Scholarlink

<p>Stress detection is crucial because it shows how the brain responds to stimuli during the conceptual assignment of neuronal activity. Stress accomplished while doing mental arithmetic tasks (MAT) can be determined using an electroencephalogram (EEG). In this work, the publicly available MAT EEG dataset was used. The first stage involve decomposing the EEG signal into intrinsic mode functions (IMFs) using variational mode decomposition (VMD). In the second stage, entropy-based features were valuate from the IMFs. Next, the extracted features were fed into Henry gas solubility optimization (HGSO) for optimal selection of features. In the third phase, machine learning (ML) classifiers, namely, decision tree (DT), logistic regression (LR), K-nearest neighbor (KNN), support vector machine (SVM), ensemble learner classifier (EL), and its variants, were employed to classify the features selected by HGSO. The highest classification accuracy (AC %), sensitivity (SE %), and specificity (SP %) were achieved, with 95.1%, 96.1%, and 97.2% respectively. When compared to previous state-of-the-art techniques, the proposed technique (VMD+HGSO+EL) has proven to be more successful for stress detection.</p>

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

Stress detection via EEG signals using Henry gas solubility optimization feature selection

  • 期刊论文

  • 发表日期：

    2025年11月11日

查看全文
  • Jammisetty Yedukondalu,
  • G. Sai Sravanthi,
  • K. Srinivasa Rao,
  • Y. Murali Krishna,
  • M. Jaya,
  • Kalyani Sunkara

摘要

Stress detection is crucial because it shows how the brain responds to stimuli during the conceptual assignment of neuronal activity. Stress accomplished while doing mental arithmetic tasks (MAT) can be determined using an electroencephalogram (EEG). In this work, the publicly available MAT EEG dataset was used. The first stage involve decomposing the EEG signal into intrinsic mode functions (IMFs) using variational mode decomposition (VMD). In the second stage, entropy-based features were valuate from the IMFs. Next, the extracted features were fed into Henry gas solubility optimization (HGSO) for optimal selection of features. In the third phase, machine learning (ML) classifiers, namely, decision tree (DT), logistic regression (LR), K-nearest neighbor (KNN), support vector machine (SVM), ensemble learner classifier (EL), and its variants, were employed to classify the features selected by HGSO. The highest classification accuracy (AC %), sensitivity (SE %), and specificity (SP %) were achieved, with 95.1%, 96.1%, and 97.2% respectively. When compared to previous state-of-the-art techniques, the proposed technique (VMD+HGSO+EL) has proven to be more successful for stress detection.