NeuroCom

Brain-machine interfaces connect the brain to an automated system without requiring any movement from the user. Movement is central to our interactions with machines (e.g., typing on a keyboard, grasping, and manipulating joysticks), leading to numerous applications of brain-machine interfaces in both civilian and defence domains. One can imagine intuitive and efficient control interfaces that could, for example, control prosthetics/exoskeletons or communicate with tools/vehicles in contexts where hands cannot be used (e.g., wearing gloves, using other tools, etc.). Moreover, these interfaces offer unique possibilities for human augmentation through movement augmentation. However, the need for surgical intervention to place electrodes near the brain's motor areas is a significant barrier to the development of these interfaces beyond specific medical applications. It becomes clear that the development of brain-machine interfaces must rely on the non-invasive identification of movement intentions, a challenge underscored by a specific research program of the DARPA (Defense Advanced Research Projects Agency). It exists several non-invasive approaches based on electroencephalography, magnetoencephalography, or surface electromyography. However, these techniques have several major limitations that likely explain the limited performance of interfaces using them. For example, surface electromyography only provides very indirect (and noisy) information about the neural drive sent to skeletal muscles. The breakthrough and originality of our project lie in detecting the movement command at the peripheral level rather than at the cortical level. Specifically, we will identify the neural code of movement at the spinal motor neurons (alpha) level, which constitute the final pathway of motor control. Our team pioneered the approach to identify the activity of spinal motor neurons by decomposing electromyographic signals, making these motor neurons the only neurons whose activity (discharge rate) can be measured non-invasively. Through four interconnected objectives, this project aims to develop the next generation of human-machine interfaces: • Objective #1: Identify motor neuron synergies controlling the hand in a control space involving varied mechanical and neural constraints. This task will help delineate the smallest units of movement control. • Objective #2: Develop an intuitive neural human-machine interface that decodes the movement command from motor neuron synergies. • Objective #3: Determine the level of plasticity of voluntary control of spinal motor neurons. This will involve understanding whether control units can be split into several independent control units, an essential element for movement augmentation. • Objective #4: Facilitate the transfer of these methods from the laboratory to the field by simplifying the measurement of myoelectric activity through the development of hybrid electrodes combining textile and electronic tattoo.