Intel and Brown University Are Developing an AI System to Help Paralyzed Patients
Spinal cord injuries alter someone's life forever, and they can happen to anyone. When someone suffers a serious spinal cord injury, they most typically become paralyzed from that point down, or up.
The reason behind this devastating paralysis is because the human body isn't able to regenerate severed nerve fibers. The brain loses its signal to alert muscles to move, and paralysis occurs.
However, with the help of artificial intelligence (AI) and new technologies, the medical field is working toward helping patients with spinal cord injuries to move again.
A team of scientists from Intel and Brown University in the U.S. has linked up to work on AI-driven technology in a bid to assist paralyzed people to feel and move freely.
What will the research entail?
Offered a grant of $6.3 billion from the U.S. Defense Advanced Research Projects Agency (DARPA), the team of researchers, led by Brown University, are embarking upon a two-year project.
David Borton, lead researcher of the project and assistant professor at Brown's School of Engineering, said: "We know that circuits around a spinal cord lesion often remain active and functional."
Borton and his team aim to develop AI-driven technology — an 'intelligent spinal interface' — that will allow patients with spinal cord injuries to move their limbs once more and regain bladder control. Bladder control is reported to be one of the main concerns for spinal cord injury sufferers.
Borton continued, "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."
To begin with, the team, which consists of Brown medical researchers, AI experts from Intel, physicians from Rhode Island Hospital, and partners from Micro-Leads Medical, will use a device, implanted into patients' spinal cords, that is controlled by external hardware.
Intel will bring the software, hardware, and research support to the table. AI and machine learning tools will be developed to process the signals traveling up and down the spinal cord above the injury site.
In reverse, signals moving up the spinal cord from beneath the injury site could potentially be used to stimulate movement above the site.
In order to understand how to communicate the right motor commands, the team will record motor and sensory signals directly from the spinal cord. The team will collect data from the Rhode Island Hospital by implanting electrode arrays into volunteer patients' spinal cords.
These patients will undergo regular physical therapy, during which their implanted device will record and stimulate the spine.
The final aim is to create a fully-implantable device that works on a long-term basis. The device would enable the severed nerves to communicate immediately upon receiving the brain's electrical signals, just like those without spinal cord injuries.