Unsupervised Deep Feature Extraction for Neonatal Sleep Stage Classi cation

Abstract

Preterm birth carries many risks for an infant. These infants are accomodated in a specialized unit of the hospital: the Neonatal Intensive Care Unit, NICU. The NICU utilizes many tools to provide care and assess maturation of these infants. Sleep stage monitoring is such an important tool. EEG provides a non-invasive way of performing the monitoring task. However, correctly identifying sleep stages from EEG recordings is a challenging and time-consuming task even for experts.

-

N. Seeuws et al. et al. [ 3 ] The unlabeled dataset consists of seven additional recordings providing an additional 109 hours of EEG recordings. This data was recorded under the same modalities as the labeled dataset. The signal is ltered and downsampled to 30Hz from 250Hz for use in this work. Altough the full dataset provides a multichannel signal, this work focused on single channel bipolar EEG data to simplify implementation and experimentation. It used the di erence between the C3 and C4 electrode based on the international 10-20 system[ 2 ] as a signal.

The unsupervised approach used a Variational Auto-Encoder, VAE, as introduced by Kingma et al.[ 5 ] to extract features from the EEG segments. After training on the combined dataset of EEG recordings, the VAE de nes a posterior probability distribution in latent space given an EEG segment and the mean value of this posterior was used as the extracted feature vector of a segment. A gradient boosting classi er performed the nal classi cation using the extracted features of the labeled segments.

The semi-supervised approach made use of a Generative Adversarial Network, GAN, as introduced by Goodfellow et al.[ 4 ] The discriminator of the GAN was extended with an additional output acting as the desired classi er as proposed by Salimans et al.[ 7 ] This model can make direct use of labeled and unlabeled information and does not need an additional classi er to detect sleep stages.

The results are benchmarked against two other approaches. The rst makes use of a set of features deemed relevant for sleep stage identi cation as proposed by Piryatinska et al.[ 6 ] mainly focused on spectral information of the EEG segments. The features are classi ed by a gradient boosting model. The second approach makes use of a convolutional neural network in a traditional supervised setting proposed by Ansari et al.[ 1 ] Several performance metrics were computed but models were mainly compared using Cohen's kappa coe cient. The VAE based model scored 0:47 on the test set while the GAN based model scored 0:64. Comparing these kappa values to the 0:43 scored by the classi er making use of spectral features[ 6 ] and the 0:60 reported for the supervised CNN by Ansari et al.[ 1 ] one can see the performance of a sleep stage classi er can be improved by making use of unlabeled data.

The VAE based model succeeded at extracting slightly better features compared to classical spectral features based on the improved classi cation performance but failed to improve upon the performance of an end-to-end deep learning approach. The GAN based model did show improved performance to supervised deep learning when making use of unlabeled information.

A further investigation of the feature space for both the VAE based model and the GAN based model leads to the conclusion that an unsupervised model struggles to separate factors of variation corresponding to sleep stage where a semi-supervised approach does identify such relevant factors.

The results show a possibility to improve sleep stage classi cation performance by leveraging unlabeled recordings. Increased performance is currently only observed for semi-supervised models.

References

1. Ansari , A. , De Wel , O. , Pillay , K. , Dereymaeker , A. , Jansen , K. , Van Hu el, S., Naulaers , G. , De Vos , M.: A convolutional neural network outperforming state-ofthe-art sleep staging algorithms for both preterm and term infants . Internal Report 19-65 , ESAT-STADIUS, KU Leuven (Leuven,Belgium) (2019 (submitted for publication))

2. Cherian , P.J. , Swarte , R.M. , Visser , G.H. : Technical standards for recording and interpretation of neonatal electroencephalogram in clinical practice . Annals of Indian Academy of Neurology 12 ( 1 ), 58 ( 2009 )

3. Dereymaeker , A. , Pillay , K. , Vervisch , J., Van Hu el, S., Naulaers , G. , Jansen , K. , De Vos , M.: An automated quiet sleep detection approach in preterm infants as a gateway to assess brain maturation . International journal of neural systems 27(06) , 1750023 ( 2017 )

4. Goodfellow , I. , Pouget-Abadie , J. , Mirza , M. , Xu , B. , Warde-Farley , D. , Ozair , S. , Courville , A. , Bengio , Y. : Generative adversarial nets . In: Advances in neural information processing systems . pp. 2672 { 2680 ( 2014 )

5. Kingma , D.P. , Welling , M. : Auto-encoding variational bayes . arXiv preprint arXiv:1312.6114 ( 2013 )

6. Piryatinska , A. , Terdik , G. , Woyczynski , W.A. , Loparo , K.A. , Scher , M.S. , Zlotnik , A. : Automated detection of neonate eeg sleep stages . Computer methods and programs in biomedicine 95(1) , 31 { 46 ( 2009 )

7. Salimans , T. , Goodfellow , I. , Zaremba , W. , Cheung , V. , Radford , A. , Chen , X. : Improved techniques for training gans . In: Advances in neural information processing systems . pp. 2234 { 2242 ( 2016 )