WaveSleepNet: An interpretable deep convolutional neural network for the continuous classification of mouse sleep and wake

MLS-Net: An Automatic Sleep Stage Classifier Utilizing Multimodal Physiological Signals in Mice

Over the past decades, feature-based statistical machine learning and deep neural networks have been extensively utilized for automatic sleep stage classification (ASSC). Feature-based approaches offer clear insights into sleep characteristics and require low computational power but often fail to capture the spatial-temporal context of the data. In contrast, deep neural networks can process raw sleep signals directly and deliver superior performance. However, their overfitting, inconsistent accuracy, and computational cost were the primary drawbacks that limited their end-user acceptance. To address these challenges, we developed a novel neural network model, MLS-Net, which integrates the strengths of neural networks and feature extraction for automated sleep staging in mice. MLS-Net leverages temporal and spectral features from multimodal signals, such as EEG, EMG, and eye movements (EMs), as inputs and incorporates a bidirectional Long Short-Term Memory (bi-LSTM) to effectively capture the spatial-temporal nonlinear characteristics inherent in sleep signals. Our studies demonstrate that MLS-Net achieves an overall classification accuracy of 90.4% and REM state precision of 91.1%, sensitivity of 84.7%, and an F1-Score of 87.5% in mice, outperforming other neural network and feature-based algorithms in our multimodal dataset.

Sleep-Deep-Learner is taught sleep-wake scoring by the end-user to complete each record in their style

Sleep-wake scoring is a time-consuming, tedious but essential component of clinical and preclinical sleep research. Sleep scoring is even more laborious and challenging in rodents due to the smaller EEG amplitude differences between states and the rapid state transitions which necessitate scoring in shorter epochs. Although many automated rodent sleep scoring methods exist, they do not perform as well when scoring new datasets, especially those which involve changes in the EEG/EMG profile. Thus, manual scoring by expert scorers remains the gold standard. Here we take a different approach to this problem by using a neural network to accelerate the scoring of expert scorers. Sleep-Deep-Learner creates a bespoke deep convolution neural network model for individual electroencephalographic or local-field-potential (LFP) records via transfer learning of GoogLeNet, by learning from a small subset of manual scores of each EEG/LFP record as provided by the end-user. Sleep-Deep-Learner then automates scoring of the remainder of the EEG/LFP record. A novel REM sleep scoring correction procedure further enhanced accuracy. Sleep-Deep-Learner reliably scores EEG and LFP data and retains sleep-wake architecture in wild-type mice, in sleep induced by the hypnotic zolpidem, in a mouse model of Alzheimer's disease and in a genetic knock-down study, when compared to manual scoring. Sleep-Deep-Learner reduced manual scoring time to 1/12. Since Sleep-Deep-Learner uses transfer learning on each independent recording, it is not biased by previously scored existing datasets. Thus, we find Sleep-Deep-Learner performs well when used on signals altered by a drug, disease model, or genetic modification.

Sleep-Deep-Net learns sleep wake scoring from the end-user and completes each record in their style

Sleep-wake scoring is a time-consuming, tedious but essential component of clinical and pre-clinical sleep research. Sleep scoring is even more laborious and challenging in rodents due to the smaller EEG amplitude differences between states and the rapid state transitions which necessitate scoring in shorter epochs. Although many automated rodent sleep scoring methods exist, they do not perform as well when scoring new data sets, especially those which involve changes in the EEG/EMG profile. Thus, manual scoring by expert scorers remains the gold-standard. Here we take a different approach to this problem by using a neural network to accelerate the scoring of expert scorers. Sleep-Deep-Net (SDN) creates a bespoke deep convolution neural network model for individual electroencephalographic or local-field-potential records via transfer learning of GoogleNet, by learning from a small subset of manual scores of each EEG/LFP record as provided by the end-user. SDN then automates scoring of the remainder of the EEG/LFP record. A novel REM scoring correction procedure further enhanced accuracy. SDN reliably scores EEG and LFP data and retains sleep-wake architecture in wild-type mice, in sleep induced by the hypnotic zolpidem, in a mouse model of Alzheimer's disease and in a genetic knock-down study, when compared to manual scoring. SDN reduced manual scoring time to 1/12. Since SDN uses transfer learning on each independent recording, it is not biased by previously scored existing data sets. Thus, we find SDN performs well when used on signals altered by a drug, disease model or genetic modification.

Selective Continuous Positive Airway Pressure Withdrawal With Supplemental Oxygen During Slow-Wave Sleep as a Method of Dissociating Sleep Fragmentation and Intermittent Hypoxemia-Related Sleep Disruption in Obstructive Sleep Apnea

Obstructive sleep apnea (OSA) is considered to impair memory processing and increase the expression of amyloid-β (Aβ) and risk for Alzheimer’s disease (AD). Given the evidence that slow-wave sleep (SWS) is important in both memory and Aβ metabolism, a better understanding of the mechanisms by which OSA impacts memory and risk for AD can stem from evaluating the role of disruption of SWS specifically and, when such disruption occurs through OSA, from evaluating the individual contributions of sleep fragmentation (SF) and intermittent hypoxemia (IH). In this study, we used continuous positive airway pressure (CPAP) withdrawal to recapitulate SWS-specific OSA during polysomnography (PSG), creating conditions of both SF and IH in SWS only. During separate PSGs, we created the conditions of SWS fragmentation but used oxygen to attenuate IH. We studied 24 patients (average age of 55 years, 29% female) with moderate-to-severe OSA [Apnea-Hypopnea Index (AHI); AHI4% > 20/h], who were treated and adherent to CPAP. Participants spent three separate nights in the laboratory under three conditions as follows: (1) consolidated sleep with CPAP held at therapeutic pressure (CPAP); (2) CPAP withdrawn exclusively in SWS (OSA_SWS) breathing room air; and (3) CPAP withdrawn exclusively in SWS with the addition of oxygen during pressure withdrawal (OSA_SWS + O₂). Multiple measures of SF (e.g., arousal index) and IH (e.g., hypoxic burden), during SWS, were compared according to condition. Arousal index in SWS during CPAP withdrawal was significantly greater compared to CPAP but not significantly different with and without oxygen (CPAP = 1.1/h, OSA_SWS + O2 = 10.7/h, OSA_SWS = 10.6/h). However, hypoxic burden during SWS was significantly reduced with oxygen compared to without oxygen [OSA_SWS + O₂ = 23 (%min)/h, OSA_SWS = 37 (%min)/h]. No significant OSA was observed in non-rapid eye movement (REM) stage 1 (NREM 1), non-REM stage 2 (NREM 2), or REM sleep (e.g., non-SWS) in any condition. The SWS-specific CPAP withdrawal induces OSA with SF and IH. The addition of oxygen during CPAP withdrawal results in SF with significantly less severe hypoxemia during the induced respiratory events in SWS. This model of SWS-specific CPAP withdrawal disrupts SWS with a physiologically relevant stimulus and facilitates the differentiation of SF and IH in OSA.