Effective and Efficient Intracortical Brain Signal Decoding with Spiking Neural Networks

TL;DR

This paper introduces a novel spiking neural network framework with local synaptic stabilization and channel-wise attention, achieving higher accuracy and significantly lower energy consumption in intracortical brain signal decoding for BCIs.

Contribution

The study presents LSS-CA-SNN, a new SNN architecture with innovative stabilization and attention mechanisms, plus a data augmentation method, improving BCI decoding performance and energy efficiency.

Findings

LSS-CA-SNN outperforms state-of-the-art ANNs in accuracy.

Achieves 14.78-43.86 times energy savings.

Demonstrates effectiveness on invasive macaque datasets.

Abstract

A brain-computer interface (BCI) facilitates direct interaction between the brain and external devices. To concurrently achieve high decoding accuracy and low energy consumption in invasive BCIs, we propose a novel spiking neural network (SNN) framework incorporating local synaptic stabilization (LSS) and channel-wise attention (CA), termed LSS-CA-SNN. LSS optimizes neuronal membrane potential dynamics, boosting classification performance, while CA refines neuronal activation, effectively reducing energy consumption. Furthermore, we introduce SpikeDrop, a data augmentation strategy designed to expand the training dataset thus enhancing model generalizability. Experiments on invasive spiking datasets recorded from two rhesus macaques demonstrated that LSS-CA-SNN surpassed state-of-the-art artificial neural networks (ANNs) in both decoding accuracy and energy efficiency, achieving 0.80-3.87% performance gains and 14.78-43.86 times energy saving. This study highlights the potential of LSS-CA-SNN and SpikeDrop in advancing invasive BCI applications.