Mahfuzah, Mustafa and Nor Anis Aneza, Lokman (2013) kNN Classification of Epilepsy Brainwaves. In: Malaysian Technical Universities Conference on Engineering & Technology (MUCET 2013) , 3-4 December 2013 , Kuantan, Pahang. pp. 1-2..
![[img]](http://umpir.ump.edu.my/style/images/fileicons/application_pdf.png)
|
PDF
EE-003.pdf Download (184kB) |
Abstract
Epilepsy is a disorder of the normal brain function by the existence of abnormal synchronous discharges in large groups of neurons in brain structures and it is estimated about 1% of the world’s population suffers from this disease [Tzallas et al., 2009]. It has been reported that the brainwave of Epilepsy patient mostly in sharp, spike and complex wave pattern [Tzallas et al., 2009]. In addition, Epilepsy brainwaves pattern lies in wide variety of Electroencephalogram (EEG) signals in formed of low-amplitude and polyspikes activity [Vargas et al., 2011]. Generally, this disease was examined through the brainwaves or EEG signals by clinical neurulogists. An EEG is a device to record the brainwaves in term of electrical activity from the brain. Brain patterns from wave shapes that are commonly sinusoidal and measured from peak to peak that range from 0.5 μV to 100 μV in amplitude. Moreover, the brainwaves have been categorized into four frequency bands, Beta (>13 Hz), Alpha (8-13 Hz), Theta (4-8 Hz) and Delta (0.5-4 Hz). All the frequency bands will be used to characterize the Epilepsy brainwave in terms of amplitude (voltage) and frequency [Mustafa et al., 2013]. The Epilepsy brainwaves were downloaded from http://www.vis.caltech.edu/~rodri/data.htm of Fp1 and Fp2 channels which is from rats. The brainwaves consists Epilepsy and non-Epilepsy samples. Then, the brainwaves were pre-processed to remove artefact (noise). Various methods had been introduced to detect spike-wave discharge in Epilepsy patient brainwave. Brainwave is nonstationary signal, therefore, time-frequency analysis is appropriate methods to analyse the signals[Tzallas et al., 2009, Vargas et al., 2011]. One of the most popular time-frequency analyses is ShortTime Fourier Transform (STFT). After the brainwaves were pre-processed, STFT was employed to the clean brainwaves. The STFT spectrogram was generated for four frequency bands of the samples.
| Item Type: | Conference or Workshop Item (Speech) |
|---|---|
| Uncontrolled Keywords: | kNN; Epilepsy; Brainwaves; EEG Signals |
| Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering |
| Faculty/Division: | Faculty of Electrical & Electronic Engineering |
| Depositing User: | ina |
| Date Deposited: | 12 Aug 2014 07:07 |
| Last Modified: | 11 Apr 2018 01:57 |
| URI: | http://umpir.ump.edu.my/id/eprint/5724 |
| Download Statistic: | View Download Statistics |
Actions (login required)
 |
View Item |