Threshold based MEG data classification for healthy and epileptic subjects

The most commonly used clinical tool for initial diagnosis of epilepsy is electroencephalogram (EEG). Recent advances in magnetoencephalography (MEG) technology provide a new source of information to analyze brain activities. In order to determine whether or not particular subjects' brain signals exhibit epileptic activities, epileptologists often spend considerable amount of time to review MEG recordings. This paper proposes a new algorithm for automatic classification of MEG data into two classes: data that belongs to healthy subjects and data that belongs to epileptic subjects. The classifier makes use of energy values of Delta and Theta bands. The effectiveness of proposed classifier has been tested using real MEG data obtained from 35 healthy subjects and 35 epileptic patients. Results obtained from the classifier show that the proposed classifier can be used as a tool in the initial diagnosis phase of epilepsy.

[1]  S. Taulu,et al.  Spatiotemporal signal space separation method for rejecting nearby interference in MEG measurements , 2006, Physics in medicine and biology.

[2]  Mojtaba Bandarabadi,et al.  Epileptic seizure prediction using relative spectral power features , 2015, Clinical Neurophysiology.

[3]  Fathi E. Abd El-Samie,et al.  EEG seizure detection and prediction algorithms: a survey , 2014, EURASIP J. Adv. Signal Process..

[4]  D. Cohen.  Magnetoencephalography: Detection of the Brain's Electrical Activity with a Superconducting Magnetometer , 1972, Science.

[5]  E. Halgren,et al.  Spatiotemporal mapping of brain activity by integration of multiple imaging modalities , 2001, Current Opinion in Neurobiology.

[6]  A. Schulze-Bonhage,et al.  Views of patients with epilepsy on seizure prediction devices , 2010, Epilepsy & Behavior.

[7]  N. Thakor,et al.  Quantitative EEG analysis methods and clinical applications , 2009 .

[8]  G. Barkley,et al.  Controversies in neurophysiology. MEG is superior to EEG in localization of interictal epileptiform activity: Pro , 2004, Clinical Neurophysiology.

[9]  K. Abdel-Aziz,et al.  Automatic Epileptic Seizure Detection Using Scalp EEG and Advanced Artificial Intelligence Techniques , 2015, BioMed research international.

[10]  E. Halgren,et al.  Dynamic Statistical Parametric Mapping Combining fMRI and MEG for High-Resolution Imaging of Cortical Activity , 2000, Neuron.

[11]  Saleh A. Alshebeili,et al.  MEG data classification for healthy and epileptic subjects using linear discriminant analysis , 2015, 2015 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT).

[12]  Gonzalo Alarcón,et al.  The Value of Magnetoencephalography to Guide Electrode Implantation in Epilepsy , 2013, Brain Topography.

[13]  Saleh A. Alshebeili,et al.  Enhancing the reliability of epileptic seizure alarms for scalp EEG signals , 2015, 2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP).