Scholarlink

<p>Healthcare is increasingly utilising Industrial Internet of Things (IIoT) technologies. However, it also raises important issues regarding security, dependability, and privacy. Several techniques fall short in preventing manipulation and cyberattacks on sensitive data. We present MedShield, a threat detection framework with a privacy focus that integrates edge computing, blockchain, and machine learning to address these issues. MedShield employs IPFS and blockchain technology to safely store and exchange healthcare models and data, making them impenetrable. Additionally, a deep autoencoder is used to conceal private network data. Three layers make up the framework’s architecture: IIoT devices, edge, and the cloud. For quicker processing, machine learning tasks are managed at the edge, while the blockchain is guaranteed via the cloud for safe data storage. Additionally, MedShield facilitates distributed model training across edge devices managed by a central node, thereby enhancing data privacy protection. Trials using open IIoT Datasets demonstrate that MedShield preserves user privacy, detects threats with high accuracy, and operates effectively at scale. Because of this, it is a robust and dependable option for protecting contemporary smart healthcare systems.</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

Medshield a privacy enhanced threat detection framework for secure industrial IoT healthcare

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月27日

查看全文
  • Dileep Kumar Murala,
  • K. Madhura,
  • Gopi Adapa,
  • Yadaiah Balagoni,
  • Bananeza Romeo

摘要

Healthcare is increasingly utilising Industrial Internet of Things (IIoT) technologies. However, it also raises important issues regarding security, dependability, and privacy. Several techniques fall short in preventing manipulation and cyberattacks on sensitive data. We present MedShield, a threat detection framework with a privacy focus that integrates edge computing, blockchain, and machine learning to address these issues. MedShield employs IPFS and blockchain technology to safely store and exchange healthcare models and data, making them impenetrable. Additionally, a deep autoencoder is used to conceal private network data. Three layers make up the framework’s architecture: IIoT devices, edge, and the cloud. For quicker processing, machine learning tasks are managed at the edge, while the blockchain is guaranteed via the cloud for safe data storage. Additionally, MedShield facilitates distributed model training across edge devices managed by a central node, thereby enhancing data privacy protection. Trials using open IIoT Datasets demonstrate that MedShield preserves user privacy, detects threats with high accuracy, and operates effectively at scale. Because of this, it is a robust and dependable option for protecting contemporary smart healthcare systems.