Home | Publications and software | Personal


Frølich, Laura and Andersen, Tobias S. and Mørup, Morten (2015). Classification of Independent Components of EEG Into Multiple Artifact Classes. Psychophysiology. DOI: 10.1111/psyp.12290. (link to paper on journal web site)

Nielsen, Bo Friis and Frølich, Laura and Nielsen, Otto Anker and Filges, Dorte. (2013). Estimating passenger numbers in trains using existing weighing capabilities. Transportmetrica A: Transport Science. DOI: 10.1080/23249935.2013.795199 (link to paper on journal web site)

Eskelund, Kasper and Frølich, Laura, and Andersen, Tobias S. (2013). Facial configuration and audiovisual integration of speech: a mismatch negativity study. In Proceedings of ISAAR2013.

Master's thesis (2011) (link to PDF)

Conferences with contributions

Laura Frølich, Irene Winkler, Klaus-Robert Müller, and Wojciech Samek (2015) Investigating effects of different artefact types on Motor Imagery BCI. Proceedings of 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2015. To appear.

Laura Frølich, Tobias S. Andersen, and Morten Mørup (2013) Classification of Independent Components of EEG Into Multiple Artifact Classes, Abstract at the International Conference on Basic and Clinical Multimodal Imaging (BaCI), p. 50 (link to abstract)

Multi-class classification of independent components of EEG "Breaking news" poster at AISTATS, 2012

Parameters in ensemble predictions Oral presentation at ANEMOS.plus: 4th Contractors Meeting, Crete, 2009

Matlab Code: IC_MARC (classification of Independent Components of EEG into Multiple ARtifact Classes)

Works as plugin for EEGLab if placed in the folder "plugins" within the EEGLab directories. A Quick Start Guide is included in the folder and also available here.
Version 1.6, corrects the list of features used in the established feature set.
Version 1.5, fixes some remaning graphics issues in pop_ICMARC_prop().
Version 1.4, fixes error from DIPFIT that occurs in a data consistency check by FieldTrip if the EEG.icaact field is empty when the EEG struct is converted to the FieldTrip format.
Version 1.3, fixes graphics issues so that IC_MARC now also runs under Matlab versions 2014b and above.
Version 1.2
Version 1.2 causes problems with a FieldTrip function that IC_MARC relies on through DIPFIT. The error occurs in a data consistency check made by FieldTrip. The latest version of EEGLab that this was not a problem in was Version of EEGLab runs under Matlab versions prior to 13b.
To just plot and manually classify and reject components using the GUI, choose the "Plot scalpmaps with classes" menu item in the IC_MARC menu within the "Plot" menu in EEGLab. This allows you to manually choose the classifications of ICs and to determine whether or not to mark ICs for rejection. The classifications are saved in EEG.reject.classtype and the rejection status is set in EEG.reject.gcompreject.

Matlab Code: Tensor classification

Tensor classification, code to run CMDA, DATER, BDA as well as functions that calculate gradients for BDCA and the methods described in a forthcoming paper. The optimisation toolboxes ManOpt and immoptibox were used to optimise the objective functions calculated by these functions, as described in the paper.