Adversarial Domain Adaptation for Stable Brain-Machine Interfaces

TL;DR

This paper introduces an adversarial domain adaptation approach for brain-machine interfaces that stabilizes neural decoding over long periods, reducing the need for recalibration and improving user experience.

Contribution

It proposes a novel adversarial domain adaptation network that outperforms existing methods in stabilizing neural signal decoding for BMIs with minimal data.

Findings

Adversarial domain adaptation outperforms CCA and distribution matching methods.

The proposed method requires fewer data points for effective adaptation.

Significant improvement in BMI stability over long-term recordings.

Abstract

Brain-Machine Interfaces (BMIs) have recently emerged as a clinically viable option to restore voluntary movements after paralysis. These devices are based on the ability to extract information about movement intent from neural signals recorded using multi-electrode arrays chronically implanted in the motor cortices of the brain. However, the inherent loss and turnover of recorded neurons requires repeated recalibrations of the interface, which can potentially alter the day-to-day user experience. The resulting need for continued user adaptation interferes with the natural, subconscious use of the BMI. Here, we introduce a new computational approach that decodes movement intent from a low-dimensional latent representation of the neural data. We implement various domain adaptation methods to stabilize the interface over significantly long times. This includes Canonical Correlation Analysis used to align the latent variables across days; this method requires prior point-to-point correspondence of the time series across domains. Alternatively, we match the empirical probability distributions of the latent variables across days through the minimization of their Kullback-Leibler divergence. These two methods provide a significant and comparable improvement in the performance of the interface. However, implementation of an Adversarial Domain Adaptation Network trained to match the empirical probability distribution of the residuals of the reconstructed neural signals outperforms the two methods based on latent variables, while requiring remarkably few data points to solve the domain adaptation problem.