Epilepsy Diagnosis

In a recent submission to Annals of Neurology (currently available on bioarx), the algorithm that EpiScalp was shown to be able to detect epilepsy from resting-state scalp EEG.

While scalp EEG is important for diagnosing epilepsy, a single routine EEG is limited in its diagnostic value in current standard practice. Only a small percentage of routine EEGs show interictal epileptiform discharges (IEDs) and overall misdiagnosis rates of epilepsy are 20-30%. We developed EpiScalp to demonstrate how analyzing network properties in EEG recordings can be used to improve the speed and accuracy of epilepsy diagnosis – even in the absence of IEDs. EpiScalp produces a risk score between 0-1 from minutes of EEG data and does not rely on the presence of EEG abnormalities. Instead, our novel algorithm predicts epilepsy in resting-state (no seizure) brain networks using a dynamic network model (DNM).

In this multicenter study, we analyzed routine scalp EEGs from 203 patients with suspected epilepsy and normal initial EEGs. The patients’ diagnoses were later confirmed based on an epilepsy monitoring unit (EMU) admission. About 47% ultimately being diagnosed with epilepsy and 53% with non-epileptic conditions. EpiScalp achieved definitive diagnoses for 90 of the 203 patients with very low (epilepsy unlikely) and very high (epilepsy likely) risk scores, demonstrating remarkable accuracy, sensitivity, and specificity at 94%, 93%, and 93% respectively. For all 203 patients, EpiScalp could have reduced misdiagnoses to 26% (a 50% reduction) while preserving mid-range risk scores diagnosed by clinicians.

EpiScalp provides accurate diagnostic aid from a single initial EEG recording, even in more challenging epilepsy cases with normal initial EEGs. This may represent a paradigm shift in epilepsy diagnosis by deriving an objective measure of epilepsy likelihood from previously uninformative EEGs.