Grid Computing Technology and the Recurrence Quantification Analysis to Predict Seizure Occurrence in Patients Affected by Drug-Resistant Epilepsy

Abstract

Nowadays, a hot topic in the field of epilepsy research is the detection of any reliable marker, embedded in the electroencephalograms (EEGs), that can be exploited to predict the seizure with a sufficient advance notice. A useful analytical tool which may help epileptologists to unveil significant patterns in EEGs of people suffering from epilepsy is the Recurrence Quantification Analysis (RQA). This technique can be easily exploited by researchers since RQA software applications and related source codes are freely available. Nevertheless, the analysis of extensive EEGs can be considerably CPU-time-consuming so researchers are often obliged to strongly reduce the amount of data RQA is applied to. High throughput computing appears as the best solution to solve this problem. In this paper we present the preliminary results of the RQA performed on the EEGs of four epileptic patients who underwent pre-surgical evaluation for the resection of epileptic foci. In this study, EEGs were segmented in epochs of proper length each one analysed independently from the others using a Grid computing infrastructure.