Abstract
Reliable automation of the labor-intensive manual task of scoring animal sleep can facilitate the analysis of long-term sleep studies. In recent years, deep-learning-based systems, which learn optimal features from the data, increased scoring accuracies for the classical sleep stages of Wake, REM, and Non-REM. Meanwhile, it has been recognized that the statistics of transitional stages such as pre-REM, found between Non-REM and REM, may hold additional insight into the physiology of sleep and are now under vivid investigation. We propose a classification system based on a simple neural network architecture that scores the classical stages as well as pre-REM sleep in mice. When restricted to the classical stages, the optimized network showed state-of-the-art classification performance with an out-of-sample F1 score of 0.95 in male C57BL/6J mice. When unrestricted, the network showed lower F1 scores on pre-REM (0.5) compared to the classical stages. The result is comparable to previous attempts to score transitional stages in other species such as transition sleep in rats or N1 sleep in humans. Nevertheless, we observed that the sequence of predictions including pre-REM typically transitioned from Non-REM to REM reflecting sleep dynamics observed by human scorers. Our findings provide further evidence for the difficulty of scoring transitional sleep stages, likely because such stages of sleep are under-represented in typical data sets or show large inter-scorer variability. We further provide our source code and an online platform to run predictions with our trained network.
Highlights
Sleep is one of the most fundamental and not yet well understood processes that can be observed in humans and most animals
The second study[36] reported a convolutional neural network to show better classification performance compared to a random forest trained on engineered features but was not based on human-annotated data
The third s tudy[24] trained a convolutional neural network to predict sleep stages in narcoleptic and wild-type mice where cataplectic events were derived based on rules operating on the predictions (Wake, REM, or NREM) of the network
Summary
Sleep is one of the most fundamental and not yet well understood processes that can be observed in humans and most animals. Systems that can create and use learned features have repeatedly been demonstrated to achieve superior classification performance in related fields such as speech or image recognition[29,30] These successes indicate that learned features may be effective and expressive representations of data with respect to the classification challenges at hand, as well as various EEG-based challenges including brain-computer interfaces, epileptic seizure detection, or sleep s coring[31]. Deep neural networks introduced in two studies[32,33] were inspired by image recognition systems and used preprocessed spectrograms of EEG and the electromyogram (EMG) as input “images” These systems yielded state-of-the-art classification performances but needed an additional hidden Markov m odel[32] that constrained the output of the system on physiologically plausible sleep state transitions or used mixture z-scoring[33] which needs an initial sample of human-annotated data for each mouse. All discussed neural networks were trained to distinguish between the main sleep stages Wake, REM, and NREM only
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
