|
Activity Number:
|
385
- SPEED: Statistical Methods and Applications in Medical Research, Risk Analysis, and Marketing Part 1
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 10, 2022 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Imaging
|
|
Abstract #322958
|
|
|
Title:
|
Club Exco: Clustering Brain Extreme Communities from Multi-Channel EEG Data
|
|
Author(s):
|
Matheus Bartolo Guerrero* and Raphael Huser and Hernando Ombao
|
|
Companies:
|
King Abdullah University of Science and Technology and King Abdullah University of Science and Technology (KAUST) and King Abdullah University of Science and Technology
|
|
Keywords:
|
Multivariate time series;
extreme value theory;
clustering;
nonstationarity;
non-stationary processes;
spectral measure
|
|
Abstract:
|
Epilepsy is a chronic neurological disorder affecting more than 50 million people globally. An epileptic seizure acts like a temporary shock to the neuronal system, disrupting regular electrical activity in the brain. Epilepsy is frequently diagnosed with electroencephalograms (EEGs). Current cluster methods for characterizing brain connectivity (e.g., coherence and partial coherence) are most influenced by the bulk of the EEG distribution rather than the tails. We develop the Club Exco method, which uses a spherical k-means procedure applied to the pseudo-angles derived from extreme amplitudes of EEG signals during an epileptic seizure to cluster brain extreme communities from multi-channel EEG data. With this approach, cluster centers can be interpreted as extremal prototypes, revealing the extremal dependence structures of communities of EEG channels. The clustering of channels can then be used as an exploratory tool to classify EEG channels into mutually asymptotically independent or asymptotically dependent groups. We apply the Club Exco method to investigate the differences in EEG brain connectivity networks between the pre- and post-seizure phases.
|
Authors who are presenting talks have a * after their name.
