Back
Full-text
Journal article

Automatic Human Sleep Stage Scoring Using Deep Neural Networks.

  • Malafeev A Chronobiology and Sleep Research, Institute of Pharmacology and Toxicology, University of Zurich, Zurich, Switzerland.
  • Laptev D Information Science and Engineering, Institute for Machine Learning, ETH Zurich, Zurich, Switzerland.
  • Bauer S Information Science and Engineering, Institute for Machine Learning, ETH Zurich, Zurich, Switzerland.
  • Omlin X Neuroscience Center Zurich, University of Zurich and ETH Zurich, Zurich, Switzerland.
  • Wierzbicka A Sleep Disorders Center, Department of Clinical Neurophysiology, Institute of Psychiatry and Neurology in Warsaw, Warsaw, Poland.
  • Wichniak A Third Department of Psychiatry and Sleep Disorders Center, Institute of Psychiatry and Neurology in Warsaw, Warsaw, Poland.
  • Jernajczyk W Sleep Disorders Center, Department of Clinical Neurophysiology, Institute of Psychiatry and Neurology in Warsaw, Warsaw, Poland.
  • Riener R Neuroscience Center Zurich, University of Zurich and ETH Zurich, Zurich, Switzerland.
  • Buhmann J Information Science and Engineering, Institute for Machine Learning, ETH Zurich, Zurich, Switzerland.
  • Achermann P Chronobiology and Sleep Research, Institute of Pharmacology and Toxicology, University of Zurich, Zurich, Switzerland.
Show more…
  • 2018-11-22
Published in:
  • Frontiers in neuroscience. - 2018
EEG
artificial neural networks
automatic scoring
deep learning
features
random forest
raw data
sleep
English The classification of sleep stages is the first and an important step in the quantitative analysis of polysomnographic recordings. Sleep stage scoring relies heavily on visual pattern recognition by a human expert and is time consuming and subjective. Thus, there is a need for automatic classification. In this work we developed machine learning algorithms for sleep classification: random forest (RF) classification based on features and artificial neural networks (ANNs) working both with features and raw data. We tested our methods in healthy subjects and in patients. Most algorithms yielded good results comparable to human interrater agreement. Our study revealed that deep neural networks (DNNs) working with raw data performed better than feature-based methods. We also demonstrated that taking the local temporal structure of sleep into account a priori is important. Our results demonstrate the utility of neural network architectures for the classification of sleep.
Language
  • English
Open access status
gold
Identifiers
Persistent URL
https://sonar.ch/global/documents/185758
Statistics
Document views: 46 File downloads:
  • Full-text: 0