Spiking Neural Network for On-line Cognitive Activity Classification Based on EEG Data
The paper presents a method for the classification of EEG data recorded in two cognitive scenarios, a relaxing and memory task. The method uses a reservoir of spiking neurons that are activated by the spatio-temporal EEG data. The states of the reservoir are periodically read out and classified producing in a continuous classification result over time. After suitable optimization of the model parameters, we achieve a test accuracy of 82% on a small data set. Future applications of the proposed model are discussed including its use for an early detection of a cognitive impairment such as in Alzheimers disease.
KeywordsSpiking Neural Networks Liquid State Machines Reservoir Computing EEG data classification Cognitive tasks
Unable to display preview. Download preview PDF.
- 4.Jaeger, H.: The “echo state” approach to analysing and training recurrent neural networks. Tech. rep., Fraunhofer Institute for Autonomous Intelligent Syst. (2001)Google Scholar
- 6.Kindermans, P.J., Buteneers, P., Verstraeten, D., Schrauwen, B.: An uncued brain-computer interface using reservoir computing. In: Workshop: Machine Learning for Assistive Technologies, Proceedings, p. 8. Ghent University, Department of Electronics and information systems (2010)Google Scholar