CHICAGO — Elizabeth Halsne, CPO, research prosthetist at the Center for Bionic Medicine at the Rehabilitation Institute of Chicago, presented strategies to collect high-quality surface electromyography data from an amputee’s limb to control powered lower limb prostheses at the American Academy of Orthotists and Prosthetists Annual Meeting and Scientific Symposium.
The purpose of these studies was to record muscle activity from the residual limb, then add that input from the user to decode their intended motion.
“We were focused on not only high-quality, but consistent EMG [electromyography] data that can be collected within the prosthesis itself,” Halsne said.
Halsne and colleagues demonstrated that gel liners with embedded electrodes and low-profile systems, integrated directly into the socket, can successfully provide this data.
Elizabeth Halsne
“We have attempted different approaches,” she said, “and have successfully recorded EMG and used it to improve control of powered prostheses, both at the transtibial and transfemoral levels.”
Study participants performed a number of ambulation tasks, including walking on different terrains at varying speeds, stair ascent and descent, standing, sitting and navigating ramps. They also controlled the prosthesis in a seated position to ensure the control was viable across a wide range of activities.
Tactic one
The first tactic employed self-adhesive electrodes applied directly to the skin of the residual limb. The electrodes were attached to wire leads to monitor the muscle signals.
“The advantage of using the self-adhesive electrodes is that you have reliable contact with the residual limb because it is adhered directly to it,” she said. “This affords the most high-quality EMG recording.”
However, this method presented some practical challenges.
“Not only is there a time intensive process of localizing muscle sites and applying each pair of electrodes individually, but it would be difficult to fit all these sensors and wires inside a socket. It is not a feasible option for use with a prosthesis,” she said.
“With this in mind, we were faced with the challenge of finding the best way to collect EMG information within the socket for individuals with transtibial amputations,” Halsne said.
Tactic two
The second approach was tested with a cohort of eight individuals with transtibial amputation, both male and female, with trauma and infection etiologies. They all used pin locking gel liners for prosthetic suspension and interface.
Elastomeric gel liners are embedded with
fabric electrodes.
Image: Elizabeth Halsne
Elastomeric gel liners were embedded with conductive silver fabric electrodes and leads. To do this, conductive leads were attached to the bare internal fabric of an ALPS South liner prior to the gel injection, so the finished product insulates most of the leads with gel and leaves the desired electrode locations exposed.
“Once the liner has undergone the standard gel injection process, all that is left are these small areas of silver fabric and we collect EMG from these electrode sites,” Halsne said.
This method also offered reliable skin contact, as the liner presses the electrode against the limb, so it is not susceptible to loss of contact from changes in limb shape.
Benefits of this system included repeatable electrode sites and easy use because the liner is donned in a reliable and familiar fashion. Drawbacks surfaced when these gel liners were used for an extended time, according to Halsne.
Participants with compromised skin integrity experienced issues at some electrode sites after walking for several hours, such as irritation at the edges of the fabric electrodes. Another issue noted with this design was that the silver fabric became susceptible to corrosion over time.
“It is actually the sweat that is corroding the fabric,” she said. “The silver fabric corrodes, the signal degrades and we no longer have high-quality EMG.”
Tactic three
The final method needed to offer a solution for transfemoral amputees, according to Halsne. Fifteen individuals, both male and female, with transfemoral amputations participated in using EMG control with their suction sockets. “We didn’t feel like gel liners were a practical solution for them,” she said.
Electrode domes were mounted through the socket wall and attached to wire leads on the other side. These were applied in test sockets and definitive laminations with flexible inner liners.
Advantages of this method included repeatable electrode sites and easy use since this system, like the liners, is also contained within the prosthesis without additional donning burden on the user.
As expected, disadvantages noted were loss of contact with skin and signal noise with motion.
“There are volume and muscle shape changes over the course of different activities that result in the skin pulling away from the domes,” Halsne said. “Unlike liners, we don’t have anything holding these electrodes against the skin except for the socket wall.” However, with advanced signal processing, the EMG data can still be decoded to recognize the participant’s intended motion.
Future studies
Halsne said although these methods could potentially enhance the control of powered prostheses, further study is needed to improve the quality of EMG data within a prosthesis.
“We need additional refinements so we can ensure that we are recording useful EMG information instead of noise,” Halsne said. “We have parallel development to create options for different populations that achieve this goal and preserve the ease of use of these systems for our patients.”
One improvement is to eliminate the silver fabric as the contact against the skin, but keep it as a flexible conductive lead that is insulated from exposure to sweat, she said.
Future research for socket-mounted domes includes comparing the size, material and profile, and improving the signal transmission between the skin contact and the processor.
“We hope by doing this we can enhance the control for these prosthetic users, so that they end up with a more accurate and reliable experience with powered lower limb devices,” she said. — by Shawn M. Carter
Disclosure: Halsne has no relevant financial disclosures.
