Wasim Q. Malik, Harvard Medical School, Boston, USA

Clinical investigation of neural prosthetics for people with severe SCI

Funded in: 2014, 2015, 2016


Back to overview

Problem: Movement commands from the brain fail to reach the limbs when the spinal pathway of downstream neural signals is disrupted due to SCI, resulting in the inability to move the limbs.

Target: Extraction of movement related signals in the brain for the control of a prosthetic device

Goal: High-performance neural prosthetics suitable for clinical use in rehabilitation of people with chronic SCI

Movement commands from the brain fail to reach the limbs when the spinal pathway of downstream neural signals is disrupted due to SCI, resulting in the inability to move the limbs. Providing an alternative pathway, a neural prosthetic device conveys movement-related signals generated in the brain to an assistive device, such as a robotic arm or computer cursor, restoring some degree of motor function. Current neural prosthetic prototypes provide impressive proof-of-concept demonstrations, but some practical issues yet remain.

The inherent variability of neuronal signal recordings poses a major challenge as it severely affects the longterm performance of a neural prosthetic. Action potential recordings from single-neuronal ensembles provide precise movement information, but are highly unstable. Signals recorded from larger volumes of brain tissue, such as local field potentials, are more stable but convey relatively coarse information.

 

The project team investigates the extraction of movement commands jointly from neuronal action potentials and local field potentials to combine the benefits of both. The aim is to develop a mathematical framework for optimal information combining from different types of movement-related neural signals. In an ongoing pilot clinical trial of the BrainGate neural prosthetic, the performance of the multi-signal approach will be tested with the help of tetraplegic participants.

It is expected that the joint signal processing approach will provide both precise control and longterm stability in neural prosthetics, helping to overcome one of the major challenges of this technology and advancing it in its translation to the clinic. High-performance neural prosthetics suitable for clinical use may play a key role in the rehabilitation of people with chronic SCI, helping them regain a degree of independence and productivity, and improving their quality of life.