Scholarlink

<p>The Internet of Medical Things (IoMT) consists of interconnected devices such as wearable sensors, medical monitors, and diagnostic tools that collect and transmit real-time health data. These networks play a crucial role in modern healthcare by enabling continuous patient monitoring and data-driven decision-making. However, the increasing reliance on medical IoT devices exposes them to various cybersecurity threats<b>,</b> such as data breaches, unauthorized access, and denial-of-service attacks. The sensitive nature of healthcare data makes these networks attractive targets for malicious actors, which compromise patient safety and privacy. Existing approaches to attack detection in medical IoT networks often struggle to provide accurate and timely detection of cyber-attacks due to the uncertainty and dynamic nature of IoT data. This paper addresses these limitations by introducing a novel Fuzzy Adaptive Support Vector Machine (F-ASVM) that combines fuzzy logic with adaptive SVM to detect attacks accurately. The integration of fuzzy logic allows the model to handle uncertainty in sensor data, providing a more accurate classification of normal behavior and potential attacks. Meanwhile, the adaptive SVM ensures that the model adjusts to new attack patterns over time, making it adaptable to emerging threats. The study also incorporates the Seagull Optimization Algorithm (SOA) with a local strategy to fine-tune the hyperparameters of the SVM, improving the model’s performance and ensuring efficient convergence. The effectiveness of the proposed approach is evaluated using a diabetic patient dataset<b>,</b> with simulation results showing its ability to detect attacks with high accuracy, enhancing the security and reliability of critical medical IoT infrastructure.</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

Evolving security measures for IoT medical data in cloud environments

  • 期刊论文

  • 发表日期：

    2025年6月28日

查看全文
  • C. Dhaya,
  • G. Niranjana,
  • B. Prakash

摘要

The Internet of Medical Things (IoMT) consists of interconnected devices such as wearable sensors, medical monitors, and diagnostic tools that collect and transmit real-time health data. These networks play a crucial role in modern healthcare by enabling continuous patient monitoring and data-driven decision-making. However, the increasing reliance on medical IoT devices exposes them to various cybersecurity threats, such as data breaches, unauthorized access, and denial-of-service attacks. The sensitive nature of healthcare data makes these networks attractive targets for malicious actors, which compromise patient safety and privacy. Existing approaches to attack detection in medical IoT networks often struggle to provide accurate and timely detection of cyber-attacks due to the uncertainty and dynamic nature of IoT data. This paper addresses these limitations by introducing a novel Fuzzy Adaptive Support Vector Machine (F-ASVM) that combines fuzzy logic with adaptive SVM to detect attacks accurately. The integration of fuzzy logic allows the model to handle uncertainty in sensor data, providing a more accurate classification of normal behavior and potential attacks. Meanwhile, the adaptive SVM ensures that the model adjusts to new attack patterns over time, making it adaptable to emerging threats. The study also incorporates the Seagull Optimization Algorithm (SOA) with a local strategy to fine-tune the hyperparameters of the SVM, improving the model’s performance and ensuring efficient convergence. The effectiveness of the proposed approach is evaluated using a diabetic patient dataset, with simulation results showing its ability to detect attacks with high accuracy, enhancing the security and reliability of critical medical IoT infrastructure.