Malin Björnsdotter
Department of Physiology, Institute of Neuroscience and Physiology, University of Gothenburg, Sweden
Simon Beckmann
Department of Physiology, Institute of Neuroscience and Physiology, University of Gothenburg, Sweden
Erik Ziegler
Department of Physiology, Institute of Neuroscience and Physiology, University of Gothenburg, Sweden
Johan Wessberg
Department of Physiology, Institute of Neuroscience and Physiology, University of Gothenburg, Sweden
Download article
Published in: The Swedish AI Society Workshop May 20-21; 2010; Uppsala University
Linköping Electronic Conference Proceedings 48:4, p. 9-14
Published: 2010-05-19
ISBN:
ISSN: 1650-3686 (print), 1650-3740 (online)
The appealing properties of artificial intelligence (AI) methods are being increasingly acknowledged by the neuroimaging community; as evidenced by the recent surge of brain activity pattern recognition studies [19]. Supervised learning and classification; in particular; are appreciated tools for localizing and distinguishing intricate brain response patterns and making predictions about otherwise undetectable neural states. Our group refines and applies such methods in order to implement sensitive and dynamic tools for characterization of neurophysiological data. Specifically; we employ support vector machines (SVMs); particle swarm optimization (PSO); independent component analysis (ICA); as well as both genetic and memetic algorithms on functional magnetic resonance imaging (fMRI) and electroencephalography (EEG) data. This paper provides a brief overview of our recent advances in the development and utilization of Aibased analysis; with the particular aspiration to characterize human brain activation patterns produced
by touch.
