Scholarlink

<p>The integration of artificial intelligence (AI) with smart textiles has enabled the development of self-learning wearables for personalized health monitoring. This study presents a comprehensive framework for AI-driven adaptive textiles capable of real-time physiological monitoring and autonomous adjustments based on biometric data. The system incorporates flexible biosensors, edge AI processing, and wireless communication to continuously assess user health metrics. A dataset from 30 participants was collected under varying activity conditions (resting, walking, exercising), revealing high accuracy of smart textile sensors when compared to conventional medical devices. Statistical analysis demonstrated strong correlations between smart textile sensor readings and medical-grade devices, with Pearson correlation coefficients of 0.96 for heart rate, 0.94 for body temperature, and 0.99 for hydration level. Mean Absolute Error (MAE) values before calibration were 1.90&#xa0;bpm for heart rate, 0.11&#xa0;°C for body temperature, and 0.72% for hydration level. Post-adjustment of sensor algorithms using Kalman filtering, moving average smoothing, and adaptive thresholding, the system achieved error reductions of up to 25%, significantly improving real-time performance. These findings highlight the viability of self-learning smart textiles as an effective tool for continuous, non-invasive health monitoring. The integration of AI-driven adjustments enhances accuracy and reliability, making these wearables a promising solution for personalized healthcare applications. Further research will focus on optimizing energy efficiency and expanding the system’s predictive capabilities using advanced AI techniques.</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

Self-Learning Smart Textiles: AI-Driven Adaptive Wearables for Personalized Health Monitoring

  • 期刊论文

  • 发表日期：

    2025年10月15日

查看全文
  • Md. Eanamul Haque Nizam,
  • Ayub Nabi Khan,
  • A. N. M. Ahmed Ullah,
  • Nujhat Kawsari

摘要

The integration of artificial intelligence (AI) with smart textiles has enabled the development of self-learning wearables for personalized health monitoring. This study presents a comprehensive framework for AI-driven adaptive textiles capable of real-time physiological monitoring and autonomous adjustments based on biometric data. The system incorporates flexible biosensors, edge AI processing, and wireless communication to continuously assess user health metrics. A dataset from 30 participants was collected under varying activity conditions (resting, walking, exercising), revealing high accuracy of smart textile sensors when compared to conventional medical devices. Statistical analysis demonstrated strong correlations between smart textile sensor readings and medical-grade devices, with Pearson correlation coefficients of 0.96 for heart rate, 0.94 for body temperature, and 0.99 for hydration level. Mean Absolute Error (MAE) values before calibration were 1.90 bpm for heart rate, 0.11 °C for body temperature, and 0.72% for hydration level. Post-adjustment of sensor algorithms using Kalman filtering, moving average smoothing, and adaptive thresholding, the system achieved error reductions of up to 25%, significantly improving real-time performance. These findings highlight the viability of self-learning smart textiles as an effective tool for continuous, non-invasive health monitoring. The integration of AI-driven adjustments enhances accuracy and reliability, making these wearables a promising solution for personalized healthcare applications. Further research will focus on optimizing energy efficiency and expanding the system’s predictive capabilities using advanced AI techniques.