A Biohybrid Setup for Coupling Biological and Neuromorphic Neural Networks

Abstract

Developing technologies for coupling neural activity and artificial neural components, is key for advancing neural interfaces and neuroprosthetics. We present a biohybrid experimental setting, where the activity of a biological neural network is coupled to a biomimetic hardware network. The implementation of the hardware network (denoted NeuroSoC) exhibits complex dynamics with a multiplicity of time-scales, emulating 2880 neurons and 12.7 M synapses, designed on a VLSI chip. This network is coupled to a neural network in vitro , where the activities of both the biological and the hardware networks can be recorded, processed, and integrated bidirectionally in real-time. This experimental setup enables an adjustable and well-monitored coupling, while providing access to key functional features of neural networks. We demonstrate the feasibility to functionally couple the two networks and to implement control circuits to modify the biohybrid activity. Overall, we provide an experimental model for neuromorphic-neural interfaces, hopefully to advance the capability to interface with neural activity, and with its irregularities in pathology.

Neurostimulation stabilizes spiking neural networks by disrupting seizure-like oscillatory transitions

“…Additional future work could investigate how in vitro neuromodulatory stimuli can be optimized so as not to induce oscillatory activity, with our analysis here providing initial evidence www.nature.com/scientificreports/ that stimulation with a sufficiently high frequency may accomplish this through selectively activating inhibitory cells in a non-rhythmic fashion. The use of bio-hybrid systems [92][93][94][95] could provide an useful tool in such research. A paramount use of these mathematical findings is towards the articulation of a theoretical understanding of how neuromodulation acts to mitigate seizure onset, and in turn potential pathways through which these treatments might be improved.…”

Memristive synapses connect brain and silicon spiking neurons

“…An intriguing method for implementing the synaptors involves using the standardised interface of the Internet, which has been previously trialled for non-synaptic network communications [14][15][16] . We thus instantiated our example of synaptor-linked circuit in a geographically-distributed manner.…”

Periodicity Pitch Perception

“…A promising candidate is Spikey with its PyNN application programming interface, which allows execution of NEST and NEURON code (Pfeil et al, 2013). NeuroSoc seems to be the ideal candidate because NMDA, AMPA, and GABA channel kinetics are supported (Mayr et al, 2015;Keren et al, 2019). An accelerated analog neuromorphic hardware system emulating NMDA-and calcium-based non-linear dendrites is a promising candidate too (Schemmel et al, 2017).…”