Electronic Theses and Dissertation

Electroencephalography (eeg) adalah metode non-invasif yang digunakan untuk merekam aktivitas kelistrikan otak dan sering dimanfaatkan dalam analisis kondisi neurologis, seperti autism spectrum disorder (asd). namun, sinyal eeg sering terkontaminasi noise dari sumber internal dan eksternal, seperti pergerakan mata, aktivitas otot, dan detak jantung, yang dapat mengurangi kualitas sinyal serta menyulitkan proses analisis. penelitian ini bertujuan untuk membandingkan performansi metode principal component analysis (pca) dan multiscale principal component analysis (mspca) dalam mereduksi noise pada sinyal eeg penyandang asd dan individu normal. penelitian ini menggunakan dataset eeg sekunder dari universitas king abdul aziz, yang diolah menggunakan bahasa pemrograman python. evaluasi performansi dilakukan berdasarkan empat parameter utama, yaitu mean squared error (mse), signal-to-noise ratio (snr), dan percentage root mean square difference (prd). hasil penelitian ini diharapkan dapat mengidentifikasi metode yang lebih unggul dalam menghilangkan noise tanpa merusak struktur utama sinyal eeg. temuan ini diharapkan memberikan kontribusi signifikan dalam pengembangan teknik pengolahan sinyal eeg, khususnya pada populasi penyandang asd, serta mendukung analisis dan interpretasi sinyal eeg yang lebih akurat. kata kunci: electroencephalography (eeg), principal component analysis (pca), multiscale principal component analysis (mspca), autism spectrum disorder (asd), pengurangan derau, pengolahan sinyal.

Electroencephalography (EEG) serves as a non-invasive technique to capture the brain’s electrical activity and is frequently applied in studying neurological disorders such as Autism Spectrum Disorder (ASD). Nevertheless, EEG recordings are prone to contamination from various internal and external noise sources, including eye movements, muscle contractions, and cardiac signals. Which impair signal clarity and complicate subsequent analyses. This research focuses on evaluating and comparing the effectiveness of Principal Component Analysis (PCA) and Multiscale Principal Component Analysis (MSPCA) in denoising EEG signals obtained from individuals with ASD and neurotypical controls. The study utilizes secondary EEG data acquired from King Abdul Aziz University and processes it using Python programming. Perfomance assessment is based on key metrics including Mean Squared Error (MSE), Signal-to-Noise Ratio (SNR), and Percentage Root Mean Square Difference (PRD). The objective is to identify the method that optimally reduces noise while preserving the intrinsic EEG signal structure. The anticipated outcomes are expected to advance EEG signal processing methodologies, particulary for the ASD population, therby enhacing the precision of EEG signal analysis and interpretation. Keywords: Electroencephalography (EEG), Principal Component Analysis (PCA), Multiscale Principal Component Analysis (MSPCA), Autism Spectrum Disorder (ASD), Noise Reduction, Signal Processing.