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

• Published in: IEEE sensors letters Vol. 9; no. 2; pp. 1 - 4
• Main Authors: Sartipi, Shadi, Andersen, Mie, Hauglund, Natalie, Kjaerby, Celia, Untiet, Verena, Nedergaard, Maiken, Cetin, Mujdat
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.02.2025; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accuracy; Artificial neural networks; autoencoder–decoder; Brain modeling; Convolution; Convolutional neural networks; Decoding; Distributed memory; electroencephalogram (EEG); Electroencephalography; Feature extraction; Filters; Long short term memory; long-short term memory (LSTM); Mice; Neural networks; REM sleep; Sensor applications; Sleep; sleep scoring; temporal transition; Training
• ISSN: 2475-1472; 2475-1472
• DOI: 10.1109/LSENS.2024.3523427

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 letter 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-REM 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.