Fully Convolutional Signal Transcription for Artifact Removal and Normalization

Abstract

Electrophysiological recordings are usually prone to artifacts that are commonly removed by visual inspection and signal decomposition techniques such as ICA in clinical and research contexts. In this work, we propose a generative deep learning model for EEG signals inspired by the WaveNet architecture conditioned on biosignals from other modalities (EOG, EMG and ECG) that are expected to capture relevant context information on artifacts that are simultaneously recorded in the EEG signal. As our main contributions, we show that it is possible to capture non-trivial correlations between time-series from multiple modalities in a fully-convolutional neural network. By dampening the context signals during inference, it is possible to remove the caused artifacts in the EEG signal by (1) generating a possibly uncorrelated EEG sequence from a dampened version of the context signal (2) estimating the context signal’s contribution to a particular kind of artifact and removing this effect from the original EEG sequence, preserving the signal information. In this poster, we present preliminary results on the openly available MASS cohort-3 sleep database, comprising recordings from various patients recorded during sleep.

Publication
Poster Presentation at the 13th Bernstein Conference 2017, Göttingen, Germany
Date
Links

A. Proposed WaveNet-inspired network architecture. Using recordings from ECG, EOG and EMG recordings as context signals, the network generates an EEG sequence matching the context signals. B. Demonstration of artifact removal C. Network learning curve Figure 1: A. Proposed WaveNet-inspired network architecture. Using recordings from ECG, EOG and EMG recordings as context signals, the network generates an EEG sequence matching the context signals. B. Demonstration of artifact removal C. Network learning curve

References

[1] Oord, Aaron van den, et al. "Wavenet: A generative model for raw audio." arXiv preprint arXiv:1609.03499 (2016).
[2] O'reilly, Christian, et al. "Montreal Archive of Sleep Studies: an open‐access resource for instrument benchmarking and exploratory research." Journal of sleep research 23.6 (2014): 628-635., 10.1111/jsr.12169
[3] Gal, Yarin. Uncertainty in deep learning. Diss. PhD thesis, University of Cambridge, 2016.
[4] Massetani, Roberto, et al. "Alteration of cardiac function in patients with temporal lobe epilepsy: different roles of EEG‐ECG monitoring and spectral analysis of RR variability." Epilepsia 38.3 (1997): 363-369., 10.1111/j.1528-1157.1997.tb01129.x