Abstract:
Electroencephalography is one of the most common electrophysiological methods. Regardless of its ubiquity, the problem of automatic EEG data processing and analysis is a ...Show MoreMetadata
Abstract:
Electroencephalography is one of the most common electrophysiological methods. Regardless of its ubiquity, the problem of automatic EEG data processing and analysis is a topic of an ongoing research. In this paper we discuss an algorithm for detection of frequency components in EEG signal with the use of Bayesian statistics. Proposed approach creates a model of a signal, which consists of only statistically relevant components. Model parameters are estimated in a way that avoids getting stuck in local minima. Operation of the algorithm is illustrated with examples of real and artificial signals.
Published in: 2016 21st International Conference on Methods and Models in Automation and Robotics (MMAR)
Date of Conference: 29 August 2016 - 01 September 2016
Date Added to IEEE Xplore: 26 September 2016
ISBN Information: