Clinical electroencephalography (EEG) records vast amounts of human complex data yet is still reviewed primarily by human readers. Deep Belief Nets (DBNs) are a relatively new type of multi-layer neural network commonly tested on two-dimensional image data, but are rarely applied to times-series data such as EEG. We apply DBNs in a semi-supervised paradigm to model EEG waveforms for classification and anomaly detection. DBN performance was comparable to standard classifiers on our EEG dataset, and classification time was found to be 1.7 to 103.7 times faster than the other high-performing classifiers. We demonstrate how the unsupervised step of DBN learning produces an autoencoder that can naturally be used in anomaly measurement. We compare the use of raw, unprocessed data—a rarity in automated physiological waveform analysis—to hand-chosen features and find that raw data produces comparable classification and better anomaly measurement performance. These results indicate that DBNs and raw data inputs may be more effective for online automated EEG waveform recognition than other common techniques.
Intracranial electroencephalography (IEEG) plays an important role in guiding epilepsy surgery in pediatric epilepsy patients. Recently, there has been increased interest in higher frequency components of clinical IEEG recordings and their potential relationship to epileptogenic brain tissue. We employ a previously validated, automated discrete gamma oscillation (GO) detection algorithm to determine the prevalence of discrete gamma events over prolonged, representative segments of IEEG recorded from ten patients. Approximately 8 h of IEEG, 16 randomly selected 30-min segments of continuous interictal IEEG per patient, were analyzed. The electrodes within the seizure onset zone were found to have significantly higher mean GO activity averaged across these 16 segments in five of the ten patients. There was observed variability between individual 30-min segments in these patients, indicating that longer recordings of interictal activity improved localization. Our data suggest this method of automated GO detection across long periods may be useful in planning epilepsy surgery in certain children with intractable epilepsy. Further research is required to help determine which patients would benefit from this technique.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.