With $6.3 million in DARPA funding, Intel, Brown and other partners will work on building the AI-driven hardware and software needed to treat spinal cord injuries.
When someone suffers a severe spinal injury, their brain’s electrical signals can get cut off from their muscles, leading to paralysis. It’s a devastating problem, especially since the human body cannot regenerate severed nerve fibers. But with the help of the right AI-driven technologies, medical professionals may eventually be able to help spinal injury victims regain muscle control and sensation.
With a new $6.3 million grant from the US Defense Advanced Research Projects Agency (DARPA), a team of Brown University researchers is leading a two-year effort to develop this technology.
“We know that circuits around a spinal lesion often remain active and functional,” David Borton, assistant professor at Brown’s School of Engineering and a researcher at the University’s Carney Institute for Brain Science, said in a statement.
Borton is leading the effort to develop and test an “intelligent spinal interface” that could specifically help restore limb movement and bladder control for people who have suffered spinal cord injuries. Bladder control is the top concern reported by people with spinal cord injuries.
“This exploratory study aims to build the toolset—the mix of hardware, software and functional understanding of the spinal cord—to make such a system possible,” he said.
The two-year study will bring together Brown’s medical researchers, AI experts from Intel, physicians from Rhode Island Hospital and partners from Micro-Leads Medical, a company that has developed high-resolution spinal cord stimulation technology.
Intel will provide the hardware, software and research support needed for developing AI and machine learning tools that could decode spinal signals recorded from the spine. As signals travel down the spinal cord above an injury site, they could be used to drive electrical stimulation below the injury. Signals coming up the spinal cord from below the injury site could be used to drive stimulation above it. Researchers will record motor and sensory signals from the spinal cord and use artificial neural networks to learn how to communicate the right motor commands.
To collect this data, surgeons at Rhode Island Hospital will implant electrode arrays into volunteers with spinal cord injuries. The implanted device will record and stimulate the spine as the patients participate in standard physical therapy.
Ultimately, researchers how to demonstrate their device can target the neural circuits that influence limb movement and bladder control.
The device used in this phase of the project will rely on an external computer system to decode the spinal signals. The aim is to eventually develop a fully-implantable device for long-term use, which could allow the severed nerves to communicate in real time.
By Stephanie Condon for Between the Lines