Scholarlink

<p>Disease identification services provided by the healthcare cloud in the “mobile healthcare network” are inaccessible due to security issues and a lack of resources on “internet of things” nodes. Secure medical services are facilitated by the emergence of blockchain and artificial intelligence technologies. However, there is a chance for data hacking and privacy breaches. Therefore, an efficient privacy preservation system is developed in this work using the cryptographic algorithm to safeguard the confidentiality of the data and secure the private information of patients. It also used the deep learning model to predict the patient’s disease. Initially, the multimodal data including images, signals, or data are gathered from standard sources. Then, the collected multimodal data are converted into 2D images. Better visualization is achieved by this 2D image conversion. The converted 2D images are encrypted on the transmitter side through an adaptive 2D logistic chaotic model. This technique is capable of generating secret keys to improve the safety of data. The same approach is used for performing decryption at the receiver side. An optimal key is generated during the encryption to safeguard the data using a modified red-billed blue magpie optimizer. Finally, the disease is predicted using the attention-based residual DenseNet with a gated recurrent unit. This technique consumes much less amount of memory and provides fast execution. The experimental evaluation is carried out to demonstrate the effectiveness of the recommended model over traditional models.</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

Ensuring the privacy of data in modern healthcare systems using adaptive encryption technique and disease prediction via attention-based residual DenseNet with GRU

  • 期刊论文

  • 发表日期：

    2025年6月5日

查看全文
  • Pankaj Kumar Sharma,
  • Hemant Kumar Vijayvergia,
  • Amit Garg,
  • Varun Prakash Saxena,
  • Shyam Sundar Agrawal,
  • Meeta Sharma

摘要

Disease identification services provided by the healthcare cloud in the “mobile healthcare network” are inaccessible due to security issues and a lack of resources on “internet of things” nodes. Secure medical services are facilitated by the emergence of blockchain and artificial intelligence technologies. However, there is a chance for data hacking and privacy breaches. Therefore, an efficient privacy preservation system is developed in this work using the cryptographic algorithm to safeguard the confidentiality of the data and secure the private information of patients. It also used the deep learning model to predict the patient’s disease. Initially, the multimodal data including images, signals, or data are gathered from standard sources. Then, the collected multimodal data are converted into 2D images. Better visualization is achieved by this 2D image conversion. The converted 2D images are encrypted on the transmitter side through an adaptive 2D logistic chaotic model. This technique is capable of generating secret keys to improve the safety of data. The same approach is used for performing decryption at the receiver side. An optimal key is generated during the encryption to safeguard the data using a modified red-billed blue magpie optimizer. Finally, the disease is predicted using the attention-based residual DenseNet with a gated recurrent unit. This technique consumes much less amount of memory and provides fast execution. The experimental evaluation is carried out to demonstrate the effectiveness of the recommended model over traditional models.