Scholarlink

Electroencephalography (EEG) serves as a crucial tool for analyzing cognitive ability, providing insights about brain activity associated with various mental states. The present study investigates cognitive performance in both resting and active states among sports and non-sports individuals through real-time EEG signals collected from frontal and parietal electrodes (F3, F4, P3, and P4). A rigorous preprocessing pipeline, including filtering techniques and Independent Component Analysis (ICA), was employed to enhance data quality by reducing noise and artifacts. Utilizing transfer learning approaches, four pre-trained models such as VGG16, VGG19, INCEPTIONV3, and CNN were compared to identify cognitive ability differences. Notably, the CNN model achieved an accuracy of 98.6%, demonstrating the power of transfer learning techniques in accurately classifying cognitive ability. The study findings reveal significant differences in cognitive performance between sports and non-sports individuals, with the sports group exhibiting enhanced cognitive function due to regular physical activities.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Transfer Learning Framework for Cognitive Ability Classification Using Electroencephalography

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • N. Keerthika,
  • V. Kiruthika,
  • Sukriti,
  • Vasundhara Challagundla,
  • Sai Rithika Nandyala,
  • Muppuri Saranya

摘要

Electroencephalography (EEG) serves as a crucial tool for analyzing cognitive ability, providing insights about brain activity associated with various mental states. The present study investigates cognitive performance in both resting and active states among sports and non-sports individuals through real-time EEG signals collected from frontal and parietal electrodes (F3, F4, P3, and P4). A rigorous preprocessing pipeline, including filtering techniques and Independent Component Analysis (ICA), was employed to enhance data quality by reducing noise and artifacts. Utilizing transfer learning approaches, four pre-trained models such as VGG16, VGG19, INCEPTIONV3, and CNN were compared to identify cognitive ability differences. Notably, the CNN model achieved an accuracy of 98.6%, demonstrating the power of transfer learning techniques in accurately classifying cognitive ability. The study findings reveal significant differences in cognitive performance between sports and non-sports individuals, with the sports group exhibiting enhanced cognitive function due to regular physical activities.