HEAR is a simple, yet efficient algorithm to remove transient, high-variance artifacts from multivariate time-series signals (e.g., electroencephalographic (EEG) signals).
Kobler, R. J., Sburlea, A. I., Mondini, V. & Müller-Putz, G. R. HEAR to remove pops and drifts: the high-variance electrode artifact removal (HEAR) algorithm. In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, 2019 accepted version
HEAR can be applied offline and online. The repository contains a reference implementation in Matlab and a dataset of simulated EEG signals. The demonstration dataset is stored in the eeglab format.
- Download HEAR and open the downloaded folder.
- Startup the eeglab toolbox:
- Open the
train_HEAR.m
script. The script loads a calibration dataset (demo_simrest.set
) that contains simulated artifact-free EEG signals. Then HEAR is fit to the data. The parameteris_causal
defines if HEAR should be used onlineis_causal = true
or offlineis_causal = false
. The calibrated model is stored to the disk 'hear_mdl.mat'. - The script
apply_HEAR.m
uses the calibrated model to correct pop and drift artifacts in a second dataset (demo_simreach.set
).
Feel free to contact me at reinmar.kobler@tugraz.at.
This work was supported by the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (Consolidator Grant 681231 'Feel Your Reach').