with similar frustrations and have returned to a lower-tech simple hook, or learn to live as a one-armed person.

Another limitation of current arms is degrees of freedom, engineering speak for movement. Your arm has more than 30 degrees of freedom, from flapping your shoulder to giving a thumbs up. Current prosthetic arms have three. This lab's first prototype had eight. Prototype II will be a bigger leap.

Mr. HARSHBARGER: We're going from eight degrees of freedom to 27.

ROBERTS: Again, project manager Stuart Harshbarger.

Mr. HARSHBARGER: And we're going from 11 pounds, for eight degrees of freedom to another 10 and half or so down to, you know, maybe something lower than nine. So we're continuing to drive, function up and reduce the overall weight.

ROBERTS: The lead engineer on the second prototype is Tom van Doren(ph). He picks up a key component - a shiny metal cylinder a few inches long with wires protruding from it, which will form the shoulder.

Mr. TOM VAN DOREN (Lead Engineer, Johns Hopkins University): This shoulder here is designed to match the size of a 50th percentile female shoulder and give the force of a roughly a 50th percentile male. So this particular piece that I have in my hand has what we refer to as two degrees of freedom, which means two motions, to bring your arm up forward out in front of you and then to bring your arm up to the side like you were doing a Jumping Jack.

ROBERTS: And a 50th percentile female is what, about 5-foot-4?

Mt. VAN DOREN: I'm not sure what her height is. I probably can quote the length of her humerus and the length of her fingers, but very small when you're designing the mechanical components. I think we are - all the mechanical engineers were very surprised when we finally started getting our components back from the machine shop at just how small they are.

The weights of this device for the two shoulder components that we have here is very similar to the human arm, and we'll achieve speeds and forces that are very similar to the human arm. So that right there is the challenge for Proto II - was getting the force out of this arm and the load capacity out of the arm in the same size and mass as a human arm.

We're not good enough mechanical engineers to design something that beats the human arm or trying to meet it.

ROBERTS: Of course, these stubborn design challenges are like catnip for mechanical engineers. It's working with amputees like test patient Jesse Sullivan that get them back in the lab day after day. Matt Kozlowski was lead engineer for Prototype I.

Mr. KOZLOWSKI: To see Jesse there as a bilateral amputee without anything on, and just seeing his chest, his torso, and realizing that he can't do anything without the assistance of a device such as this, it really is amazing because he can't open a doorknob, he can't go and get a drink out of the water fountain, things that you and I take for granted on, you know, completely, daily.

When Bob and I sat there in the field and worked with Jesse and realized, you know, he'd been sitting there for an hour and a half and he haven't even drink, you know, you'd need to give him a drink because he can't do it himself, and that, sort of, brought it all home as far as how important some sort of advancements in prosthetics are.

The key for the program is in the utility and neurologically interfacing it so that it's really a viable technology. You know, we can make robots all day long, but the reality of it in terms of usability, it's probably where we're going to make our breakthroughs and revolutionize to state-of-art.

ROBERTS: Prototype II is due for completion in July. And assuming all goes well, in the fall, DARPA will sign off on another $25 million to $30 million to take this project to completion in 2009. The ultimate goal: FDA approval of an arm that weighs about seven pounds that can pick up a 50-pound weight, maybe even type; an arm that will really help people return to their daily lives.

Copyright © 2007 NPR. All rights reserved. Visit our website terms of use and permissions pages at for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.