Neural interface restores touch sensation after spinal cord injury

NEW YORK (Reuters Health) - A sensorimotor neural interface successfully restored touch sensation in a patient with quadriplegia resulting from spinal cord injury (SCI), researchers report.

“Neurotechnology and brain-computer interfaces are becoming an effective way to leverage residual neural signals for functional benefit following SCI, stroke, and several other dysfunctional states,” Dr. Patrick D. Ganzer of Battelle Memorial Institute, in Columbus, Ohio, told Reuters Health by email.

An estimated 50% of patients with a clinically complete SCI have a “sensory discomplete” SCI, where tactile stimuli that the patients cannot feel nevertheless evoke changes in cortical activity. Brain-computer interfaces (BCIs) can reanimate paralyzed muscles after SCI, but whether they can restore touch was unknown.

Dr. Ganzer and colleagues used a BCI could to simultaneously reanimate both motor function and touch sensation in a chronically paralyzed patient with a clinically complete SCI who could already use a BCI to move the hand.

While the patient was unable to perceive mechanical sensory stimuli below spinal level C6, sensory stimuli to the hand robustly modulated neural activity in the contralateral primary motor cortex (M1).

This residual sensory neural activity was reliably decoded from M1 using a support vector machine (SVM), and a vibrotactile array on the affected bicep enabled sensory feedback that restored conscious touch perception to a detection rate over 90%.

A modified grasp and release test demonstrated real-time sensorimotor demultiplexing where the participant was able to perform the task using the system but not without using the system, the team reports in Cell.

“It was initially surprising when we first discovered the subperceptual touch signal,” Dr. Ganzer said. “The participant has a very severe spinal cord injury that could have blocked this signal traveling from the hand to the brain. These results demonstrate that even a small contingent of spared spinal fibers can be leveraged for functional benefit. Lastly, the study’s overall results are interesting because the brain implant was not originally intended to record both touch and movement neural signals.”

“Participants that have a ‘clinically complete’ SCI may have a small contingent of spared fibers remaining that are still transmitting a neural signal,” he said. “Therefore, a very small quantity of spared fibers can potentially be leveraged for benefit, even though they might only be transmitting a faint signal.”

“Additionally,” Dr. Ganzer said, “the study’s findings can potentially inform future neural implant locations. Future neural interfaces can be placed in areas that encode a multitude of mixed neural signals that might be of value for the given technology. Regardless, information from neural interfaces should be maximized to enable new functional benefits in patients, even in new and unintended ways.”

He added, “The NeuroLife team at Battelle is currently working with our collaborators to develop a take-home BCI system. This would allow for the BCI to be used during activities of daily living outside of the laboratory setting. One of our recent achievements in this domain was a demonstration in the participant’s home using a portable miniatured version of the ‘muscle stimulation system.’ A computer tablet was able to control muscle stimulation to elicit hand movements during home activities.”

Dr. Andrew Jackson, professor of neural interfaces at Newcastle University, in Newcastle-upon-Tyne, U.K., told Reuters Health by email, “The study rests on a surprising finding: touch stimuli that are imperceptible to a spinal cord-injured participant can nevertheless be ‘decoded’ from brain activity in the motor cortex. Previous brain imaging studies have suggested such signals can reach the brain even after injury, but this is the first study to reveal them in the activity of individual brain cells.”

“There are two ways this (approach) could work in practice,” he said. “First, the decoded sensory signals could be used to trigger some kind of sensory substitute, in this case a vibrotactile cuff acting on a part of the body that the patient can consciously feel. Second, the sensory signals could directly influence the electrical muscle stimulation, thereby implementing a sensorimotor feedback loop similar to the unconscious reflexes that allow us to maintain our grip on an object without having to think about it.”

“While there are other ways to achieve this in principle, the nice thing about the current approach is that it only requires a single surgical implant to achieve both motor and sensory restoration,” Dr. Jackson said.

“These approaches are still at an early stage of development,” he said. “I anticipate seeing more studies like this one that find clever ways of exploiting and enhancing these surviving connections to restore function.”

SOURCE: Cell, online April 23, 2020.