Saidatul Ardeenawatie Awang
Thumbnail Image
Preferred name
Saidatul Ardeenawatie Awang
Official Name
Saidatul Ardeenawatie, Awang
Alternative Name
Saidatul, Ardeenawatie
Awang, S. A.
Awang, Saidatul Ardeenawatie
Awang, Saidatul Ardeenaawatie
Awang, Saidatul Ardeenawatie Binti
Awang, Saidatul Ardeenawatiebinti
Saidatul, A.
Main Affiliation
Scopus Author ID
57205231792
Researcher ID
CCJ-6771-2022

Research Output

Filters

2 1
2
Reset filters

Settings
Now showing 1 - 2 of 2
  • Publication
    EEG signal processing using deep learning for motor imagery tasks: Leveraging signal images
    (Springer, 2025)
    Saidatul Ardeenawatie Awang
    ;
    Abdul Hamid Adom
    ;
    Erdy Sulino Mohd Muslim Tan
    ;
    Husna Najeha Amran
    ;
    Arni Munira Markom
    A novel approach to processing electroencephalography (EEG) signals has emerged, leveraging the utilization of signal images. The application of deep learning techniques in bypassing complex signal and image processing tasks has generated significant interest in this field. However, challenges remain in signal image processing, particularly in handling significant features and image sizes. This study presents a comprehensive investigation of EEG motor imagery signal processing, focusing on the classification of three tasks: eating, drinking, and seeking assistance. Fast Fourier Transform (FFT) is employed to extract signal image features, which are subsequently utilized in a deep learning framework. EEG data were collected from five subjects, and four transfer functions of deep learning models, namely VGG16, VGG19, ResNet50, and ResNet101, were employed for training and classification purposes. The performance of the four models was meticulously evaluated and compared. Notably, VGG16 exhibited superior performance in accurately classifying the EEG motor imagery tasks, achieving an impressive accuracy of 90%, sensitivity of 84%, and specificity of 92%. In conclusion, this study underscores the efficacy of EEG signal image processing through deep learning-based classification techniques. The findings highlight the potential of utilizing signal images in EEG analysis for motor imagery tasks, thereby contributing to the advancement of brain-computer interface technology and enhancing our understanding of neural dynamics.
  • Publication
    Development of neurometric acute stress assessment based on EEG signals
    ( 2014)
    Saidatul Ardeenawatie Awang
    Nowadays, stress is one of the major issues where too much stress may lead to depression, fatigue and insomnia. Stress can be divided into two types called Eustress and Distress. Eustress or positive stress refers to the positive stress which helps to improve the performance of an individual. In contrast, Distress or negative stress can devastate a person by creating depression and damage the quality of life. It is essential to comprehend and to figure out the state of current stress in numerical index. The development of a reliable data acquisition protocol is a crucial part to elicit mental stress in different level of stress. In this study, some modification on the existing Mental Arithmetic Task (MAT) has been made to ensure the designed protocol is capable to induce the different intensity of stress such as low, moderate and high. The dynamical excitation protocol and time pressure concept are proposed in this work. There are three validation methods have been used, namely, K Nearest Neighbor (KNN), Alpha Brain Asymmetry and statistical analysis (Paired T-test). As a result of this study, it was found that the proposed experimental protocol is comparable as the verification has been made with the following: (i) The t-test result based on physiological changes during pre and post experiment were found to be statistically significant (p<0.01) (ii) The mean value of Alpha Brain Asymmetry are comparable and have a potential to discriminate between levels and (iii) the classification accuracy of 84% confirmed that the proposed protocol have potential in classifying the mental stress level. Besides that, the preprocessing technique applying elliptic filters with 256 data per frame is the most suitable technique. Five types of spectral estimator (Welch, Burg, Yule Walker, Modified Covariance and Multiple Signal Classification) based feature extraction is performed on the normalized signals. The extracted features are cross validated using 10-fold cross validation and classified using KNN and have been proved using statistical analysis (ANOVA). The maximum mean classification rate of 86.75% is achieved using Modified Covariance feature derived from alpha waves using KNN. Besides that, this study found that F3 and F4 are the most informative electrodes with the classification rate of 93.50%. Last but not least, a new algorithm has been proposed based on the more established index, Alpha Asymmetry Score (AAS), as a reference. Modifications have been made in term of the frequency band as a variable in the stress index. The classification accuracy of the proposed Stress Asymmetry Score (SAS) is approximately 96% which is 10% higher than AAS. The development of the stress index promises new era of stress brain related research for future people’s benefit.
      18  1