Working Memory Performance Classification in Children Using Electroencephalogram (EEG) and VGGNet

• Published in: TEM Journal
• Main Author: Abidin N.A.Z.; Yassin A.I.M.; Mansor W.; Jahidin A.H.; Azhan M.N.M.; Ali M.S.A.M.
• Format: Article
• Language: English
• Published: UIKTEN - Association for Information Communication Technology Education and Science 2024
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85210921359&doi=10.18421%2fTEM134-05&partnerID=40&md5=b5b3dab64bd005425c136d759167ffef

This study investigates the relationship between EEG and different levels of working memory performance in children. A total of two hundred thirty subjects have volunteered for the study. Initially, the students are required to answer psychometric tests to gauge their working memory performance. Based on the scores obtained, the students are then segregated in high, medium, and low working memory performance groups. Resting EEG is recorded from prefrontal cortex and pre-processed for noise removal. Synthetic EEG is then generated to balance out and enhance the number of samples to two hundred for every control group. Next, short-time Fourier transform is applied to convert the signal to spectrogram. The feature image is used to train the VGGNet model. The deep learning model has been successfully developed with 100% accuracy for training, and 85.8% accuracy for validation. These indicate the potential of assessing working memory performance alternatively using EEG and VGGNet model. © 2024 Nabila Ameera Zainal Abidin et al; published by UIKTEN. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 License. The article is published with Open Access at https://www.temjournal.com/