Scholarlink

<p>The integration of Internet of Medical Things (IoMT) technology is revolutionizing patient monitoring by enabling real-time and remote assessment. It utilized various health sensors and wireless technologies to communicate with patients and transmit health records to cloud systems for analysis and disease identification. This study proposes an AI-driven framework for disability detection using secured IoMT, leveraging motion analysis, efficient data routing, and secure cloud storage. The framework captures motion data through simulated IoT-enabled wearable devices, represented by open-source datasets such as publicly available PAMAP2 and MHEALTH. To identify movement patterns connected to disability and train the model, the motion data is first preprocessed using noise reduction and normalization techniques. The proposed framework utilizes a Support Vector Machine to classify the patterns due to its lightweight features, ultimately providing a rapid, real-time analysis in crucial health circumstances. It processes the extracted features and predicts whether a movement pattern indicates normal or disability related human behavior. Moreover, health records are transmitted from IoMT devices to the cloud using network optimization by exploring LPWAN/LoRaWAN protocols, ensuring energy-efficient, low-latency communication. By combining intelligent learning with optimized network protocols and secure cloud integration, the proposed framework gives a practical approach to the healthcare domain to access timely insights into patient health. The proposed framework is simulated in NS3 for performance evaluation and provides significant outcomes as compared to existing approaches in terms of energy consumption by an average of 47%, attack detection rate by an average of 35%, packet drop ratio by an average of 48% and false positive rate by an average of 42%.</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

Real-time e-health framework for efficient AI-driven disability monitoring using secured Internet of Medical Things

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月2日

查看全文
  • Ghadah Naif Alwakid,
  • Mamoona Humayun,
  • Khalid Haseeb,
  • Ayesha Shafique

摘要

The integration of Internet of Medical Things (IoMT) technology is revolutionizing patient monitoring by enabling real-time and remote assessment. It utilized various health sensors and wireless technologies to communicate with patients and transmit health records to cloud systems for analysis and disease identification. This study proposes an AI-driven framework for disability detection using secured IoMT, leveraging motion analysis, efficient data routing, and secure cloud storage. The framework captures motion data through simulated IoT-enabled wearable devices, represented by open-source datasets such as publicly available PAMAP2 and MHEALTH. To identify movement patterns connected to disability and train the model, the motion data is first preprocessed using noise reduction and normalization techniques. The proposed framework utilizes a Support Vector Machine to classify the patterns due to its lightweight features, ultimately providing a rapid, real-time analysis in crucial health circumstances. It processes the extracted features and predicts whether a movement pattern indicates normal or disability related human behavior. Moreover, health records are transmitted from IoMT devices to the cloud using network optimization by exploring LPWAN/LoRaWAN protocols, ensuring energy-efficient, low-latency communication. By combining intelligent learning with optimized network protocols and secure cloud integration, the proposed framework gives a practical approach to the healthcare domain to access timely insights into patient health. The proposed framework is simulated in NS3 for performance evaluation and provides significant outcomes as compared to existing approaches in terms of energy consumption by an average of 47%, attack detection rate by an average of 35%, packet drop ratio by an average of 48% and false positive rate by an average of 42%.