Microstate analysis of the EEG using local global graphs

Michalopoulos, Kostas; Bourbakis, Nikolaos G.

doi:10.1109/bibe.2013.6701583

Cited by 5 publications

(4 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The signal to noise ratio on the single trial data is also very low, constituting the single trial analysis a difficult task in general. In our case, we are using the technique described in [12]. We are taking under consideration the spatial relationship between the electrodes and we are working in the topographic image in order to find similarities between the maps.…”

Section: A Modeling Of the Eeg Topographymentioning

confidence: 99%

“…We are using the marked watershed segmentation algorithm [13] in order to identify the dominant peaks of the topography and reduce the effect of noise in the measurements. The Local Global graph (LG graph) structure [14] is used as descriptor of the topography and the corresponding measures of similarity between LG graphs are used as described in [12].…”

Section: A Modeling Of the Eeg Topographymentioning

confidence: 99%

See 1 more Smart Citation

Using Dynamic Bayesian Networks for modeling EEG topographic sequences

Michalopoulos¹,

Bourbakis²

2014

2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society

Self Cite

View full text Add to dashboard Cite

In this work we present a methodology for modeling the trajectory of EEG topography over time, using Dynamic Bayesian Networks (DBNs). Based on the microstate model we are using DBNs to model the evolution of the EEG topography. Analysis of the microstate model is being usually limited in the wide band signal or an isolated band. We are using Coupled Hidden Markov Models (CHMM) and a two level influence model in order to model the temporal evolution and the coupling of the topography states in three bands, delta, theta and alpha. We are applying this methodology for the classification of target and non-target single trial from a visual detection task. The results indicate that taking under consideration the interaction among the different bands improves the classification of single trials.

show abstract

Section: A Modeling Of the Eeg Topographymentioning

confidence: 99%

Section: A Modeling Of the Eeg Topographymentioning

confidence: 99%

Using Dynamic Bayesian Networks for modeling EEG topographic sequences

Michalopoulos¹,

Bourbakis²

2014

2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society

Self Cite

View full text Add to dashboard Cite

show abstract

“…The main approach in this problem is to first define a measure of similarity between topographies and then apply a clustering algorithm in order to represent the set of topographies using the cluster centroids. In [18], we introduced a new measure of similarity based on the Local Global Graph (LG graph) which was applied for the segmentation of the average ERP. This measure treats the topographic map as an image and uses segmentation in order to extract the LG graph.…”

Section: A Eeg As Sequence Of Field Topography Mapsmentioning

confidence: 99%

“…A modified k-means algorithm has been applied in [17]. Also hierarchical clustering algorithms and soft clustering algorithms as Gaussian mixture models [15], [18] have been successfully applied for the identification of dominant topographies. Some implementations incorporate temporal filtering of the results in order to remove isolated topographies in time and create a smooth temporal segmentation.…”

Section: ) Modelling Using Hidden Markov Modelsmentioning

confidence: 99%

Fusion of EEG Topograhic Features and fMRI Using Canonical Partial Least Squares

Michalopoulos

Bourbakis

2014

2014 IEEE International Conference on Bioinformatics and Bioengineering

Self Cite

View full text Add to dashboard Cite

In this paper we present a novel method for describing the EEG as a sequence of topographies, based on the notion of microstates. We use Hidden Markov Models (HMM) to model the temporal evolution of the topography of the average Event Related Potential (ERP) and we calculate the Fisher score of the sequence by taking the gradient of the trained model parameters given the sequence. In this context, the average Event Related Potential (ERP) is described as a sequence of topographies and the Fisher score describes how this sequence deviates from the learned HMM. This alternative modeling of the ERP is used to fuse EEG information, as expressed by the temporal evolution of the topography, and Functional Magnetic Resonance Imaging (fMRI). We use Canonical Partial Least Squares (CPLS) for the fusion of the Fisher score with fMRI features. In order to test the effectiveness of this method, we compare the results of this methodology with the results of CPLS using the average ERP signal of a single channel. Using this methodology we are able to derive components that co-vary between EEG and fMRI and present significant differences between the two tasks. The results indicate that this descriptor effectively characterizes the temporal evolution of the ERP topography and can be used for fusing EEG and fMRI for the discrimination of the brain activity on different tasks.

show abstract