2019/02/28 by Melisa Maidana Capitán, Nuria Cámpora, Capitán, Melisa Maidana +7
Computer Science · Neuroscience · #Blind Source Separation Techniques #EEG and Brain-Computer Interfaces #FOS: Biological sciences #Neural dynamics and brain function #Neurons and Cognition (q-bio.NC)
paper · pdf · doi:10.48550/arxiv.1902.11236
openalex publication_date 2019/02/28 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28
The amount of power in different frequency bands of the electroencephalogram\n(EEG) carries information about the behavioral state of a subject. Hence,\nneurologists treating epileptic patients monitor the temporal evolution of the\ndifferent bands. We propose a covariance-based method to detect and\ncharacterize epileptic seizures operating on the band-filtered EEG signal. The\nalgorithm is unsupervised, and performs a principal component analysis of\nintra-cranial EEG recordings, detecting transient fluctuations of the power in\neach frequency band. Its simplicity makes it suitable for online\nimplementation. Good sampling of the non-ictal periods is required, while no\ndemands are imposed on the amount of data during ictal activity. We tested the\nmethod with 32 seizures registered in 5 patients. The area below the resulting\nreceiver-operating characteristic curves was 87 % for the detection of seizures\nand 91 % for the detection of recruited electrodes. To identify the\nbehaviorally relevant correlates of the physiological signal, we identified\ntransient changes in the variance of each band that were correlated with the\ndegree of loss of consciousness, the latter assessed by the so-called\nConsciousness Seizure Scale, summarizing the performance of the subject in a\nnumber of behavioral tests requested during seizures. We concluded that those\ncrisis with maximal impairment of consciousness tended to exhibit an increase\nof variance approximately 40 seconds after seizure onset, with predominant\npower in the theta and alpha bands, and reduced delta and beta activity.\n