EEG Signal Clustering With Learned Features Using Deep Autoencoder
DOI:
https://doi.org/10.54139/revinguc.v28i1.18Keywords:
Epileptic EEG signals, autoencoder, K-means, SVC.Abstract
This work proposes a convolutional autoencoder based non-supervised feature extractor, to find clusters of electroencephalographic signal (EEG) to supporting to the physicists to diagnose the epilepsy condition. Three autoencoders were designed with input dimensions of 4096×1, 2048×2 and 768×6, to analyze the impact of the signal length on latent representation generated by autoencoders. Latent representation was used as input to the clustering algorithms K-means and support vector clustering. Latent representation was mapped onto a two-dimensional space of mean and standard deviation to visualize it, and to apply the clustering algorithms. Results showed a good latent representation of the three autoencoders, with a maximum reconstruction error of 1,47 % for the worst case. Clustering algorithms got visually consistent clusters compared with the ground-truth distribution onto the two-dimension latent space. The best performance was achieved with the K-means algorithm and the best latent representation of the input signal. Resultant clusters were impacted by the length of the input segment, where K-means with an input length of 4096 samples had the best performance.
Downloads
References
A. Neligan and J. W. Sander, Epidemiology of Seizures and Epilepsy. John Wiley & Sons, Ltd, 2014, ch. 4, pp. 28-32. https://doi.org/10.1002/9781118456989.ch4
D. Shorvon, Simon, Handbook of epilepsy treatment. Forms, causes and therapy in children and adults, 2nd ed. Blackwell Publishing, 2005.
S. Miyamoto, H. Ichihashi, and K. Honda, Algorithms for Fuzzy Clustering: Methods in c-Means Clustering with Applications, 1st ed. Springer Publishing Company, 2008.
G. E. Hinton and R. R. Salakhutdinov, "Reducing the dimensionality of data with neural networks," Science, vol. 313, no. 5786, pp. 504-507, 2006. https://doi.org/10.1126/science.1127647
E. Min, X. Guo, Q. Liu, G. Zhang, J. Cui, and J. Long, "A survey of clustering with deep learning: From the perspective of network architecture," IEEE Access, vol. 6, pp. 39 501-39 514, August 2018. https://doi.org/10.1109/ACCESS.2018.2855437
J. B. MacQueen, "Some methods for classification and analysis of multivariate observations," in Proc. of the fifth Berkeley Symposium on Mathematical Statistics and Probability, L. M. L. Cam and J. Neyman, Eds., vol. 1. University of California Press, 1967, pp. 281-297.
A. Ben-Hur, D. Horn, H. T. Siegelmann, and V. Vapnik, "Support vector clustering," Journal of Machine Learning Research, vol. 2, pp. 125-137, march 2001.
S.-H. Lee and K. M. Daniels, "Gaussian kernel widths selection and fast cluster labeling for support vector clustering," University of Massachusetts Lowell, Tech. Rep., 2005.
B. Schölkopf, J. C. Platt, J. Shawe-Taylor, A. J. Smola, and R. C. Williamson, "Estimating the support of a high- dimensional distribution," Neural computation, vol. 13, no. 7, pp. 1443-1471, 2001. https://doi.org/10.1162/089976601750264965
B. Schölkopf, R. C. Williamson, A. J. Smola, J. Shawe-Taylor, J. C. Platt et al., "Support vector method for novelty detection." in NIPS, vol. 12. MIT Press, 1999, pp. 582-588.
V. Vapnik, The Nature of Statistical Learning Theory, 2nd ed. Springer, 2000, vol. 1. https://doi.org/10.1007/978-1-4757-3264-1_1
L. Rokach, "A survey of clustering algorithms," in Data Mining and Knowledge Discovery Handbook, 2nd ed., O. Maimon and L. Rokach, Eds. Springer, 2016, ch. 14, pp. 269-297. https://doi.org/10.1007/978-0-387-09823-4_14
H. W. Kuhn, "The hungarian method for the assignment problem," Naval research logistics quarterly, vol. 2, no. 1-2, pp. 83-97, 1955. https://doi.org/10.1002/nav.3800020109
P. A. Estévez, M. Tesmer, C. A. Perez, and J. M. Zurada, "Normalized mutual information feature selection," IEEE Transactions on Neural Networks, vol. 20, no. 2, pp. 189-201, February 2009. https://doi.org/10.1109/TNN.2008.2005601
R. Andrzejak, K. Lehnertz, F. Mormann, C. Rieke, P. David, and C. Elger, "Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity: Dependence on recording region and brain state," Physical Review E, vol. 64, no. 061907, pp. 061 907-1-061 907-8, 2001. https://doi.org/10.1103/PhysRevE.64.061907
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2023 Revista Ingeniería UC
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
