KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Settings

Subject: search.filters.subject.Brain-Computer Interface (BCI)

Search Results

Now showing 1 - 4 of 4
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    PSD-EEGRepNet: A CNN Architecture with Multi-Branch RepBlocks for Power Spectral Density-Based Motor Imagery EEG Classification in BCI
    (2025-01-01)
    Thangthong, Kantham
    ;
    Asadi, Fawad
    ;
    Tungjitkusolmun, Supan
    Motor Imagery (MI) based Brain-Computer Interfaces (BCIs) utilizing electroencephalography (EEG) offer significant potential, yet progress can be hindered by the computational demands of deep learning classifiers. This study introduces and evaluates a novel lightweight, multi-branch Convolutional Neural Network (CNN), inspired by efficient design principles, specifically for classifying MI tasks from Power Spectral Density (PSD) EEG features. Our objective was to achieve strong classification performance coupled with favorable development phase computational characteristics. Evaluated on 10 subjects from a public PhysioNet dataset using 5-fold cross-validation and two data overlap conditions (80%, 90%), the proposed model (~6.32M parameters, ~20 MFLOPs) demonstrated high mean classification accuracies (80.38% for 80% overlap, 85.69% for 90% overlap) and efficient training times (avg. 8.8s and 20.4s per fold, respectively). While performance scaled positively with data augmentation, inter-subject variability was noted. We conclude that the proposed architecture effectively balances high accuracy with significant offline computational efficiency offering a valuable tool for BCI research and a promising foundation for developing practical MI-BCI systems.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Common Spatial Pattern Variants for Feature Extraction in Multi-Class Motor Imagery BCI
    (2025-01-01)
    Asadi, Fawad
    ;
    Thangthong, Kantham
    ;
    Tungjitkusolmun, Supan
    Brain-Computer Interfaces (BCIs) based on Motor Imagery (MI) offer intuitive control but face challenges due to noisy electroencephalography (EEG) signals that often contain artifacts, necessitating effective feature extraction. Common Spatial Pattern (CSP) is a standard technique, yet its limitations in multi-class MI scenarios motivate various extensions. This paper investigates the classification performance, computational cost, and the crucial performance versus cost trade-off inherent in practical BCI applications of five prominent CSP variants, including standard CSP, Regularized CSP (RCSP), Filter Bank CSP (FBCSP), Sparse CSP, and Kernel CSP, when applied to multi-class MI classification. Using the PhysioNet MI dataset, we employed a consistent preprocessing pipeline (18 sensorimotor channels, filtering, epoching) and evaluated each variant by feeding extracted features into a Support Vector Machine (SVM) classifier with an RBF kernel. The evaluation revealed that standard CSP and RCSP achieved the highest peak average accuracy (~97.8%) with near-efficient computational cost. FBCSP yielded high accuracy (~96.7%) but incurred the highest computational cost. Sparse CSP was computationally efficient but achieved lower peak accuracy (~95.3%), while Kernel CSP showed the lowest peak accuracy (~92.8%) at moderate cost. The findings highlight the performance profiles and efficiency of these CSP methods for this multi-class MI task, indicating that standard CSP and RCSP offer a particularly effective combination of high accuracy and manageable computational load under the evaluated conditions.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    DETECTION OF DRIVER DROWSINESS FROM EEG SIGNALS USING WEARABLE BRAIN SENSING HEADBAND
    (2021-05-31)
    Chan, Khune Satt Nyein
    ;
    Srisurangkul, C.
    ;
    Depaiwa, N.
    ;
    Pangkreung, S.
    Driver drowsiness detection plays an important role in the field of road safety and advanced driver assistance system. Electroencephalogram (EEG) signals are one of the most accurate and reliable indicators of fatigue and drowsiness but in the case of detecting drowsiness, its medical graded measuring system can be intrusive to the driver. The purpose of this research is to test the feasibility and usability of the consumer graded EEG sensor to use in a driver drowsiness detection system. The experiment was carried out by using MUSE S brain sensing headband. Fast Fourier Transform (FFT) method was used to extract features from EEG signals. The extracted feature data are then used to build two classification model, the Support Vector Machine (SVM) and Artificial Neural Network (ANN). The detection of drowsiness is the binary classification task which is to classify between drowsy epochs and alert epochs. In the case of detecting only drowsy epochs, the SVM model detected 82.7% of the drowsy epochs which was better than the ANN model which can only detect 81.25% of the drowsy epochs. But in the detection of both drowsy and alert epochs, the ANN model performed better than that of SVM. The SVM model was tested with different kernel function and Fine Gaussian SVM model showed the highest accuracy of 87.8%. The ANN model performed slightly higher than the SVM model with an accuracy of 87.9%. The ability of consumer graded EEG sensor to use in drowsiness detection system was validated in this research.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Time-frequency analysis of red-green visual flickers based on steady-state visual evoked potential recording
    (2017-02-21)
    Dechwechprasit, Panisa
    ;
    Phothisonothai, Montri
    ;
    Tantisatirapong, Suchada
    The study of brain activity can be done by visual stimulus flickering at specific frequencies, Steady-State Visual Evoked Potential or as known as SSVEP. SSVEP is to stimulate the EEG signal to locate the desired goal of the experiment when a visual stimulus flickering with different constant frequencies and same duration. We aim to present a case study of electroencephalogram (EEG) signal by analyzing the frequency response of the red and green light. The stimulation is based on SSVEP by dividing the trial into two trials: single light and two lights. We considered three parameters that are light color, frequency and epoch interval. The optimal experimental results showed the classification accuracy rate of 74% and 75% for single and two color lights, respectively. The results can be considerably applied to the brain-computer interface (BCI) system.