•  
  •  
 

Abstract

Electroencephalogram (EEG) is a convenient neuroimaging technique due to its non-invasive setup, practical usage, and high temporal resolution. EEG allows to detect brain electrical activity to diagnose neurological disorders. Epilepsy is a crucial neurologic disorder that is reasoned from occurrence of sudden and repeated seizures. The goal of this paper is to classify the focal (epileptogenic area) and non-focal (non-epileptogenic area) EEG records with cepstral coefficients and machine learning algorithms. Analysis is carried out using publicly available Bern-Barcelona EEG dataset. Mel Frequency Cepstral Coefficients (MFCC) are calculated from EEG epochs. Feature sets are normalized with z-score and dimension reduction is realized using Principal Component Analysis. Fine Tree, Quadratic Discriminant Analysis, Logistic Regression, Gaussian Naïve Bayes, Cubic Support Vector Machine, weighted k-nearest neighbors, and Bagged Trees are applied for classification stage. A value of k=10 is used for cross validation. All focal and non-focal EEG pairs are perfectly classified with acc., sen., spe., and F1-score of 100% and AUC with 1 via. Quadratic Discriminant Analysis, Logistic Regression, Cubic SVM and Weighted k-NN. Proposed work recommends MFCCs as a single marker and this provides less computation workload, practicality, and direct processing of focal / non-focal EEG time series. Proposed methodology in this paper serves one of the highest achievements to literature and can assist neurologist and physicians to validate their diagnosis.

Reviewers Suggestions

None

DOI

10.24012/dumf.1002081

Share

COinS