A deep learning architecture to detect events in EEG signals during sleep

TL;DR

This paper introduces a deep learning model that simultaneously detects and classifies sleep-related EEG events, improving efficiency and consistency over traditional event-specific methods.

Contribution

A novel convolutional neural network architecture that jointly predicts locations, durations, and types of sleep EEG events from raw signals.

Findings

Outperforms current state-of-the-art event-specific algorithms

Efficient detection of macro- and micro-events in sleep EEG

Reduces need for manual annotation by experts

Abstract

Electroencephalography (EEG) during sleep is used by clinicians to evaluate various neurological disorders. In sleep medicine, it is relevant to detect macro-events (> 10s) such as sleep stages, and micro-events (<2s) such as spindles and K-complexes. Annotations of such events require a trained sleep expert, a time consuming and tedious process with a large inter-scorer variability. Automatic algorithms have been developed to detect various types of events but these are event-specific. We propose a deep learning method that jointly predicts locations, durations and types of events in EEG time series. It relies on a convolutional neural network that builds a feature representation from raw EEG signals. Numerical experiments demonstrate efficiency of this new approach on various event detection tasks compared to current state-of-the-art, event specific, algorithms.