TechRxiv
2020_eeg_imu_journal_pre_print.pdf (2.13 MB)

Motion artefact removal in electroencephalography and electrocardiography by using multichannel inertial measurement units and adaptive filtering

Download (2.13 MB)
preprint
posted on 25.01.2021, 14:30 by Christopher Beach, Mingjie Li, Ertan Balaban, Alex Casson
This paper presents a new active electrode design for electroencephalogram (EEG) and electrocardiogram (ECG) sensors based on Inertial Measurement Units (IMUs) to remove motion artefacts during signal acquisition. Rather than measuring motion data from a single source for the entire recording unit, IMUs are attached to each individual EEG or ECG electrode to collect more local movement data. This movement data is then used to remove the motion artefact by using Normalised Least Mean Square (NLMS) adaptive filtering. Results show that the proposed active electrode design can reduce motion contamination from EEG and ECG signals in chest movement and head swinging motion scenarios. However the performance depends on the quality of the input signal with the algorithm providing better performance on signals with lower signal-to-noise ratios. The new instrumentation hardware allows data driven artefact removal to be performed, providing a new approach compared to widely used, non-parametric, blind-source separation methods, and helps enable \emph{in the wild} EEG recordings to be performed.

Funding

Passively Powered Non-invasive Human Body Sensing on Bio-Degradable Conformal Substrates

Engineering and Physical Sciences Research Council

Find out more...

MultiSense - Devising and Manufacturing mm-Wave High Data Rate Low Latency On-Skin Technologies

Engineering and Physical Sciences Research Council

Find out more...

History

Email Address of Submitting Author

christopher.beach@manchester.ac.uk

ORCID of Submitting Author

0000-0003-4964-3173

Submitting Author's Institution

The University of Manchester

Submitting Author's Country

United Kingdom

Licence

Exports