Building a Better Hand With the HAPTIX Project
July 2015 Issue
Modern upper-limb prostheses have gotten so mechanically advanced that users can peel the skin off a grape or fold origami cranes. What has remained elusive, though, is the sensation of touch that able-bodied people take for granted. Without sensation, even the most advanced and expensive artificial hand in the world will never be able to truly replace the loss for a patient with an upper-limb amputation.
"There are some very exquisite and mechanically advanced prostheses in the world, and on the mechanical side, most of the issues have been solved," says Dan Merrill, PhD, chief clinical scientist for Ripple, Salt Lake City. "But unless people have sensory feedback, they will treat their prosthesis like a tool hanging off their tool belt. There are a substantial amount of studies that demonstrate that once we overcome that hurdle of embodiment-allowing the prosthesis to become sensed as part of the self-people will start to use it as part of their normal life, not just picking it up when they need it."
The Defense Advanced Research Projects Agency (DARPA) has recognized this void for patients with amputations and in response has awarded prime contracts for Phase 1 of its Hand Proprioception and Touch Interfaces (HAPTIX) program. The ultimate goal of the program is to create a prosthetic hand that will be able to feel and move like a natural hand. Within four years, the agency seeks to have a U.S. Food and Drug Administration-approved prosthesis system ready for take-home trials.
To reach that goal, DARPA has awarded several Phase 1 grants. Those that are most successful will move on to Phase 2, which aims to integrate the technology into a complete HAPTIX test system.
One of the Phase 1 grants was awarded to Ripple under Coprincipal Investigators Merrill and Daniel McDonnall, PhD, director of research. The company hopes to use the knowledge it has gained from its prior work creating an implantable electrode array to help with motor function in order to incorporate sensation using a similar design. Ripple plans to utilize technologies from other organizations that have also received Phase 1 HAPTIX funding, including a tiny, implantable electrode array called the Utah Slanted Electrode Array, developed by a team at the University of Utah.
In the end, Merrill says, he hopes to have a device that would make the patient feel like he or she has a true hand, rather than just a useful tool.
"When that happens, they almost won't feel like an amputee anymore," he says.
The Problem of Living Without Limb Sensation
The problem of living without limb sensation is both functional and psychological, says Dustin Tyler, PhD, associate professor at Case Western Reserve University (CWRU), who is leading that team's Phase 1 HAPTIX effort.
"Essentially, amputees using traditional prosthetics are disconnected from whatever they are doing and don't have good interaction with their task," Tyler says. "They can't feel it when they touch someone's hand. That's what amputees want to do-to reconnect with people-and they lose that ability when they don't have a hand."
Chris Lake, CPO/L, FAAOP, clinical director, Lake Prosthetics and Research, Euless, Texas, says that a lack of sensation is the number one complaint he hears from his patients.
"There's not a day that goes by where there's not another example of how a patient would benefit from sensory feedback, how much they want to know how hard they are gripping," says Lake, who is providing clinical prosthetic consulting to Ripple.
That lack of feeling creates a host of problems, he says. Patients can squeeze a cup so much that it breaks or they might inadvertently squeeze a loved one's hand too much. To prevent these problems, they have to physically watch every motion they perform with their prostheses, and such vigilance mentally drains them after a while, he says.
To compensate, amputees tend to use their sound arms instead of their prostheses and, as a result, do not fully integrate the prostheses into their daily functional lives.
"If they can't feel it, they don't trust it," Lake says.
Beyond being useful, sensation is part of what connects people to their bodies, Merrill says. That sense of embodiment is just as important for a patient with an amputation as the functional aspect of sensation. Studies show that if a patient feels a prosthesis is a part of him or her, it won't feel like a tool that can be set aside when he or she is done using it.
"It just becomes a part of them, and in the long run, they no longer think of themselves as missing something," he says.
Having Sensation for the First Time
Already, researchers have started to bring the feeling of sensation back to patients on a limited basis. At CWRU, Tyler's team has figured out how to send patterns of electric signals between the brain and a prosthetic arm, giving patients sensation from their prostheses for the first time.
To do it, the CWRU team has equipped prosthetic hands with multiple sensors that send signals to electrodes placed outside nerves in patients' arms. Electrical signals from the prostheses' sensors activate the arm nerves, which send the signals to the brain.
"The brain can't tell the difference between a signal from a natural hand and a signal from our electrodes, so the brain thinks it's from the hand," says Tyler, who is also a biomedical engineer at the Louis Stokes Cleveland VA Medical Center.
Because the sensors can only detect pressure and not textures that come from a true sense of touch, the CWRU team has developed algorithms that convert the input from the sensors into different signals sent to the brain. The researchers have gotten the sense of touch so precise that, when activated, their two test patients can be blindfolded and have the ability to hold a cherry and remove its stem with their prostheses.
Without sensation and still using the blindfold, the patients crushed most of the cherries.
Aside from sensation, the team at CWRU stumbled upon another huge discovery for patients with amputations: Both test subjects said that their phantom limb pain was virtually eliminated after feeling sensation in their prostheses.
