Electroencephalographic (EEG) signals present a myriad of challenges to analysis, beginning with the detection of artifacts. This simple algorithm is based on a statistical physics method, multiscale entropy (MSE) analysis, which quantifies the complexity of a signal. Noise corrupted EEG signals have lower information content, and, therefore, reduced complexity compared with their noise free counterparts. The method allows to identify epochs of corrupted noise on a given EEG signal. On an epoch-by-epoch base, the complexity index (CI) of the signal is computed as the sum of the MSE on the first 5 scales. Epochs having a CI lower than a pre-set threshold are scored as containing an artifact.
The code employs the function msentropy.m, from the WFDB Matlab toolbox.
Mandatory inputs to the function are:
Matlab R2015a and later versions. Please note that the WFDB Matlab toolbox must be installed and in your Matlab path.
The MSE-based artifact detector software package was developed at the Wyss Institute for Biologically Inspired Engineering at Harvard University and Beth Israel Deaconess Medical Center/Harvard Medical School by Sara Mariani, Filipa Borges and Teresa S. Henriques.
When referencing this software, please cite: Sara Mariani, Ana F. T. Borges, Teresa Henriques, Ary L. Goldberger, Madalena D. Costa: Use of multiscale entropy to facilitate artifact detection in electroencephalographic signals. EMBC 2015: 7869-7872
Please report bugs and questions at firstname.lastname@example.org.