Independent Component Analysis of Ictal EEG in Medial Temporal Lobe Epilepsy

Abstract

Summary:Purpose: Application of independent component analysis (ICA) to interictal EEGs and to event-related potentials has helped noise reduction and source localization. However, ICA has not been used for the analysis of ictal EEGs in partial seizures. In this study, we applied ICA to the ictal EEGs of patients with medial temporal lobe epilepsy (TLE) and investigated whether ictal components can be separated and whether they indicate correct lateralization.Methods: Twenty-four EEGs from medial TLE patients were analyzed with the extended ICA algorithm. Among the resultant 20 components in each EEG, we selected components with an ictal nature and reviewed their corresponding topographic maps for the lateralization. We then applied quantitative methods for the verification of increased quality of the reconstructed EEGs.Results: All ictal EEGs were successfully decomposed into one or more ictal components and nonictal components. After EEG reconstruction with exclusion of artifacts, the lateralizing power of the ictal EEG was increased from 75 to 96%.Conclusions: ICA can separate successfully the manifold components of ictal rhythms and can improve EEG quality. Key Words: Independent component analysis-Ictal component-Ictal EEG-Medial temporal lobe epilepsyArtifact.EEG recording is an essential step to localize irritative and ictal-onset zones, especially when epilepsy surgery is being considered. To localize the ictal-onset zone, a well-trained epileptologist visually inspects the EEG recordings. Identification of an unambiguous ictal-onset zone is often difficult because of unwanted artifacts arising from muscle contraction or eye movements. Recently a new data-processing technique, independent component analysis (ICA), was developed for the purpose of resolving multiple mixtures of data into statistically independent components (1,2). If we apply this method of analysis to EEG data, the EEG can be decomposed into spatiotemporal components that have fixed potential field distributions and maximally independent waveforms. The method has been found to be quite successful in separating artifacts or noisy components from the multichannel interictal EEG (3), recordings of event-related potentials (4), and functional magnetic resonance imaging (MRI) (5).We applied ICA to multichannel ictal EEG recordings in medial temporal lobe epilepsy (TLE). The ICA transforms the multichannel EEG data into an equal number of spatial patterns and their associated temporal waveforms, which are statistically independent of each other. The purpose of this study was to provide answers for the following two questions: Can ICA separate successfully ictal rhythms from noisy ictal EEGs of partial seizures? Can ICA help to lateralize medial TLE? MATERIALS AND METHODSWe applied ICA to 11 ictal EEGs from seven right medial TLE patients and to 13 EEGs from seven left medial TLE patients. All patients had been seizure free for more than a year after standard anterior temporal lobectomy with amygdalohippocampectomy. The diagnosis of each patien...

Independent Component Analysis Removing Artifacts in Ictal Recordings

“…In our study, the use of ICA in ictal recordings was clearly shown to be beneficial for recognizing the ictal pattern, the time of the ictal onset, and the exact localization of the ictal focus. In other ictal studies, ICA has proven its usefulness in separating components of the seizure itself (25). Nam et al (25) found that ICA was useful to find components of the EEG with good correlation with the focus of the seizure, as demonstrated by correspondence with the resulting topographic maps for those components in temporal lobe seizures.…”

“…ICA may be helpful in identifying components generated from different brain regions. In EEG or MEG signals, ICA has been used to separate components of the epileptiform activity (20,21), to eliminate artifacts (22,23), to detect spikes automatically (24), and to locate components of ictal activity (25). The possibility of its combined use with other techniques is interesting, as a means of attaining deeper knowledge of the genesis of brain activity (26).…”

How Many Electrodes Are Really Needed for EEG-Based Mobile Brain Imaging?

“…Another area in which ICA of EEG has proved useful is seizure detection and localization [30]. Seizures are often marked by large amounts of movement artifact that can pose similar challenges as locomotion to the interpretation of EEG recordings [31].…”

Block term decomposition for modelling epileptic seizures

“…The topographic maps corresponding to the ictal component can indicate the lateralisation of the seizure [12]. CPD of channel × time × frequency EEG tensors can provide even more accurate spatial information: it was shown that the topographic map of the ictal source coincides with the clinically determined seizure onset region [11,13].…”