Fast Automated Interictal Spike Detection in iEEG/ECoG Recordings

article in a collection out of WoS and Scopus

English

MOTIVATION Interictal spikes have been established as an im- portant biomarker in surface EEG and intracranial iEEG recordings for some time (Staley et al., 2011). Spikes are used for clinical practice and research of epilepsy, ADHD and also in other areas(Barkmeier et al., 2012a). Although gold standard for interictal spike detection has been and still mainly is manual evaluation, it has been shown that higher consistency of results can be achieved by automated detection al- gorithm (Barkmeier et al., 2012b) also saving enor- mous amount of work for reviewers thus providing faster data analysis for research or even clinical prac- tice. OBJECTIVES Computational efficiency is not so important, when recordings from only few channels are processed and real-time detection is not necessary. Example of those would be recordings from rodents(Ovchinnikov et al., 2010). However, when processing intracranial record- ings from humans, in as much as 150 channels with 5 KHz sampling rate, which are in average 30 minutes long, computational time requirements gain great deal of importance. This algorithm has been designed to address this very issue. While several terabytes(just our institution) of such recordings are ready for pro- cessing, detection algorithm must have been designed to allow fast offline processing of intracranial record- ings or even real-time detection in at least hundreds of channels simultaneously. In order to process large signal data, memory access is crucial bottleneck for CPU processing, which puts high requirements on ef- fective cache utilization, reducing frequency of access to RAM.

KEŠNER, F.; CIMBÁLNÍK, J.; DOLEŽALOVÁ, I.; BRÁZDIL, M.; SEKANINA, L.

17. 10. 2015

Lisabon

1

4

3