Principal Components Analysis of EEG Signals for Epileptic Patient Identification

Abstract

According to the behavior of its neuronal connections, it is possible to determine if the brain suffers from abnormalities such as epilepsy. This disease produces seizures and alters the patient’s behavior and lifestyle. Neurologists employ the electroencephalogram (EEG) to diagnose the disease through brain signals. Neurologists visually analyze these signals, recognizing patterns, to identify some indication of brain disorder that allows for the epilepsy diagnosis. This article proposes a study, based on the Fourier analysis, through fast Fourier transformation and principal component analysis, to quantitatively identify patterns to diagnose and differentiate between healthy patients and those with the disease. Subsequently, principal component analysis can be used to classify patients, employing frequency bands as the signal features. Besides, it is made a classification comparison before and after using principal component analysis. The classification is performed via logistic regression, with a reduction from 5 to 4 dimensions, as well as from 8 to 7, achieving an improvement when there are 7 dimensions in the precision, recall, and F1 score metrics. The best results obtained, without PCA are: precision 0.560, recall 0.690, and F1 score 0.620; meanwhile, the best values obtained using PCA are: precision 0.734, recall 0.787, and F1 score 0.776.

Análise de técnicas de aprendizado de máquina em dados de eletroencefalograma com uso de PCA

“…Contudo, para a base de dados de EGG utilizada neste estudo, e com os algoritmos testados, não se percebeu uma melhora significativa da aplicação da técnica de PCA. Mesmo o trabalho mais similar realizado a este [Guerrero et al 2021] o qual utiliza uma base com menor quantidade de atributos, não obteve resultados tão promissores quanto o desejado. Desta forma, ainda são necessários estudos para verificar se o PCA é uma técnica que pode auxiliar na descoberta de conhecimento em sinais do tipo EEG.…”

Implementation of Closing Eyes Detection with Ear Sensor of Muse EEG Headband using Support Vector Machine Learning

“…Easy diagnosis is very helpful for patients with epilepsy. The commonly used diagnosis for this disease is by using an electroencephalography (EEG) signal which is suggested in [4]. With EEG, biopotential signals from the brain can be obtained from the surface of the patient's head.…”

Using Explainable Artificial Intelligence to Obtain Efficient Seizure-Detection Models Based on Electroencephalography Signals

“…Given the significance and intricacy of seizure detection, considerable efforts have been directed towards machine learning (ML)-based methodologies for the automated analysis of EEG signals. Supervised ML classifiers are commonly employed in this domain [ 7 , 8 , 9 , 10 , 11 , 12 ], along with deep learning (DL) techniques [ 13 , 14 ]. These approaches are geared towards developing models adept at binary classification [ 7 , 8 , 9 , 11 , 12 ] as well as multiclass classification tasks [ 15 , 16 ].…”

“…Supervised ML classifiers are commonly employed in this domain [ 7 , 8 , 9 , 10 , 11 , 12 ], along with deep learning (DL) techniques [ 13 , 14 ]. These approaches are geared towards developing models adept at binary classification [ 7 , 8 , 9 , 11 , 12 ] as well as multiclass classification tasks [ 15 , 16 ].…”

“…Time domain techniques often employ statistical metrics such as standard deviation, kurtosis, skewness, and mean [ 8 , 15 , 16 ]. Frequency domain approaches typically involve the use of Fast Fourier Transform (FFT) or Discrete Wavelet Transform (DWT) coefficients, either directly or through derived statistical measures of these coefficients [ 7 , 9 , 10 , 11 , 12 , 16 , 18 , 19 ]. Owing to the inherent non-linearity of EEG signals [ 20 ], certain studies also focus on extracting non-linear features, including Sample Entropy, Wavelet Entropy [ 21 ], and the Lyapunov exponent [ 22 ].…”