Precise identification of epileptogenic zones in patients with intractable drug-resistant epilepsy is critical for successful epilepsy surgery. Numerous source-imaging algorithms for localizing epileptogenic zones based on scalp electroencephalography (EEG) and magnetoencephalography (MEG) have been developed and validated in simulation and experimental studies. Recently, intracranial EEG (iEEG)-based imaging of epileptogenic sources has attracted interest as a promising tool for presurgical evaluation of epilepsy; however, most iEEG studies have focused on localization of epileptogenic zones in focal epilepsy. In the present study, we investigated whether iEEG source imaging is a useful supplementary tool for identifying extended epileptogenic sources in secondary generalized epilepsy such as Lennox- Gastaut syndrome (LGS). To this end, we applied four different cortical source imaging algorithms, namely minimum norm estimation (MNE), low-resolution electromagnetic tomography (LORETA), standardized LORETA (sLORETA), and L p-norm estimation (p = 1.5, referred to as L p1.5), to artificial iEEG datasets generated assuming various source sizes and locations. We also applied these four algorithms to clinical ictal iEEG recordings acquired from a pediatric patient with LGS. Interestingly, the traditional MNE algorithm outperformed the other imaging algorithms in most of our experiments, particularly in cases when larger-sized sources were activated. Although sLORETA outperformed both LORETA and Lp1.5, its performance was not as good as that of MNE. Compared to the other algorithms, the performance of Lp1.5 decayed most rapidly as the source size increased. Our findings suggest that iEEG source imaging using MNE is a promising auxiliary tool for the identification of epileptogenic zones in secondary generalized epilepsy. We anticipate that our results will provide useful guidelines for selection of an appropriate imaging algorithm for iEEG source imaging studies.
Bibliographical noteFunding Information:
Acknowledgments This work was supported by a grant of the Korea Healthcare Technology R&D Project, Ministry for Health, Welfare & Family Affairs, Republic of Korea (A090579-0901-0000201).
All Science Journal Classification (ASJC) codes
- Radiological and Ultrasound Technology
- Radiology Nuclear Medicine and imaging
- Clinical Neurology