LG-Sleep: Local and Global Temporal Dependencies for Mice Sleep Scoring

TL;DR

LG-Sleep is a novel deep neural network that captures local and global temporal dependencies in EEG signals for accurate, subject-independent sleep scoring in mice, even with limited training data.

Contribution

Introduces LG-Sleep, a deep neural network architecture that effectively models local and global temporal features for mice sleep scoring, enhancing generalization and performance.

Findings

LG-Sleep outperforms conventional neural networks in sleep scoring accuracy.

The model generalizes well across different subjects and limited training samples.

It effectively captures temporal transitions in EEG signals for sleep stage classification.

Abstract

Efficiently identifying sleep stages is crucial for unraveling the intricacies of sleep in both preclinical and clinical research. The labor-intensive nature of manual sleep scoring, demanding substantial expertise, has prompted a surge of interest in automated alternatives. Sleep studies in mice play a significant role in understanding sleep patterns and disorders and underscore the need for robust scoring methodologies. In response, this study introduces LG-Sleep, a novel subject-independent deep neural network architecture designed for mice sleep scoring through electroencephalogram (EEG) signals. LG-Sleep extracts local and global temporal transitions within EEG signals to categorize sleep data into three stages: wake, rapid eye movement (REM) sleep, and non-rapid eye movement (NREM) sleep. The model leverages local and global temporal information by employing time-distributed convolutional neural networks to discern local temporal transitions in EEG data. Subsequently, features derived from the convolutional filters traverse long short-term memory blocks, capturing global transitions over extended periods. Crucially, the model is optimized in an autoencoder-decoder fashion, facilitating generalization across distinct subjects and adapting to limited training samples. Experimental findings demonstrate superior performance of LG-Sleep compared to conventional deep neural networks. Moreover, the model exhibits good performance across different sleep stages even when tasked with scoring based on limited training samples.