Improving the detection of sleep slow oscillations in electroencephalographic data

TL;DR

This paper introduces a new tool for labeling sleep slow oscillations and compares various algorithms for detecting them, finding that machine learning methods perform best.

Contribution

A custom tool for manual labeling of sleep slow oscillations and improved machine learning methods for their detection.

Findings

The custom tool reduced manual labeling time and enabled inspection of over 96,000 potential SO events.

Machine learning and deep learning algorithms outperformed traditional methods with up to 99.20% accuracy.

SO density and amplitude increased with sleep depth across all methods.

Abstract

We aimed to build a tool which facilitates manual labeling of sleep slow oscillations (SOs) and evaluate the performance of traditional sleep SO detection algorithms on such a manually labeled data set. We sought to develop improved methods for SO detection.

SOs in polysomnographic recordings acquired during nap time from ten older adults were manually labeled using a custom built graphical user interface tool. Three automatic SO detection algorithms previously used in the literature were evaluated on this data set. Additional machine learning and deep learning algorithms were trained on the manually labeled data set.

Our custom built tool significantly decreased the time needed for manual labeling, allowing us to manually inspect 96,277 potential SO events. The three automatic SO detection algorithms showed relatively low accuracy (max. 61.08%), but results were qualitatively similar, with SO density and amplitude increasing with sleep depth. The machine learning and deep learning algorithms showed higher accuracy (best: 99.20%) while maintaining a low prediction time.

Accurate detection of SO events is important for investigating their role in memory consolidation. In this context, our tool and proposed methods can provide significant help in identifying these events.

Figures