Classification of Local Field Potentials using Gaussian Sequence Model

TL;DR

This paper develops a spectrum-based classification method for local field potentials (LFPs) recorded from macaque prefrontal cortex, using Gaussian sequence models and Fourier coefficients with shrinkage, achieving high decoding accuracy for eye movement goals.

Contribution

It introduces a novel LFP classification approach based on minimax estimators and Gaussian sequence models, emphasizing the importance of phase information and systematic feature selection.

Findings

High decoding accuracy achieved on LFP data

Fourier coefficients with shrinkage outperform power spectrum features

Phase information is crucial for classification

Abstract

A problem of classification of local field potentials (LFPs), recorded from the prefrontal cortex of a macaque monkey, is considered. An adult macaque monkey is trained to perform a memory-based saccade. The objective is to decode the eye movement goals from the LFP collected during a memory period. The LFP classification problem is modeled as that of classification of smooth functions embedded in Gaussian noise. It is then argued that using minimax function estimators as features would lead to consistent LFP classifiers. The theory of Gaussian sequence models allows us to represent minimax estimators as finite dimensional objects. The LFP classifier resulting from this mathematical endeavor is a spectrum based technique, where Fourier series coefficients of the LFP data, followed by appropriate shrinkage and thresholding, are used as features in a linear discriminant classifier. The classifier is then applied to the LFP data to achieve high decoding accuracy. The function classification approach taken in the paper also provides a systematic justification for using Fourier series, with shrinkage and thresholding, as features for the problem, as opposed to using the power spectrum. It also suggests that phase information is crucial to the decision making.