Neonatal Seizure Detection Using Deep Convolutional Neural Networks.

Amir H Ansari,Alexander Caicedo,Gunnar Naulaers,Perumpillichira J Cherian,Sabine Van Huffel,Maarten De Vos

doi:10.1142/s0129065718500119

Abstract

Identifying a core set of features is one of the most important steps in the development of an automated seizure detector. In most of the published studies describing features and seizure classifiers, the features were hand-engineered, which may not be optimal. The main goal of the present paper is using deep convolutional neural networks (CNNs) and random forest to automatically optimize feature selection and classification. The input of the proposed classifier is raw multi-channel EEG and the output is the class label: seizure/nonseizure. By training this network, the required features are optimized, while fitting a nonlinear classifier on the features. After training the network with EEG recordings of 26 neonates, five end layers performing the classification were replaced with a random forest classifier in order to improve the performance. This resulted in a false alarm rate of 0.9 per hour and seizure detection rate of 77% using a test set of EEG recordings of 22 neonates that also included dubious seizures. The newly proposed CNN classifier outperformed three data-driven feature-based approaches and performed similar to a previously developed heuristic method.

Highlights

Neonatal seizures usually indicate serious neurological dysfunction, and could potentially worsen underlying brain injury.1,2 The majority of neonatal seizures are acute symptomatic events, unlike the unprovoked epileptic seizures observed in older children and adults.2,3 These seizures may have nonexistent or subtle clinical manifestations, which may resemble normal behavior, such as lip smacking, sucking, chewing, and blinking
The convolutional neural network (CNN) features which were removed by the pruning process are marked with an asterisk (∗)
A novel neonatal seizure detector using convolutional neural networks and random forest was introduced in this paper

Summary

Introduction

Neonatal seizures usually indicate serious neurological dysfunction, and could potentially worsen underlying brain injury. The majority of neonatal seizures are acute symptomatic events, unlike the unprovoked epileptic seizures observed in older children and adults. These seizures may have nonexistent or subtle clinical manifestations, which may resemble normal behavior, such as lip smacking, sucking, chewing, and blinking. The majority of neonatal seizures are acute symptomatic events, unlike the unprovoked epileptic seizures observed in older children and adults.. The majority of neonatal seizures are acute symptomatic events, unlike the unprovoked epileptic seizures observed in older children and adults.2,3 These seizures may have nonexistent or subtle clinical manifestations, which may resemble normal behavior, such as lip smacking, sucking, chewing, and blinking. It has been shown that the most accurate method for their detection is visual interpretation of continuous multichannel EEG along with video by an expert clinical neurophysiologist.. It has been shown that the most accurate method for their detection is visual interpretation of continuous multichannel EEG along with video by an expert clinical neurophysiologist.1 Such interpretation is extremely labor-intensive, time-consuming, and importantly, needs special expertise which is not available around the clock in many neonatal intensive care units (NICUs). A reliable and accurate automated neonatal seizure detector using multi-channel continuous EEG can be a very helpful supportive tool, for the NICUs

Objectives

Methods

Results

Discussion

Conclusion