Please use this identifier to cite or link to this item:
For citation please use:
Main Title: Towards classifier visualisation in 3D source space
Author(s): Krol, Laurens R.
Mousavi, Mahta
De Sa, Virginia
Zander, Thorsten O.
Type: Conference Object
Abstract: In the context of brain-computer interfacing, it is important to investigate what regions of the brain a classifier focuses on. For one, this will clarify to what extent the classifier relies on brain activity, as opposed to undesirable non-cortical signals. More generally, the practice is informative as it allows conclusions to be drawn about the cortical regions-and thus, cortical functions-that contribute to the effect under investigation. In this study, we start to investigate different methods to visualise the regions of interest of classifiers based on windowed means and on common spatial patterns. Specifically, we take individually reconstructed source spaces and transform the classifier filter weights into relevance weights indicating the relative contribution of each source to the classifier. This is visualised across participants in an average brain. By decomposing the classifier weights into separate sources and localising these in the brain, this method provides a tool to evaluate classifiers and test hypotheses.
Subject(s): scalp
brain modeling
feature extraction
Issue Date: 17-Jan-2019
Date Available: 6-Jan-2021
Is Part Of: 10.14279/depositonce-10656
Language Code: en
DDC Class: 153 Kognitive Prozesse, Intelligenz
004 Datenverarbeitung; Informatik
Proceedings Title: 2018 IEEE International Conference on Systems, Man, and Cybernetics (SMC)
Publisher: IEEE
Volume: 2018
Article Number: 18410764
Publisher DOI: 10.1109/SMC.2018.00022
EISSN: 2577-1655
ISBN: 978-1-5386-6650-0
ISSN: 1062-922X
TU Affiliation(s): Fak. 5 Verkehrs- und Maschinensysteme » Inst. Psychologie und Arbeitswissenschaft » FG Biopsychologie und Neuroergonomie
Appears in Collections:Technische Universit├Ąt Berlin » Publications

Files in This Item:
Format: Adobe PDF | Size: 1.96 MB
DownloadShow Preview

Item Export Bar

Items in DepositOnce are protected by copyright, with all rights reserved, unless otherwise indicated.