Auto-adaptive Neuromorphic Brain Machine Interface: toward fully embedded neuroprosthetics

Introduction

Nearly 746,000 people sustain a spinal cord injury every year, with dramatic human, societal, and economical costs, leading to impairment or even complete loss of motor functions.

Motor Brain-Machine Interfaces (BMIs)

Motor Brain-Machine Interfaces (BMIs) translate brain neural signals into commands to external effectors. BMIs raise hopes that limb mobility may be restored, providing patients with control over orthoses, prostheses, or over their own limbs using electrical stimulation.

Challenges in Neuroprosthetics

In spite of spectacular results, taking neuroprosthetics into daily practice has proven difficult. Currently, neuroprosthetics are restricted to assisted trials in laboratories and require regular retraining of a decoder in a supervised manner within controlled environments. They include various components:

• Recording device
• Antennas
• Base station
• Computers connected to effectors

These components are complicated to install and use.

Project Overview

Building on the consortium's unique experience in clinical chronic BMIs, the project will address major methodological and technological breakthroughs to achieve the first assistance-free motor neuroprosthetics system.

Objectives of the NEMO BMI Project

The NEMO BMI project will conduct the exploration of assistance-free and easy-to-use portable neuroprosthetics, including:

1. Wireless neuronal activity recorder
2. Real-time neuronal activity decoder based on integrated technologies
3. Spinal cord stimulator

Usability Improvement

A first objective is the crucial improvement of usability by introducing an auto-adaptive framework to train the decoder in an adaptive manner during the neuroprosthetics' unsupervised use.

Brain-Guided Spinal Cord Stimulation

Brain-guided spinal cord stimulation activating patients' limbs with an automatic stimulus pattern optimization is the second project objective.

Miniaturized Embedded Solutions

A third objective is the exploration of miniaturized embedded solutions by taking advantage of a novel neuromorphic hardware architecture.

Clinical Trials

NEMO BMI technologies will be studied offline and online in two ongoing clinical trials and will be critical to specify the next-generation assistance-free BMI.