Identifikasi Keadaan Mata Berdasarkan Sinyal Elektroensefalografi (EEG) Berbasis Metode Transformasi Wavelet Diskrit (TWD)

Azmi, Dzakia Fathimatul (2025) Identifikasi Keadaan Mata Berdasarkan Sinyal Elektroensefalografi (EEG) Berbasis Metode Transformasi Wavelet Diskrit (TWD). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003211161-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Penelitian ini bertujuan untuk mengidentifikasi keadaan mata menggunakan sinyal elektroensefalografi (EEG) dengan pendekatan Transformasi Wavelet Diskrit (TWD). Analisis berbasis metode TWD digunakan untuk mengatasi sinyal EEG yang bersifat acak dan dinamis. Perbedaan kekuatan relatif antara kondisi mata tertutup dan terbuka mencerminkan perubahan respons sensorik visual dan fokus kognitif, khususnya di area frontal otak. Data EEG yang digunakan berasal dari dua saluran elektrode bagian anterior frontal (AF3 dan AF4) yang kemudian disegmentasi dan didekomposisi hingga level keempat menggunakan TWD. Hasil dekomposisi berupa koefisien aproksimasi dan detail yang selanjutnya ditransformasi menjadi dua jenis fitur, yaitu energi wavelet relatif dan entropi Shannon. Kedua jenis fitur ini digunakan sebagai input untuk membangun model klasifikasi menggunakan dua metode, yakni Linear Discriminant Analysis (LDA) dan Support Vector Machine (SVM). Evaluasi performa model klasifikasi dilakukan dengan menggunakan Stratified K-Fold Cross Validation (K = 5). Hasil evaluasi menunjukkan bahwa model klasifikasi terbaik diperoleh dari algoritma SVM menggunakan fitur entropi Shannon dengan parameter C = 1 dan γ = 1. Model mencapai rata-rata akurasi sebesar 83,6% dengan waktu eksekusi 0,0527 detik. Penelitian ini diharapkan dapat berkontribusi pada pengembangan sistem pemantauan kesehatan yang lebih baik.
=========================================================================================================================================
This study aims to identify eye conditions using electroencephalography (EEG) signals with the Discrete Wavelet Transform (DWT) approach. The DWT method-based analysis is used to overcome the random and dynamic nature of EEG signals. Differences in relative strength between closed and open eye states reflect visual sensory responses and cognitive focus, particularly in the frontal area of the brain. The EEG data used was collected from two anterior frontal electrode channels (AF3 and AF4) which are then segmented and decomposed to the fourth level using TWD. The decomposition results in the form of approximation and detail coefficients which are then transformed into two types of features, namely relative wavelet energy and Shannon entropy. These two types of features are used as inputs to build classification models using two methods, namely Linear Discriminant Analysis (LDA) and Support Vector Machine (SVM). The performance evaluation of the classification model was carried out using Stratified K-Fold Cross Validation (K = 5). The evaluation results showed that the best classification model was obtained from the SVM algorithm using Shannon entropy features with parameters C = 1 and γ = 1. The model achieved an average accuracy of 83.6% with an execution time of 0.0527 seconds. This research is expected to contribute to the development of better health monitoring systems.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	EEG, LDA, SVM, TWD, EEG, LDA, SVM, DWT
Subjects:	H Social Sciences > HD Industries. Land use. Labor > HD108 Classification (Theory. Method. Relation to other subjects ) Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA403.3 Wavelets (Mathematics) Q Science > QP Physiology > Q376.5 Electroencephalography (EEG) T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Dzakia Fathimatul Azmi
Date Deposited:	05 Aug 2025 02:20
Last Modified:	05 Aug 2025 02:20
URI:	http://repository.its.ac.id/id/eprint/127286

Actions (login required)

View Item