French clinic hails results of neural implant-controlled full-body assistive exoskeleton, but no real-world trial yet
A 30-year-old Frenchman named Thibault paralysed from the shoulders down after a fall four years ago has taken his first steps in a laboratory thanks to an exoskeleton “suit” developed by biomedical research centre Clinatec and the University of Grenoble. The suit, which is controlled by brain implants, also helps Thibault to move his arms, but so far can only be used in laboratory with a tether attached to the ceiling to prevent it from falling.
The technology was designed by lead researcher Prof Alim-Louis Benabid, president of the Clinatec executive board, who was previously notable for developing the deep brain stimulation technique to treat the rigidity and tremors caused by Parkinson’s disease, for which he won the European Patent Office Inventors Award for research in 2016. Benabid and his colleagues published their research in The Lancet Neurology, which describes Thibault’s case as a proof of concept.
After his accident, which broke his spine and the level of the fourth and fifth cervical vertebrae at the base of the neck, Thibault (who has not revealed his surname) spent two years in hospital. The Clinatec team began their treatment by placing two implants, each incorporating 64 electrodes, onto the surface of his brain over the sections which control sensorimotor function to the upper limbs. These electrodes communicate wirelessly with computers to interpret the signals they collect.
Initially, Thibault used the implants to control characters in a computer game, but having gained experience from this he moved on to the 65kg exoskeleton. Walking was the first task he managed, with controlling the arms coming later and being more difficult because of the number of muscles involved and the complexity of coordinating all of their movements. Currently, the service does not incorporate assistive control of fingers, although this is a future target of Benabid’s team.
The Clinatec team stresses that this research is still in its early stages. The suit is extremely cumbersome and the patient needs to be strapped in, and currently it cannot prevent falls or allow the user to recover from them, hence the need for a full suspension harness. Moreover, the team is currently only using the signals from half of the electrodes, because this is the maximum they can manage to interpret in the 350ms needed to go from thoughts to movement; any longer and the suit becomes increasingly difficult to control. More powerful computers and AI systems would allow faster data analysis and the use of more channels.
This bottleneck is also currently limiting the utility of the arm controls. Currently, Thibault is 71 per cent successful in tasks requiring him to move his arms and rotate his wrists to touch targets.