New Algorithm Predicts Onset of Epileptic Seizures

New research published in PLOS ONE suggests a mathematical algorithm can predict the onset of seizures in epilepsy patients up to 20 minutes in advance.

With an aim to predict symptoms from 0-5 minutes to 20-25 minutes prior to seizure onset, Negin Moghim and David W. Corne of Heriot-Watt University in the United Kingdom analyzed existing datasets on 21 epilepsy patients who underwent invasive electroencephalography (EEG).

Throughout the study, “204 feature time series were extracted for each patient, (which translated to) 34 distinct features for each of the 6 EEG channels,” the authors penned.

Using EEG data and a prediction algorithm, the researchers created the Advance Seizure Prediction via Pre-ictal Relabeling (ASPPR), a predictive model that reported success rates of “96.30% for prediction between 1 and 6 minutes in advance, 96.13% for prediction between 8 and 13 minutes in advance, 94.5% for prediction between 14 and 19 minutes in advance, and 94.2% for prediction between 20 and 25 minutes in advance.”

According to the study authors, ASPPR classifies data into seizure, pre-seizure, between seizures, and post-seizure states in order to accurately predict symptoms. Although machine-learning treatment options for epilepsy have not widely been explored, the current findings suggest symptoms of the disorder can be detected or prevented early in a non-invasive manner, according to a statement provided by Heriot-Watt University.

“Being able to predict seizures — and coupling this information with state-of-the-art medical device technology — we should soon be able to provide unobtrusive wearable devices that provide accurate advance warning of seizures and allows patients to take prompt action to minimize the risk to themselves,” Corne commented.