Penghilangan Artefak Gerakan Mata pada Sinyal EEG dari Penderita Epilepsi menggunakan Wavelet-Independent Component Analysis

Rakhmah, Viprilia (2019) Penghilangan Artefak Gerakan Mata pada Sinyal EEG dari Penderita Epilepsi menggunakan Wavelet-Independent Component Analysis. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 07211540000028-Undergraduate_Theses.pdf Restricted to Repository staff only Download (6MB) | Request a copy

Abstract

Selama proses perekaman sinyal EEG yang berkisar kurang lebih 30 menit, pasien tentu mengalami gerakan tubuh yang disengaja maupun berasal dari instruksi dokter. Hal ini menyebabkan keabnormalan pada hasil sinyal EEG berupa sinyal asing yang terekam atau disebut dengan artefak gerak. Artefak dapat berasal dari beberapa gerakan seperti gerakan mata. Dibutuhkan suatu metode untuk menghilangkan artefak mata dari sinyal EEG agar tenaga medis tidak kesulitan dalam menganalisa hasil sinyal EEG pasien. Dalam penelitian ini diusulkan suatu sistem penghilangan artefak gerakan mata menggunakan Wavelet- Independent Component Analysis (ICA). Metode Discrete Wavelet Transform mendekomposisikan sinyal EEG menjadi banyak komponen untuk sinyal masukan ke metode ICA. Pembuangan artefak dilakukan dengan ICA, perhitungan variansi dan high pass filter. Hasil yang diperoleh dalam penelitian ini berupa sinyal EEG tanpa artefak mata dievaluasi secara kuantitatif dan kualitatif. Validasi kuantitatif dengan perhitungan normalized mean squared error (NMSE) menunjukkan nilai error rata-rata 0.048 dB dan hasil sinyal EEG yang tetap mempertahankan informasi penting berupa sinyal kejang.
==============================================================================================================================================
During the EEG recording process which take around 30 minutes, the patient absolutely does some physical movements which these movements cause the abnormalities on EEG signal recording as strange signal or motion artifacts. Motion artifacts can come from some movements like eye movement. We need a method to remove the ocular artifacts from EEG signal so medical personnel does not get difficult in analyzing patient recording’s result. This study propose ocular artifact removal system using Wavelet-Independent Component Analysis (ICA). Discrete Wavelet Transform (DWT) decompose EEG signal to many components as input of ICA method. Artifact removal is done by ICA, variance calculation and high pass filter. The result of this study is ocular artifact-free EEG signal evaluated in quantitive and qualitative. Quantitive validation with normalized mean squared error (NMSE) shows average error in 0.048 dB and clean EEG signal which maintain important data like epileptiform signal.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	EEG, artifact, ICA, normalized mean squared error
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing.
Divisions:	Faculty of Electrical Technology > Computer Engineering > 90243-(S1) Undergraduate Thesis
Depositing User:	Viprilia Rakhmah
Date Deposited:	06 Aug 2026 03:45
Last Modified:	06 Aug 2026 03:45
URI:	http://repository.its.ac.id/id/eprint/70377

Actions (login required)

View Item