Time-Varying EEG Correlations Improve Automated Neonatal Seizure Detection

The aim of this study was to develop methods for detecting the nonstationary periodic characteristics of neonatal electroencephalographic (EEG) seizures by adapting estimates of the correlation both in the time (spike correlation; SC) and time-frequency domain (time-frequency correlation; TFC). These measures were incorporated into a seizure detection algorithm (SDA) based on a support vector machine to detect periods of seizure and nonseizure. The performance of these nonstationary correlation measures was evaluated using EEG recordings from 79 term neonates annotated by three human experts. The proposed measures were highly discriminative for seizure detection (median AUCSC : 0.933 IQR: 0.821-0.975, median AUCTFC : 0.883 IQR: 0.707-0.931). The resultant SDA applied to multi-channel recordings had a median AUC of 0.988 (IQR: 0.931-0.998) when compared to consensus annotations, outperformed two state-of-the-art SDAs (p < 0.001) and was noninferior to the human expert for 73/79 of neonates.

[1]  Sampsa Vanhatalo,et al.  Detecting bursts in the EEG of very and extremely premature infants using a multi-feature approach , 2017, Medical engineering & physics.

[2]  B Boashash,et al.  A nonparametric feature for neonatal EEG seizure detection based on a representation of pseudo-periodicity. , 2012, Medical engineering & physics.

[3]  G. Lightbody,et al.  An Automated System for Grading EEG Abnormality in Term Neonates with Hypoxic-Ischaemic Encephalopathy , 2012, Annals of Biomedical Engineering.

[4]  T. Inder,et al.  Seizure detection algorithm for neonates based on wave-sequence analysis , 2006, Clinical Neurophysiology.

[5]  A. Legido,et al.  The Exact Ictal and Interictal Duration of Electroencephalographic Neonatal Seizures , 1987, Epilepsia.

[6]  A M Bye,et al.  Spatial and Temporal Characteristics of Neonatal Seizures , 1995, Epilepsia.

[7]  Hojjat Adeli,et al.  A Wavelet-Chaos Methodology for Analysis of EEGs and EEG Subbands to Detect Seizure and Epilepsy , 2007, IEEE Transactions on Biomedical Engineering.

[8]  Qi Wu,et al.  Epileptic Seizure Detection with Log-Euclidean Gaussian Kernel-Based Sparse Representation , 2016, Int. J. Neural Syst..

[9]  Marina De Vos,et al.  Improved multi-stage neonatal seizure detection using a heuristic classifier and a data-driven post-processor , 2016, Clinical Neurophysiology.

[10]  G. Lightbody,et al.  EEG-based neonatal seizure detection with Support Vector Machines , 2011, Clinical Neurophysiology.

[11]  J. Frost,et al.  A Multistage System for the Automated Detection of Epileptic Seizures in Neonatal Electroencephalography , 2009, Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society.

[12]  William P. Marnane,et al.  Neonatal Seizure Detection Using Atomic Decomposition With a Novel Dictionary , 2014, IEEE Transactions on Biomedical Engineering.

[13]  Sampsa Vanhatalo,et al.  Interobserver agreement for neonatal seizure detection using multichannel EEG , 2015, Annals of clinical and translational neurology.

[14]  S. Huffel,et al.  Automated neonatal seizure detection mimicking a human observer reading EEG , 2008, Clinical Neurophysiology.

[15]  J. Gotman,et al.  Automatic seizure detection in the newborn: methods and initial evaluation. , 1997, Electroencephalography and clinical neurophysiology.

[16]  Qi Wu,et al.  An Improved Sparse Representation over Learned Dictionary Method for Seizure Detection , 2016, Int. J. Neural Syst..

[17]  N. Kasdin.  Discrete simulation of colored noise and stochastic processes and 1/fα power law noise generation , 1995, Proc. IEEE.

[18]  Boualem Boashash,et al.  Estimating and interpreting the instantaneous frequency of a signal. I. Fundamentals , 1992, Proc. IEEE.

[19]  Scott B. Wilson,et al.  Spike detection: Inter-reader agreement and a statistical Turing test on a large data set , 2017, Clinical Neurophysiology.

[20]  G. Lightbody,et al.  A comparison of quantitative EEG features for neonatal seizure detection , 2008, Clinical Neurophysiology.

[21]  S. Vanhatalo,et al.  The effect of reducing EEG electrode number on the visual interpretation of the human expert for neonatal seizure detection , 2018, Clinical Neurophysiology.

[22]  Boualem Boashash,et al.  Automatic signal abnormality detection using time-frequency features and machine learning: A newborn EEG seizure case study , 2016, Knowl. Based Syst..

[23]  G B Boylan,et al.  Defining the gap between electrographic seizure burden, clinical expression and staff recognition of neonatal seizures , 2007, Archives of Disease in Childhood Fetal and Neonatal Edition.

[24]  G. Lightbody,et al.  Performance assessment for EEG-based neonatal seizure detectors , 2011, Clinical Neurophysiology.

[25]  B Boashash,et al.  A time-frequency approach for newborn seizure detection. , 2001, IEEE engineering in medicine and biology magazine : the quarterly magazine of the Engineering in Medicine & Biology Society.

[26]  B. Boashash,et al.  Preprocessing and time-frequency analysis of newborn EEG seizures , 2001, IEEE Engineering in Medicine and Biology Magazine.

[27]  Cristiano Rizzo,et al.  Detection of subclinical electroencephalographic seizure patterns with multichannel amplitude-integrated EEG in full-term neonates , 2009, Clinical Neurophysiology.

[28]  J. Fleiss.  Measuring nominal scale agreement among many raters. , 1971 .

[29]  Sabine Van Huffel,et al.  Weighted Performance Metrics for Automatic Neonatal Seizure Detection Using Multiscored EEG Data , 2018, IEEE Journal of Biomedical and Health Informatics.

[30]  Theiler,et al.  Spurious dimension from correlation algorithms applied to limited time-series data. , 1986, Physical review. A, General physics.

[31]  B. Boashash,et al.  A Nonlinear Model of Newborn EEG with Nonstationary Inputs , 2010, Annals of Biomedical Engineering.

[32]  Paul B. Colditz,et al.  A computer-aided detection of EEG seizures in infants: a singular-spectrum approach and performance comparison , 2002, IEEE Transactions on Biomedical Engineering.

[33]  S. E. Rose,et al.  Seizures are associated with brain injury severity in a neonatal model of hypoxia–ischemia , 2010, Neuroscience.

[34]  Scott B. Wilson,et al.  Seizure detection: correlation of human experts , 2003, Clinical Neurophysiology.

[35]  Mostefa Mesbah,et al.  A Nonstationary Model of Newborn EEG , 2007, IEEE Transactions on Biomedical Engineering.

[36]  D B Vigneron,et al.  Seizure-associated brain injury in term newborns with perinatal asphyxia , 2002, Neurology.

[37]  William P. Marnane,et al.  Robust neonatal EEG seizure Detection through Adaptive Background Modeling , 2013, Int. J. Neural Syst..

[38]  M. Brandon Westover,et al.  What it should mean for an algorithm to pass a statistical Turing test for detection of epileptiform discharges , 2017, Clinical Neurophysiology.

[39]  G B Boylan,et al.  Gaussian mixture models for classification of neonatal seizures using EEG , 2010, Physiological measurement.

[40]  T. Higuchi.  Approach to an irregular time series on the basis of the fractal theory , 1988 .

[41]  A. Liu,et al.  Detection of neonatal seizures through computerized EEG analysis. , 1992, Electroencephalography and clinical neurophysiology.