Tyler suspects that phantom limb pain comes from the brain not knowing how to respond to the lack of a limb. He thinks that without a hand to give the brain sensory information, the brain doesn't know how to fill in that missing information and, as a result, assigns it the last known sensation from the hand-which is usually a traumatic injury. With actual sensation, even with just the little bit of sensation the patients had while they were testing the devices in the lab, Tyler thinks their brains were able to reset their interpretations of what kind of feeling should exist for the hands.
"What we think is that they [the patients] just needed to send a message to their brains that their hands exist," he says. "It fills that gap for the brain and has a profound impact on the pain."
This discovery could significantly improve the lives of patients with amputations, Tyler says. Before going through testing, both patients said they had terrible phantom limb pain. One likened it to having a vise clamped down on his fist, and the other said it was like a nail being driven through his thumb.
Along with all the other benefits, Tyler says, the test patients had an entirely different approach to their devices once they felt sensation.
"When the sensation is turned off, they describe the devices as their prostheses," says Tyler. "As soon as you turn the sensation on, they describe them as working with their hands."
The Next Step: A Better Hand
Through its participation in HAPTIX, Ripple plans to develop two devices that will use implanted electrodes to better control the motor function of prostheses, as well as restore natural sensation. The company is developing the Myoelectric Implantable Recording Array (MIRA) device for motor function and the Stimulating and Recording Array (SARA) device to create sensation.
MIRA will contain 32 electrodes implanted into patients' residual muscles and connected to an electronic module under the skin. When the residual muscles contract, the electrodes will receive signals and pass them to the module, which will connect wirelessly to the prosthesis and move it in the intended fashion.
"With a computer processor, we decode those 32 channels and interpret the intent to move," Merrill says. "From that, we figure out that a person might want to open their fingers and rotate their wrist, and use that information to drive the motors in the prosthesis."
Merrill says that current artificial hands are moved by electrodes placed on top of patients' skin and that patients can only control one motion at a time, which feels unnatural.
With MIRA, Merrill says, patients will be able to move their prostheses in multiple ways simultaneously, which would feel more natural. This natural movement would help patients with amputations feel like their prostheses are more like real arms, he says.
"Feeling natural is the holy grail," Merrill says. "If we hit this home run of multiple simultaneous movements, then patients wouldn't have to think about moving their devices. They would just work the way they're supposed to, like their hands used to."
Having a lot of sensors would also be helpful, Lake says. He compares current designs to trying to send a text message using just two keys to enter all the different letters of the alphabet. Sending a message that way would take more time than having a key for every letter.
"The more muscle input we have, the more refined the prosthetic function becomes," Lake says. "You layer that with programming and algorithms to help decipher the raw input, thus creating the environment for prosthetic function to become as close to effortless and intuitive for the patient as possible. We all read quotes from research subjects about how they 'just thought about moving their arm and their arm moves,' and we're getting closer to helping patients do exactly that."
The implanted electrodes will be the key to the device, Merrill says. Currently, electrodes are placed on top of the skin and that can cause a variety of problems. Some patients don't have good skin coverage because of their injuries, and if they sweat or lose body mass, it can impact the connection between the electrodes and their skin.
Lake agrees that external electrodes aren't optimal. "The residual limb itself is dynamic," Lake says. "Its volume fluctuates throughout the day, and its shape changes as muscles contract and relax. Some parts of the day, there will be good electrode contact, and other parts of the day, not so much. When the electrodes aren't in good contact with the skin, the prosthesis may receive inadvertent signals and not operate as intended."
Each SARA device will have 64 electrodes implanted into sensory nerves. Information from sensors in the prosthesis will be processed and used to send the correct signals to the sensory nerves to enable the user's brain to correctly interpret touch and prosthesis position. They plan to use Utah Slanted Electrode Arrays, which are tiny and implanted into nerves, for the first human trials.
Though Tyler's team is working with Medtronic, Minneapolis, a medical technology provider, on a different device for the HAPTIX program, that device will also have implantable electrodes. Tyler says implantable electrodes that send wireless signals to prostheses are the next step to making a better artificial hand.
From time to time, amputees using prostheses with external electrodes have to take off their artificial arms and work to get a better connection between the electrodes on top of their skin and their devices. That won't be an issue if the electrodes are implanted, he says.
"Once implanted, those electrodes would be in the same place ten years later," Tyler says. "You wouldn't have to worry about all of those stability issues."
The HAPTIX program is highly competitive, and all of the teams are working to move on to the next phase of funding.
At Ripple, Merrill says that the team is already deep into the development of MIRA, which was envisioned by McDonnall over five years ago. Merrill hopes to have approval for clinical trials within 15 months. From there, they hope to use what they learn from that device and move on to develop the SARA. If Ripple qualifies for Phase 2 funding, its full system with both devices would still be a little under two years away from clinical trials, he says.
Other teams, including Tyler's, are using some of the same technology but have slightly different approaches.
Whatever the outcome, the HAPTIX program seeks to develop a device that would benefit patients with upper-limb amputations. While it won't be a "real" hand for them, it may be the next best thing, Tyler says.
"A more doable goal is to make the hand good enough so the users forget they lost [their hands] in the first place."
Maria St. Louis-Sanchez can be reached at .