Automatic classification of multivariate EEGs using an amount of information measure and the eigenvalues of parametric time series model features

Gersch, Will; Yonemoto, James; Naitoh, Paul

doi:10.1016/0010-4809(77)90044-1

Cited by 39 publications

(8 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Thus, the coupling is least when the chains reside in state 2/2, and most in state 0/0. This corresponds with the literature [16] showing that coupling is minimal during periods of wakefulness and maximal during deep sleep. As the two chains correspond to signal information from different sides of the brain this indicates a greater interhemispheric coupling during deep sleep.…”

Section: Sleep Eeg Datasupporting

confidence: 92%

Learning interaction dynamics with coupled hidden Markov models

Rezek

Sykacek

Roberts

2000

IEE Proceedings - Science, Measurement and Technology

View full text Add to dashboard Cite

An analysis of interactions between different physiological control systems may only possible with correlation functions if the signals have similar spectral distributions. Interactions between such signals can be modelled in state space rather than observation space, i.e we model interactions after first translating the observations into a common domain. Coupled Hidden Markov Models (CHMM) are such state-space models. They form a natural extension to standard Hidden Markov Models. In this paper we perform CHMM parameter estimation under a Bayesian paradigm, using Gibbs sampling, and in a Maximum Likelihood framework, using the expectation maximisation algorithm. The performance differences between the estimators are demonstrated on simulated data as well as biomedical data. We show that the proposed method gives meaningful results when comparing two different signals, such as respiration and EEG.

show abstract

Section: Sleep Eeg Datasupporting

confidence: 92%

Learning interaction dynamics with coupled hidden Markov models

Rezek

Sykacek

Roberts

2000

IEE Proceedings - Science, Measurement and Technology

View full text Add to dashboard Cite

show abstract

“…• • Extraction using parametric time series models: This technique is customary in the construction of some brain computer interfaces. [16][17][18][19][20][21] In our study, the EEG signals were extracted by fitting to them order-3 autoregressive models (AR(3)). 22 The resulting set of model coefficients (without the variance of the model driving noise) was subsequently used in the classification task.…”

Section: • • Fourier Analysis Based Technique: For Comparison Pur-mentioning

confidence: 99%

Protocol Design Challenges in the Detection of Awareness in Aware Subjects Using EEG Signals

et al. 2016

View full text Add to dashboard Cite

Recent studies have evidenced serious difficulties in detecting covert awareness with electroencephalography-based techniques both in unresponsive patients and in healthy control subjects. This work reproduces the protocol design in two recent mental imagery studies with a larger group comprising 20 healthy volunteers. The main goal is assessing if modifications in the signal extraction techniques, training-testing/cross-validation routines, and hypotheses evoked in the statistical analysis, can provide solutions to the serious difficulties documented in the literature. The lack of robustness in the results advises for further search of alternative protocols more suitable for machine learning classification and of better performing signal treatment techniques. Specific recommendations are made using the findings in this work.

show abstract

“…This feature can make the connection between the stimulus and the response, or between the responses observed at different scalp locations, invisible to the methods we have described-even if the two frequencies fall within the same frequency band at the base resolution of the analysis-if the record is sufficiently long (100). Various techniques have been proposed which may be capable of addressing such issues, including nonlinear cross-spectral analysis (178), approaches based on the theory of mutual information (130,179,180), and higherorder spectral analysis (181,182), all of which have been applied in at least exploratory fashion to EEG data. The theory of nonlinear dynamic systems and nonstationary random processes is also currently undergoing vigorous development in various areas of mathematics and engineering.…”

Section: Problems Of Nonstationarity and Nonlinearitymentioning

confidence: 99%