Application of Machine Learning to Sleep Stage Classification

• Published in: 2021 International Conference on Computational Science and Computational Intelligence (CSCI) Vol. 2021; pp. 349 - 354
• Main Authors: Smith, Andrew, Anand, Hardik, Milosavljevic, Snezana, Rentschler, Katherine M., Pocivavsek, Ana, Valafar, Homayoun
• Format: Conference Proceeding Journal Article
• Language: English
• Published: IEEE 01.12.2021
• Subjects: artificial intelligence; Artificial neural networks; Brain modeling; Computational modeling; Data models; Electroencephalography; electrophysiology; machine learning; neuroscience; Sleep; sleep-scoring; Wires
• ISSN: 2769-5654; 2769-5670; 2769-5654
• DOI: 10.1109/CSCI54926.2021.00130

Sleep studies are imperative to recapitulate phenotypes associated with sleep loss and uncover mechanisms contributing to psychopathology. Most often, investigators manually classify the polysomnography into vigilance states, which is time-consuming, requires extensive training, and is prone to inter-scorer variability. While many works have successfully developed automated vigilance state classifiers based on multiple EEG channels, we aim to produce an automated and openaccess classifier that can reliably predict vigilance state based on a single cortical electroencephalogram (EEG) from rodents to minimize the disadvantages that accompany tethering small animals via wires to computer programs. Approximately 427 hours of continuously monitored EEG, electromyogram (EMG), and activity were labeled by a domain expert out of 571 hours of total data. Here we evaluate the performance of various machine learning techniques on classifying 10-second epochs into one of three discrete classes: paradoxical, slow-wave, or wake. Our investigations include Decision Trees, Random Forests, Naïve Bayes Classifiers, Logistic Regression Classifiers, and Artificial Neural Networks. These methodologies have achieved accuracies ranging from approximately 74% to approximately 96%. Most notably, the Random Forest and the ANN achieved remarkable accuracies of 95.78% and 93.31%, respectively. Here we have shown the potential of various machine learning classifiers to automatically, accurately, and reliably classify vigilance states based on a single EEG reading and a single EMG reading.