Scholarlink

<p>Mental stress experienced during oral examinations can significantly impact the cognitive performance of university students. This study presents a smart bra integrated with electrocardiogram (ECG) and surface electromyography (sEMG) sensors for real-time monitoring of physiological stress responses. Eighty undergraduate students (18 males and 62 females) participated in oral exams structured according to Bloom’s Taxonomy to facilitate a controlled increase in cognitive load. The Perceived Stress Scale (PSS) and Depression Anxiety Stress Scale (DASS) were employed to establish baselines and track subjective responses during the exams. Results indicated that as exam difficulty increased, student performance significantly declined (p &lt; 0.0001), accompanied by elevated stress scores and notable physiological changes, including increased heart rate and muscle tension. Gender-specific responses were observed; males exhibited elevated muscle tension, while females demonstrated increased heart rate measures. Features extracted from ECG and sEMG signals were analyzed to classify stress levels using advanced machine learning approaches, including RoBoSS-SVM, Wave-SVM, and AdaBoost models, achieving 95.8% validation accuracy and 91.1% independent testing accuracy. This study advances stress assessment by incorporating a multi-level classification system beyond binary detection and introducing a novel stress induction protocol based on Bloom’s Taxonomy. The findings highlight the potential of integrating wearable e-textiles with machine learning for personalized stress interventions, paving the way for improved student well-being and adaptive learning environments.</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

Smart bra-based ECG and sEMG system for physiological stress detection during oral exams via machine learning classification

  • 期刊论文

  • 发表日期：

    2026年2月14日

查看全文
  • Mohamed Naeem,
  • Sahar Ali Fawzi,
  • Hamed Anwar,
  • Abdel Salam Malek

摘要

Mental stress experienced during oral examinations can significantly impact the cognitive performance of university students. This study presents a smart bra integrated with electrocardiogram (ECG) and surface electromyography (sEMG) sensors for real-time monitoring of physiological stress responses. Eighty undergraduate students (18 males and 62 females) participated in oral exams structured according to Bloom’s Taxonomy to facilitate a controlled increase in cognitive load. The Perceived Stress Scale (PSS) and Depression Anxiety Stress Scale (DASS) were employed to establish baselines and track subjective responses during the exams. Results indicated that as exam difficulty increased, student performance significantly declined (p < 0.0001), accompanied by elevated stress scores and notable physiological changes, including increased heart rate and muscle tension. Gender-specific responses were observed; males exhibited elevated muscle tension, while females demonstrated increased heart rate measures. Features extracted from ECG and sEMG signals were analyzed to classify stress levels using advanced machine learning approaches, including RoBoSS-SVM, Wave-SVM, and AdaBoost models, achieving 95.8% validation accuracy and 91.1% independent testing accuracy. This study advances stress assessment by incorporating a multi-level classification system beyond binary detection and introducing a novel stress induction protocol based on Bloom’s Taxonomy. The findings highlight the potential of integrating wearable e-textiles with machine learning for personalized stress interventions, paving the way for improved student well-being and adaptive learning environments.