Node Selection Bayesian Inference MODWPT Method for EEG Analysis
Indrawata Wardhana, Hery Afriyadi, Ahmad Nasukha · 2024
EEG signals are inherently complex and subtle, reflecting intricate brain activity patterns. However, the presence of noise can significantly impact the quality of EEG signals and the accuracy of the information derived from them. Noise in EEG encompasses various sources, including inherent noise in electronic devices, with thermal noise being a fundamental limitation that can obscure the clarity of EEG recordings. This study focuses on the critical task of denoising EEG signals to enhance their quality and facilitate more accurate analysis. It explores the application of denoising techniques, specifically Maximum Overlap Discrete Wavelet Packet Transform (MODWPT), to address the challenges posed by noise in EEG data. One of the limitations is the node selection. In order to improve this limitation, we propose a new Bayesian MODWPT. For greater accuracy, we utilize simulation signals and EEG datasets. The results conclusively illustrate the effectiveness of this approach in accurately extracting vital signal feature information while effectively filtering out unwanted noise components. This enhances the quality and reliability of EEG signal analysis in neurological research and clinical applications. This method represents a promising advancement in the field, offering improved signal fidelity and paving the way for more accurate and robust insights into brain activity and neurological conditions.