Citation

Subbramania Pattar, Deivasigamani (2020) Computer aided automatic detection and diagnosis system for epileptic seizure in brain. PhD thesis, Multimedia University.

Full text not available from this repository.

Abstract

Epileptic seizures can be detected using Electroencephalogram (EEG) signals which are captured from the brain. Localization may make epileptic surgery necessary. In this thesis, a computer aided method for the identification of EEG signals has been proposed through soft computing techniques. The EEG signal is decomposed by Dual-Tree Complex Wavelet Transform (DT-CWT) which produces the coefficient metrics. The features from this coefficient metric are used to train the Adaptive Neuro Fuzzy Inference System (ANFIS) classifier for the identification of focal EEG signals. The classifications and the diagnosis of EEG signals are also proposed through this thesis. This proposed method consists of two consecutive stages as: detection module and diagnosis module. The detection module firstly detects the focal signals from non-focal signals using the ANFIS soft computing technique. The diagnosis module then classifies the detected focal signal either into ‘Early’ or ‘Advanced’. The methods presented in this thesis for the detection and diagnosis of EEG signals require a huge number of signals from various sources, such as ANFIS and Neural Network (NN), during the training mode of this soft computing algorithm. In addition, the detection and diagnosis results of these methods, using soft computing algorithms are not optimum with respect to the radiologist’s clinical results. These limitations are now overcome by proposing a Deep Learning Neural Network (DLNN) algorithm which requires fewer signal samples. Also proposed in this research work is focal signal detection and its diagnosis procedure by using the DLNN algorithm. This method also uses Continuous Wavelet Transform (CWT), feature extraction which is followed by an optimization algorithm Genetic Algorithm (GA) and classifications using a DLNN classification approach. Next, the classified focal EEG signals are diagnosed either into ‘Early’ or ‘Advanced’ based on their severity level by using the DLNN algorithm.

Item Type:	Thesis (PhD)
Additional Information:	Call No.: TA345 .P38 2020
Uncontrolled Keywords:	Computer-aided engineering
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) > TA329-348 Engineering mathematics. Engineering analysis
Divisions:	Faculty of Engineering (FOE)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	02 Oct 2025 02:00
Last Modified:	02 Oct 2025 02:00
URII:	http://shdl.mmu.edu.my/id/eprint/14640

Downloads

Edit (login required)