||This workgroup will investigate Surface Electromyography (sEMG) data for rest position in collaboration with the workgroup "Regressing individual finger position from sEMG using a spike-based approach" using the Excitatory-Inhibitory network using the DYNAP-SE 2. The approach aligns with the studies on using balance networks to investigate epilepsy vs resting state. Following the speculation that the breakdown of the equilibrium between excitation and inhibition could lead to epilepsy.
Considering that resting and grasp are two orthogonal gestures, we would like to use this synthesized E-I spiking network as a simple binary classifier. The network should be able to control three motors on the prosthetic hand (Mia)
for the grasping gesture and "no gesture" (or reset) for the resting state.
The Dynamic Neuromorphic Asynchronous Processor (DYNAP-SE 2) is a mixed-signal Spiking Neuronal Network (SNN) Processor designed by the Neuromorphic Cognitive Systems Group at INI, University of Zurich and ETH Zurich.
Based on design principles taken from biological nervous systems, it uses analog signal processing and digital event-based communication. Each DynapSE2 chip has 1024 adaptive exponential integrate-and-fire neurons to emulate a spiking neural network. We will use DynapSE2 in combination with the Mia prosthetic hand to then trigger the intended gesture.
The main components we intend to investigate, are:
1. Network topology that supports processing an sEMG (analog-to-digital) on neuromorphic hardware.
2. E-I balance network with a population as small as 32 neurons on DYNAP-SE 2 that imposed various restrictions at the network level as well as computational level.
3. Tuning the network for motor cooperation, to trigger a grasping gesture.
4. If feasible, map the pre-trained network from other workgroups working on regression and classification of gestures using sEMG.