A Novel Internet of Things and Cloud Computing-Driven Deep Learning Framework for Disease Prediction and Monitoring

In smart cities, the e-healthcare systems aided by Internet of Things (IoT) technologies play a significant role in proficient health monitoring services. The sensitivity and number of users in health networks highlights the necessity of treating security attacks. In the era of rapid internet connectivity and cloud computing services, patient medical information is most sensitive, and its electronic representation poses privacy and security concerns. Moreover, it is challenging for the traditional classifier to process a massive amount of health data and classify patients' health statuses. To address this matter, this paper presents a novel healthcare model, IoT-CDLDPM, to estimate patients’ disease levels using original data and fuzzy entropy extracted from patients' remote locations. IoT-CDLDPM incorporates a deep learning classifier to analyze extensive patient-related data and provides efficient and accurate health status predictions. Furthermore, the proposed model presents the secured storage structure of the individual's health data in cloud servers. To give the authenticity of the health data, two new cryptographic algorithms are presented that encrypt and decrypt the data securely transmitted through the network. A comparison with existing methods reveals that the proposed system significantly reduces computation time, with a recorded time of 0.5 seconds, outperforming DSVS, PP-ESAP, and DRDA by up to 80%. Furthermore, the proposed cryptographic model enhances security levels, achieving a range between 99.4% and 99.8% across multiple experimental setups, surpassing other widely used encryption algorithms such as AES, RSA, and ECC-DH.

@article{GUO2025,
  title     = {A Novel Internet of Things and Cloud Computing-Driven Deep Learning Framework for Disease Prediction and Monitoring},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {16},
  number    = {1},
  year      = {2025},
  publisher = {The Science and Information Organization},
  author    = {Bo GUO and Lei NIU},
  doi       = {10.14569/IJACSA.2025.0160170},
  url       = {https://doi.org/10.14569/IJACSA.2025.0160170}
}