Realtime-Capable Hybrid Spiking Neural Networks for Neural Decoding of Cortical Activity

TL;DR

This paper develops a real-time capable hybrid spiking neural network for neural decoding in brain-machine interfaces, optimizing for low power and small size, and demonstrates state-of-the-art performance on a primate reaching dataset.

Contribution

It introduces a novel, optimized SNN architecture for neural decoding that exceeds current benchmarks while maintaining low resource demands and real-time capability.

Findings

Achieved state-of-the-art accuracy on the Primate Reaching dataset.

Developed a real-time capable SNN model suitable for neuromorphic hardware.

Demonstrated potential for latency-free cortical spike train decoding.

Abstract

Intra-cortical brain-machine interfaces (iBMIs) present a promising solution to restoring and decoding brain activity lost due to injury. However, patients with such neuroprosthetics suffer from permanent skull openings resulting from the devices' bulky wiring. This drives the development of wireless iBMIs, which demand low power consumption and small device footprint. Most recently, spiking neural networks (SNNs) have been researched as potential candidates for low-power neural decoding. In this work, we present the next step of utilizing SNNs for such tasks, building on the recently published results of the 2024 Grand Challenge on Neural Decoding Challenge for Motor Control of non-Human Primates. We optimize our model architecture to exceed the existing state of the art on the Primate Reaching dataset while maintaining similar resource demand through various compression techniques. We further focus on implementing a realtime-capable version of the model and discuss the implications of this architecture. With this, we advance one step towards latency-free decoding of cortical spike trains using neuromorphic technology, ultimately improving the lives of millions of paralyzed patients.