Efficient epileptic seizure detection by a combined IMF-VoE feature

Automatic seizure detection from the electroencephalogram (EEG) plays an important role in an on-demand closed-loop therapeutic system. A new feature, called IMF-VoE, is proposed to predict the occurrence of seizures. The IMF-VoE feature combines three intrinsic mode functions (IMFs) from the empirical mode decomposition of a EEG signal and the variance of the range between the upper and lower envelopes (VoE) of the signal. These multiple cues encode the intrinsic characteristics of seizure states, thus are able to distinguish them from the background. The feature is tested on 80.4 hours of EEG data with 10 seizures of 4 patients. The sensitivity of 100% is obtained with a low false detection rate of 0.16 per hour. Average time delays are 19.4s, 13.2s, and 10.7s at the false detection rates of 0.16 per hour, 0.27 per hour, and 0.41 per hour respectively, when different thresholds are used. The result is competitive among recent studies. In addition, since the IMF-VoE is compact, the detection system is of high computational efficiency and able to run in real time.

[1]  Hojjat Adeli,et al.  Principal Component Analysis-Enhanced Cosine Radial Basis Function Neural Network for Robust Epilepsy and Seizure Detection , 2008, IEEE Transactions on Biomedical Engineering.

[2]  Hasan Ocak,et al.  Automatic detection of epileptic seizures in EEG using discrete wavelet transform and approximate entropy , 2009, Expert Syst. Appl..

[3]  Ram Bilas Pachori,et al.  Discrimination between Ictal and Seizure-Free EEG Signals Using Empirical Mode Decomposition , 2008, J. Electr. Comput. Eng..

[4]  A. Aertsen,et al.  Detecting Epileptic Seizures in Long-term Human EEG: A New Approach to Automatic Online and Real-Time Detection and Classification of Polymorphic Seizure Patterns , 2008, Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society.

[5]  J. Gotman,et al.  A system to detect the onset of epileptic seizures in scalp EEG , 2005, Clinical Neurophysiology.

[6]  K. M. Kelly,et al.  Assessment of a scalp EEG-based automated seizure detection system , 2010, Clinical Neurophysiology.

[7]  Stephen M. Myers,et al.  Seizure prediction: Methods , 2011, Epilepsy & Behavior.

[8]  G. Vachtsevanos,et al.  Epileptic Seizures May Begin Hours in Advance of Clinical Onset A Report of Five Patients , 2001, Neuron.

[9]  J. Gotman,et al.  An automatic warning system for epileptic seizures recorded on intracerebral EEGs , 2005, Clinical Neurophysiology.

[10]  Xiaoli Li,et al.  Detection of Epileptic Spikes with Empirical Mode Decomposition and Nonlinear Energy Operator , 2005, ISNN.

[11]  M. De Vos,et al.  Automated artifact removal as preprocessing refines neonatal seizure detection , 2011, Clinical Neurophysiology.

[12]  Eric Laciar,et al.  An energy-based detection algorithm of epileptic seizures in EEG records , 2009, 2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[13]  Eric Laciar,et al.  Multiparametric detection of epileptic seizures using Empirical Mode Decomposition of EEG records , 2010, 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology.