A novel K-means based multivariate clustering of IC-fingerprints
-
1
G. D`Annunzio University Foundation, Institute for Advanced Biomedical Technologies, Italy
-
2
University of Chieti , ITAB, Italy
-
3
University of Potsdam, Center for Dynamics of Complex Systems, Germany
Independent component analysis (ICA) has been widely applied on functional neuroimaging data (e.g. functional magnetic resonance imaging, fMRI and magnetoencephalography, MEG) due to its data-driven nature. Since in many applications the single ICs are assigned a functional role, it is very important to develop a reliable clustering approach to assess the IC consistency across sessions and subjects to capture the spatio-temporal dynamics of active brain areas in group studies. To this aim, we developed a novel algorithm: the Multivariate Algorithm for Grouping Independent Components Kmeans based (MAGICK). It can be applied to both fMRI and MEG data. The main advantages of this approach are that the ICA independence assumption is not violated and the number of clusters is automatically estimated from an initial guess. The algorithm, based on a modified version of the fingerprints [1] exploits a set of parameters related to temporal, spatial and spectral properties of the ICs. To increase the convergence and the accuracy of MAGICK, each parameter is assigned a weight that is iteratively updated. Results on simulated and real data are presented. The real data consist of multisubject MEG recordings acquired during the median nerve stimulation and an fMRI dataset acquired during a visuomotor task. The performance of MAGICK is compared to the self-organizing group ICA (sogICA) method [2]. Our approach outperforms sogICA leading to a more accurate classification on these data.
References
1. De Martino F, Gentile F, Esposito F, Balsi M, Di Salle F, Goebel R, Formisano E. (2007) Classification of fMRI independent components using IC-fingerprints and support vector machine classifiers. NeuroImage 34:177-94.
2. Esposito F, Scarabino T, Hyvarinen A, Himberg J, Formisano E, Comani S, Tedeschi G, Goebel R, Seifritz E, Di Salle F. (2005) Independent component analysis of fMRI group studies by self-organizing clustering. NeuroImage 25: 193-205.
Conference:
Biomag 2010 - 17th International Conference on Biomagnetism , Dubrovnik, Croatia, 28 Mar - 1 Apr, 2010.
Presentation Type:
Poster Presentation
Topic:
Signal proccessing
Citation:
Spadone
S,
Pasquale
FD,
Penna
SD,
Mantini
D,
Pizzella
V and
Romani
GL
(2010). A novel K-means based multivariate clustering of IC-fingerprints.
Front. Neurosci.
Conference Abstract:
Biomag 2010 - 17th International Conference on Biomagnetism .
doi: 10.3389/conf.fnins.2010.06.00110
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
23 Mar 2010;
Published Online:
23 Mar 2010.
*
Correspondence:
Sara Spadone, G. D`Annunzio University Foundation, Institute for Advanced Biomedical Technologies, Potsdam, Italy, s.spadone@unich.it