Electronic Sleep Stage Classifiers: A Survey and VLSI Design Methodology

Abstract

First, existing sleep stage classifier sensors and algorithms are reviewed and compared in terms of classification accuracy, level of automation, implementation complexity, invasiveness, and targeted application. Next, the implementation of a miniature microsystem for low-latency automatic sleep stage classification in rodents is presented. The classification algorithm uses one EMG (electromyogram) and two EEG (electroencephalogram) signals as inputs in order to detect REM (rapid eye movement) sleep, and is optimized for low complexity and low power consumption. It is implemented in an on-board low-power FPGA connected to a multi-channel neural recording IC, to achieve low-latency (order of 1 ms or less) classification. Off-line experimental results using pre-recorded signals from nine mice show REM detection sensitivity and specificity of 81.69% and 93.86%, respectively, with the maximum latency of 39 [Formula: see text]. The device is designed to be used in a non-disruptive closed-loop REM sleep suppression microsystem, for future studies of the effects of REM sleep deprivation on memory consolidation.

A real‐time sleep scoring framework for closed‐loop sleep manipulation in mice

“…Although various studies describe methods for real‐time sleep classification, most rely on predetermined thresholds for detection (Fenzl et al., 2007; Kassiri et al., 2017; Louis et al., 2004) or train classifiers using human‐scored baseline data (Van Gelder et al., 1991), which introduce bias and increase the experimenter’s workload. Although methods which utilise feature normalisation to avoid training bias seem to be considered suitable for real‐time implementation (Bastianini et al., 2014; Stephenson et al., 2009), their actual online performance is yet to be reported.…”

Review on Biomedical Sensors, Technologies and Algorithms for Diagnosis of Sleep Disordered Breathing: Comprehensive Survey

“…Reference [8] reviewed the sensors and algorithms which were used for sleep stage classification in terms of accuracy, level of automation, implementation complexity, invasiveness and applications. This paper also includes an algorithm implementation.…”

Defining and measuring paradoxical (REM) sleep in animal models of sleep disorders

“…Standard sleep scoring methods in rodents take in consideration high activity in nuchal EMG or animal movements from video recordings to differentiate active W from sleep [2,3].…”

“…However, differentiation between NREM sleep and REM sleep necessitates measures of selected frequency bands amplitude from the parietal (sometimes with frontal) EEG -mainly delta (d: 0-4 Hz) and theta (q: 4-8Hz). Most of algorithms developed so far to score vigilance states automatically are based on such analysis [2][3][4].…”