Integration of epilepsy surgery and automatic seizure onset identification algorithm based on seizure patterns using SEEG.

Accurate identification of an epileptogenic zone (EZ) is crucial for successful epilepsy surgery. Stereotactic intracranial electroencephalography (SEEG) has contributed to improving surgical outcomes through assessing epileptic networks prior to surgery. However, SEEG data analysis has primarily relied on visual interpretation of epileptologists and clinical features of seizure types, leading to challenges in objective and quantitative surgical planning because it is difficult to visually evaluate High-Frequency Oscillation (HFO) or other epilepsy-related waveforms. Although the Epileptogenicity Index (EI) estimates epileptogenicity quantitatively based on temporal and energy parameters of brain regions, its accuracy is not always high enough. Thus, we propose a new automatic algorithm for detecting the seizure onset, referred to as TAILOR (Tailored Algorithm for Ictal Localization and Onset pRediction), which can determine the order of seizure onsets with high temporal resolution based on each patient's power spectrum patterns for tailor-made. In addition, the epileptic network can be estimated according to the order of seizure onsets determined with the proposed TAILOR. We retrospectively analyzed the clinical SEEG using TAILOR and confirmed that the actual surgical areas were ranked first as surgical areas in six out of eight cases.Clinical Relevance- We propose TAILOR as an automated algorithm for detecting seizure onset based on the power spectrum pattern. Estimating epileptic networks according to detailed seizure onset orders is anticipated to improve the accuracy of epilepsy treatment and surgery.