After nearly 60,000 votes, STAT readers have spoken: A prosthetic arm that can “feel” is 2016’s most innovative idea.
Scientists at the University of Pittsburgh and the University of Pittsburgh Medical Center who worked on the system are the winners of the first STAT Madness bracket, which pitted 32 innovative discoveries published in peer-reviewed journals last year against one another to find the best new ideas in science and medicine.
The research, published in Science Translational Medicine, describes researchers’ latest work toward developing a prosthetic arm with a brain-computer interface.
There are two sides to a brain-computer interface: interpreting brain activity to make something happen — like moving a prosthetic limb — or giving the brain information from the body or from a prosthetic. In this study, researchers focused on the latter.
The study subject was Nathan Copeland, a man with quadriplegia, meaning all four of his limbs are paralyzed and he has no sensation in them. Scientists implanted an array of electrodes just below the surface of Copeland’s brain and connected them to a prosthetic arm (not attached to Copeland’s body) which was equipped with sensors.
By precisely placing the brain electrodes, researchers were able to recreate the sensation of Copeland’s own hand being touched when a corresponding area of the prosthetic was touched.
In the study, researchers demonstrated that he was correctly able to identify which finger on the prosthetic arm researchers touched while blindfolded. Copeland also gave them feedback on how natural (or unnatural) the touch felt, which will be used to help future iterations of such devices.
Those advances, coupled with other progress in brain-controlled limbs, promise to eventually make a brain-computer interface that can both send signals out to a limb and receive sensations back. Closing this loop would very helpful — if not necessary — for someone to do things that we rely on sensory feedback to achieve, like use the arm to button a shirt or hold a pen.
Robert Gaunt, an assistant professor at the University of Pittsburgh and coauthor of the published research, said that winning the competition was meaningful because it was an opportunity to show the world how far the technology has come. The devices still have a long way to go, he said, but “I hope that the people who see this … get a sense of what could be possible in the not-so-distant future.”
The team still faces still some major engineering and science challenges — to say nothing of the regulatory hurdles the technology would need to clear to move into clinics. First of all, their study is based on just one patient’s experience; other patients with different injuries may have different results. And for now there must be wires between the array in Copeland’s brain and the arm itself; turning that into a wireless connection is a crucial step for the technology to be ready for wider use. There still need to be more studies to determine how long the electrodes can stay in a person’s brain before the quality of the signals they can send or receive deteriorates to the point of uselessness.
And there is clearly room to improve the sensation stimulations, too — Copeland reported that the vast majority of the sensations were only “possibly natural.”
Most of the project’s funding has come from the Defense Advanced Research Projects Agency, whose main interest in the project is the impact it could have on the quality of life for veterans who have lost a limb, Gaunt said. DARPA has had a dedicated prosthetic program, called Revolutionizing Prosthetics, since 2006. Results from that program have also been promising, including a similar project to produce natural-like sensations from a prosthetic arm.
In the final round of STAT Madness, Jennifer Collinger, Sharlene Flesher, and Gaunt’s team beat out a discovery from the University of Michigan about the way bacterial communities change in the lungs of patients in intensive care units. Those findings could someday be combined with rapid DNA sequencing technology to catch sepsis earlier and possibly guide personalized treatments.
A second award this year went to Tufts University for a silk-based system for stabilizing blood samples. The STAT Madness Editor’s Choice category honors the innovation deemed most exciting by STAT editorial staff.
The research, led by David Kaplan and published in the Proceedings of the National Academy of Sciences, describes how silk fibroin — a protein found in silkworm cocoons — can be mixed with blood samples to allow them to be transported to a lab without refrigeration. The silk protein protects some useful proteins in blood samples, like those used for kidney function tests as well as cancer biomarkers, from water or heat that could cause them to degrade. Currently, a sample either needs to be dried on a piece of cardboard or kept refrigerated until it reaches the lab.
Based on the results of tests run on silk-stabilized blood, the silk doesn’t seem to be interfering with the proteins or changing them. “That’s what we’ve been fairly diligent in looking at since the technology has moved from Tufts,” said Jonathan Kluge, the director of research and development at Vaxess, which has licensed the technology from Tufts.
Vaxess is currently looking for funding, possibly through the National Institutes of Health’s SBIR grant program for small businesses, to create and improve a field-ready prototype, as well as for partners to commercialize the research.
Meanwhile at Tufts, Kaplan’s lab has continued to explore other potential uses for silk, including as a way to preserve HIV drugs.