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

Stanislas Chambon (1, 2, 3) , Valentin Thorey (3) , Pierrick J Arnal (3) , Emmanuel Mignot (2) , Alexandre Gramfort (4)

1 LTCI - Laboratoire Traitement et Communication de l'Information
2 Division of Sleep Medicine [Stanford]
3 Dreem SAS
4 PARIETAL - Modelling brain structure, function and variability based on high-field MRI data

Electroencephalography (EEG) during sleep is used by clinicians to evaluate various neurological disorders. In sleep medicine, it is relevant to detect macro-events (≥ 10 s) such as sleep stages, and micro-events (≤ 2 s) 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.

Mots clés: Deep learning; EEG; Event detection; Sleep; EEG-patterns; Time series

Domaines: Intelligence artificielle [cs.AI]; Réseau de neurones [cs.NE]; Traitement du signal et de l'image [eess.SP]

Stanislas Chambon, Valentin Thorey, Pierrick J Arnal, Emmanuel Mignot, Alexandre Gramfort. A deep learning architecture to detect events in EEG signals during sleep. MLSP 2018 - IEEE International Workshop on Machine Learning for Signal Processing, Sep 2018, Aalborg, Denmark. ⟨hal-01917529⟩