Epilepsy analysis using open source EDF tools for information science and data analytics

• Published in: International journal of communication systems 2020-09, Vol.33 (13), p.n/a
• Main Authors: Gurumoorthy, Sasikumar, Muppalaneni, Naresh Babu, Sekhar, Chandra, Sandhya Kumari, Golla
• Format: Article
• Language: English
• Subjects: analytics; dual‐tree discrete wavelet transform; Echocardiography; EDF BROWSER; EEG STUDIO; Electrocardiography; Electroencephalography; Electrooculography; Epilepsy; Frequency analysis; Frequency domain analysis; information; Muscles; Nonlinear analysis; Seizures; wavelet dependent thresholding denoising; wavelet packet decomposition
• ISSN: 1074-5351; 1099-1131
• DOI: 10.1002/dac.4095

Summary Electroencephalogram (EEG) is the signals that measure the electrical variances of brain using metal electrodes. We observe the EEG signals by using European Data Format (EDF) BROWSER and EEG STUDIO. By using EDF BROWSER, we can get the mean and frequency from the filtered output signal using band‐pass filter. Using EDF BROWSER, we can also perform Root Mean Square (RMS) and signal samples. Using EEG STUDIO, we can analyze the average frequency and standard deviation. Epileptic seizure prediction and detection are done by spike detection, frequency domain analysis, and nonlinear methods. EEG signal contains different artifacts like electrooculography (EOG), EKG, and electrocardiogram (ECG). ECG signals are produced by heart. EOG signals are produced by eyes. EMG signals are produced by muscle coordination. Clinically, electroencephalogram (EEG) refers to the recording of the brain's spontaneous electrical activity over a period of time from multiple electrodes placed on the scalp. EEG data display the signals of electrical variances of brain using metal electrodes. This study is carried on epilepsy disease. People with epilepsy can experience recurrent seizures and a temporary disturbance in the messaging systems between brain cells. Epileptic seizure can be detected with the variation of spikes in the EEG data, with frequency domain analysis, and by using other nonlinear methods. EEG signal contains different artifacts like electrooculography (EOG), electrocardiogram (ECG), and electromyogram (EMG). ECG signal artifacts are produced by the function of heart. EOG signal artifacts are produced because of the movement of eyes, and these variations are noticed in the EEG data. EMG signal artifacts are produced because of the muscles coordination. The EEG signal parameters are analyzed by using EDF BROWSER and EEG STUDIO. By using EDF BROWSER, mean, frequency values, and RMS values are taken from the filtered output signal. With the help of EEG STUDIO, it is analyzed the average frequency and standard deviation of filtered output signal. In this paper, wavelet packet decomposition, wavelet dependent thresholding denoising, and dual‐tree discrete wavelet transform are the signal processing techniques that has been applied to analyze the EEG signal for epilepsy disease, where dual‐tree discrete wavelet transform has removed most of the noise that present in the signal.