Scholarlink

This research explores the revolutionary power of IoT in healthcare based on two integrated case studies, i.e., real-time clinical monitoring with multi-sensor IoT sensors (oxygen, pressure, temperature monitoring) based on machine learning (ANN, DT, SVM) for predictive diagnosis and remote medication/fitness management platforms to counter problems of accessibility and interface, with appreciable improvement in health outcomes among a trial of 30 patients. At a university, wearable student data (heart rate, activity, calories, steps) was used to create an XGBoost model with 93% activity recognition accuracy, utilizing federated learning to ensure privacy protection. Our system offers three innovations: real-time, actionable health insights through edge computing, energy-efficient wearable functionality, and scalable solutions for both clinical and consumer use. Future research involves an extension of CNN models to multi-level activity detection, elderly/chronic disease, and edge-based early warning system design. The paper suggests a preventive health model in IoT-enabled, patient-centered preventive health through technological convergence and data-driven care.

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

Multi-sensor Wearables with Machine Learning for Preventive Healthcare and Fitness Tracking

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Vinod Kumar,
  • Chander Prabha,
  • Mohammad Zubair Khan,
  • Ibrahim Aljubayri

摘要

This research explores the revolutionary power of IoT in healthcare based on two integrated case studies, i.e., real-time clinical monitoring with multi-sensor IoT sensors (oxygen, pressure, temperature monitoring) based on machine learning (ANN, DT, SVM) for predictive diagnosis and remote medication/fitness management platforms to counter problems of accessibility and interface, with appreciable improvement in health outcomes among a trial of 30 patients. At a university, wearable student data (heart rate, activity, calories, steps) was used to create an XGBoost model with 93% activity recognition accuracy, utilizing federated learning to ensure privacy protection. Our system offers three innovations: real-time, actionable health insights through edge computing, energy-efficient wearable functionality, and scalable solutions for both clinical and consumer use. Future research involves an extension of CNN models to multi-level activity detection, elderly/chronic disease, and edge-based early warning system design. The paper suggests a preventive health model in IoT-enabled, patient-centered preventive health through technological convergence and data-driven care.