Computer-assisted EEG diagnostic review for idiopathic generalized epilepsy

Epilepsy diagnosis can be costly, time-consuming, and not uncommonly inaccurate. The reference standard diagnostic monitoring is continuous video-electroencephalography (EEG) monitoring, ideally capturing all events or concordant interictal discharges. Automating EEG data review would save time resources, thus enabling more people to receive also potentially heralding a quantitative approach therapeutic outcomes. There substantial research into the automated detection of seizures epileptic activity from EEG. However, software widely used in clinic, despite numerous published algorithms, few methods have regulatory approval for detecting This study reports on deep learning algorithm computer-assisted review. Deep convolutional neural networks were trained detect discharges using preexisting dataset over 6000 labelled cohort 103 patients with idiopathic generalized epilepsy (IGE). Patients underwent 24-hour ambulatory outpatient EEG, curated confirmed independently by two specialists (Seneviratne et al., 2016). resulting was then scalp seven (four IGE three mimicking seizures) validate performance clinical setting. showed state-of-the-art mean sensitivity >95% corresponding false positive rate 1 per minute. Importantly, case studies that reduced human 80%–99%, without compromising event accuracy. presented results demonstrate increase speed accuracy assessment has potential greatly improve article part Special Issue "NEWroscience 2018".