A Spectral Clustering Approach for Modeling Connectivity Patterns in Electroencephalogram Sensor Networks
Electroencephalography (EEG) is a non-invasive low cost monitoring exam that is used for the study of the brain in every hospital and research labs. Time series recorded from EEG sensors can be studied from the perspective of computational neuroscience and network theory to extract meaningful features of the brain. In this chapter we present a network clustering approach for studying synchronization phenomena as captured by cross-correlation in EEG recordings. We demonstrate the proposed clustering idea in simulated data and in EEG recordings from patients with epilepsy.
KeywordsSensor Network Spectral Cluster Absence Epilepsy Binary Constraint Synchronization Measure
- 9.A.Y. Ng, M.I. Jordan, and Y. Weiss. On spectral clustering: Analysis and an algorithm. In Advances in Neural Information Processing Systems 14, pages 849–856. MIT Press, Boston, 2001.Google Scholar
- 11.S.E. Karisch and F. Rendl. Semidefinite programming and graph equipartition. In In Topics in Semidefinite and Interior-Point Methods, pages 77–95. AMS, Providence Rhode island, 1998.Google Scholar
- 19.J. Dauwels, F. Vialatte, and A. Cichocki. Neural Information Processinge, volume 4984 of Lecture Notes in Computer Science, chapter A Comparative Study of Synchrony Measures for the Early Detection of Alzheimers Disease Based on EEG. Springer, Berlin / Heidelberg, 2008.Google Scholar
- 20.V. Sakkalis, C.D. Giurcaneanu, P. Xanthopoulos, M. Zervakis, V. Tsiaras, Y. Yang, E. Karakonstantaki, and S. Michelyannis. Assessment of linear and nonlinear synchroniztion measures for analyzing eeg in a mild epileptic paradigm. IEEE transaction on Information technology in biomedicine, 2008. DOI:10.1109/TITB.2008.923141. 13(4):433-441, Jul 2009.Google Scholar