MC-SleepNet: Large-scale Sleep Stage Scoring in Mice by Deep Neural Networks

Yamabe, Masato; Horie, Kazumasa; Shiokawa, Hiroaki; Funato, Hiromasa; Yanagisawa, Masashi; Kitagawa, Hiroyuki

doi:10.1038/s41598-019-51269-8

Cited by 45 publications

(64 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, power spectral analysis revealed that during the possibly sleeping state, there was a peak in the delta band (1–4 Hz, Fig. 6 B), presumably corresponding to the stage of non-rapid eye movement sleep, the predominant form of sleep in mice 42 , 43 . The other state, conversely, had higher power in the gamma frequency band (30–100 Hz, Fig.…”

Section: Resultsmentioning

confidence: 99%

“…Further examination with time–frequency analysis demonstrated that one state displayed increased power mostly in a low frequency range (Fig. 6 A), potentially corresponding to sleeping periods of the animal, which are characterized by cortical EEG of higher amplitudes than awake stages 42 , 43 . In addition, power spectral analysis revealed that during the possibly sleeping state, there was a peak in the delta band (1–4 Hz, Fig.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

OSERR: an open-source standalone electrophysiology recording system for rodents

Cheng

Murari

2020

Sci Rep

View full text Add to dashboard Cite

Behavioral assessment of rodents is critical for investigation of brain function in health and disease. In vivo neurophysiological recordings are powerful tools to mechanistically dissect neural pathways that underlie behavioral changes, and serve as markers for dynamics, efficacy and safety of potential therapeutic approaches. However, most in vivo recording systems require tethers or telemetry receivers, limiting their compatibility with some behavioral tests. Here, we developed an open-source standalone electrophysiology recording system for rodents (OSERR). It is a tether-free, standalone recording device with two channels, a reference and a ground, that acquires, amplifies, filters and stores data all in itself. Thus, it does not require any cable or receiver. It is also compact and light-weight, and compatible with juvenile mice, as well as multiple recording modalities and standard electrode implantation methods. In addition, we provide the complete design of hardware, and software for operation. As an example, we demonstrated that this standalone system, when configured with a bandwidth of 1–120 Hz and gain of 1000, successfully collected EEG signals during induced seizure, extended recording, anesthesia, and social interactions in mice. The design of this system is practical, economical, and freely available. Thus, this system could enable recording of brain activity during diverse behavioral assays in a variety of arenas and settings, and allow simultaneous recordings from multiple subjects to examine social behaviors. Importantly, with the open-source documentation, researchers could customize the design of the system to their specific needs.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

OSERR: an open-source standalone electrophysiology recording system for rodents

Cheng

Murari

2020

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Although deep learning approaches have shown recent success in automated sleep scoring in both human patients 61 and animal models 62 (with large datasets and promising results), we opted for a shallow-learning approach without dimensionality reduction, allowing for greater interpretability of the results. The knowledge that the algorithm is using features with known biological relevance makes these results easier to approach and understand by the medical and research communities.…”

Section: Discussionmentioning

confidence: 99%

Sleep Identification Enabled by Supervised Training Algorithms (SIESTA): An open-source platform for automatic sleep staging of rodent polysomnographic data

Caldart

Sánchez

Ben-Hamo

et al. 2020

Preprint

View full text Add to dashboard Cite

AbstractThe temporal distribution of sleep stages is critical for the study of sleep function, regulation, and disorders in higher vertebrates. This temporal distribution is typically determined polysomnographically. In laboratory rodents, scoring of electrocorticography (ECoG) and electromyography (EMG) recordings is usually performed manually, where 5-10 second epochs are categorized as one of three specific stages: wakefulness, rapid-eye-movement (REM) sleep and non-REM (NREM) sleep. This process is laborious, time-consuming, and particularly impractical for large experimental cohorts with recordings lasting longer than 24 hours.To circumvent this problem, we developed an open-source Python toolkit, Sleep Identification Enabled by Supervised Training Algorithms (SIESTA), that automates the detection of these three main behavioral stages in mice. Our supervised machine learning algorithm extracts features from the ECoG and EMG signals, then automatically scores recordings with a hierarchical classifier based on Bagging Random Forest approaches. We evaluated this approach on data collected from wild-type mice housed under both normal and different lighting conditions, as well as from a mutant mouse line with abnormal sleep phenotypes. To validate its performance on test data, we compared SIESTA with manually scored data and obtained F1 scores of 0.92 for wakefulness, 0.81 for REM, and 0.93 for NREM.SIESTA has a user-friendly interface that can be used without coding expertise. To our knowledge, this is the first time that such a strategy has been developed using all open-source and freely available resources, and our aim is that SIESTA becomes a useful tool that facilitates further research of sleep in rodent models.

show abstract

“…Four studies proposed deep neural networks that were trained end-to-end on raw input EEG and EMG signals. One of the earliest studies 34 was based on 22 years of EEG and EMG signals from mice and reported an improvement of sleep scoring accuracy with respect to classical approaches. The system contained a bidirectional long short-term memory (LSTM) module in the classifier head to model long-range non-linear correlations 35 , and the training also included a retraining scheme.…”

Section: Introductionmentioning

confidence: 99%

Automated scoring of pre-REM sleep in mice with deep learning

Grieger

Schwabedal²,

Wendel

et al. 2021

Sci Rep

View full text Add to dashboard Cite

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.

show abstract

MC-SleepNet: Large-scale Sleep Stage Scoring in Mice by Deep Neural Networks

Cited by 45 publications

References 22 publications

OSERR: an open-source standalone electrophysiology recording system for rodents

OSERR: an open-source standalone electrophysiology recording system for rodents

Sleep Identification Enabled by Supervised Training Algorithms (SIESTA): An open-source platform for automatic sleep staging of rodent polysomnographic data

Automated scoring of pre-REM sleep in mice with deep learning

Contact Info

Product

Resources

About