Closed-Loop Neural Interfaces with Embedded Machine Learning

TL;DR

This paper reviews recent advances in embedding machine learning into neural interfaces, emphasizing design trade-offs and hardware efficiency, and introduces an optimized tree-based model for low-power neural signal classification.

Contribution

It presents a novel energy-aware, memory-efficient tree-based model that outperforms traditional models in neural signal classification tasks.

Findings

Oblique trees achieve higher accuracy with less energy consumption.

Model compression improves neural interface performance.

Effective for seizure, tremor detection, and motor decoding.

Abstract

Neural interfaces capable of multi-site electrical recording, on-site signal classification, and closed-loop therapy are critical for the diagnosis and treatment of neurological disorders. However, deploying machine learning algorithms on low-power neural devices is challenging, given the tight constraints on computational and memory resources for such devices. In this paper, we review the recent developments in embedding machine learning in neural interfaces, with a focus on design trade-offs and hardware efficiency. We also present our optimized tree-based model for low-power and memory-efficient classification of neural signal in brain implants. Using energy-aware learning and model compression, we show that the proposed oblique trees can outperform conventional machine learning models in applications such as seizure or tremor detection and motor decoding.