AI brain implant allows a paralyzed man to regain movement.

      Researchers have successfully restored hand movement and the sense of touch in a man who was paralyzed from the chest down. The findings, published in Nature Medicine, indicate that the technology has partially rewired his nervous system.

      This system, referred to as a “double neural bypass,” originates from the Feinstein Institutes for Medical Research, which is the research division of Northwell Health. It merges a brain-computer interface with AI and electrical stimulation of the brain and spinal cord.

      The participant in this study, Keith Thomas, suffered a neck injury in a diving accident in 2020, leaving him with complete tetraplegia and unable to raise his hands to his face. He joined the three-year trial 13 months post-accident.

      After undergoing training, Thomas was able to feed himself and drink using his own hand. Over the course of 35 weeks, his right arm's strength increased by 86% and his left arm by 62%, according to the researchers. He was also able to scratch his nose and wipe his mouth without assistance.

      Another technique used, known as cortical mirroring, focused on restoring the sense of touch. After about 25 weeks, Thomas regained feeling in his wrist, which had been numb since his injury.

      The longevity of these improvements is significant. Many of the gains persisted even after the stimulation ceased, remaining evident during a recent follow-up more than two years later. The researchers suggest this indicates true rewiring or neuroplasticity, rather than a temporary fix.

      Chad Bouton, the study's lead author, remarked, “We’re not just bypassing the injury; we’re actually rewiring the nervous system.” He expressed his excitement about the progress in an interview with the Guardian, adding, “Being able to feel my sister’s hand, to pet my dog and feel her fur, these experiences that the injury took away have been restored,” Thomas noted.

      In terms of how it functions, surgeons placed five microelectrode arrays into Thomas’s brain during a 15-hour surgery. AI interprets his movement intentions and activates his forearm muscles to move his hand. Sensors in a 3D-printed brace then stimulate the sensory cortex to recreate the feeling of touch.

      The decoder demonstrated an accuracy rate of 84.6% over five months without the need for retraining. Thomas managed to lift empty eggshells without breaking them 87% of the time, even while engaged in conversation.

      This research contributes to a rapidly advancing area of brain-computer interfaces. Competing efforts have used implants to help restore speech, while others are pursuing wearable or non-invasive technologies, and China has approved its first commercial brain implant.

      Globally, around 15 million people live with spinal cord injuries, with those affected by tetraplegia often prioritizing hand function. The research team aims to conduct larger trials and is exploring the use of this system for other conditions, including stroke.