Deep Learning of Simultaneous Intracranial and Scalp EEG for Prediction, Detection, and Lateralization of Mesial Temporal Lobe Seizures.

Fedor Panov,Lara Marcuse,Zan Li,Bülent Yener,Saadi Ghatan,Madeline Fields

doi:10.3389/fneur.2021.705119

Abstract

In people with drug resistant epilepsy (DRE), seizures are unpredictable, often occurring with little or no warning. The unpredictability causes anxiety and much of the morbidity and mortality of seizures. In this work, 102 seizures of mesial temporal lobe onset were analyzed from 19 patients with DRE who had simultaneous intracranial EEG (iEEG) and scalp EEG as part of their surgical evaluation. The first aim of this paper was to develop machine learning models for seizure prediction and detection (i) using iEEG only, (ii) scalp EEG only and (iii) jointly analyzing both iEEG and scalp EEG. The second goal was to test if machine learning could detect a seizure on scalp EEG when that seizure was not detectable by the human eye (surface negative) but was seen in iEEG. The final question was to determine if the deep learning algorithm could correctly lateralize the seizure onset. The seizure detection and prediction problems were addressed jointly by training Deep Neural Networks (DNN) on 4 classes: non-seizure, pre-seizure, left mesial temporal onset seizure and right mesial temporal onset seizure. To address these aims, the classification accuracy was tested using two deep neural networks (DNN) against 3 different types of similarity graphs which used different time series of EEG data. The convolutional neural network (CNN) with the Waxman similarity graph yielded the highest accuracy across all EEG data (iEEG, scalp EEG and combined). Specifically, 1 second epochs of EEG were correctly assigned to their seizure, pre-seizure, or non-seizure category over 98% of the time. Importantly, the pre-seizure state was classified correctly in the vast majority of epochs (>97%). Detection from scalp EEG data alone of surface negative seizures and the seizures with the delayed scalp onset (the surface negative portion) was over 97%. In addition, the model accurately lateralized all of the seizures from scalp data, including the surface negative seizures. This work suggests that highly accurate seizure prediction and detection is feasible using either intracranial or scalp EEG data. Furthermore, surface negative seizures can be accurately predicted, detected and lateralized with machine learning even when they are not visible to the human eye.

Highlights

Epilepsy is characterized by recurrent and unpredictable seizures
For the deep neural networks (DNN), we focused on Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) and reported the classification accuracy for the 3 class classification problems
Seizure onset time was defined as the time of the onset on intracranial EEG (iEEG) data as the scalp onset was often delayed and in 35 seizures not present at all

Summary

Introduction

Epilepsy is characterized by recurrent and unpredictable seizures This unpredictability is the core of suffering in the person with epilepsy. There is never a guarantee for a seizure free day The field, and this collection of articles, is working to address this suffering by improving the accuracy of seizure prediction, detection and forecasting. Forecasting differs from prediction by identifying a period of time, lasting hours to days, during which the person is more likely to have seizures based upon their known prior patterns and rhythms [1]. The goal is to provide a warning that a seizure is about to occur within minutes. For this warning to be useful, it must be accurate with a low false positive and a low false negative rate

Objectives

Methods

Results

Conclusion