#TITLE_ALTERNATIVE#

• Main Author: WIJAYANTI PUSPITA (NIM : 20112035) ; Pembimbing Tesis : Sapto Wahyu Indratno, Ph.D ; Prof. Dr. , JUNI
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/18777
• Institution: Institut Teknologi Bandung

Electroencephalography (EEG) is the most common test being used to diagnose epilepsy. Most of the EEG abnormal patterns in epilepsy patients are interictal waves which consist of spike and sharp waves. These two waves can be detected in detail by using Walsh transformation. Furthermore, the Bayesian classification method is applied to classify the spike waves and sharp waves resulting from Walsh transformation. Training data which consist of the original data of EEG record and the results of first and second order Walsh transformation are collected in this thesis. The gradient and the slope of the resulting data are used to construct the profile of the epileptogenic waves. The results show that the first order Walsh transformation for gradient treatment gives the best result of Bayesian classification. While the second order Walsh transformation is less sensitive to classify epileptogenic waves. This is due to the fact that the morphologic characteristics of the walsh transformation of spike and sharp waves are difficult to distinguish