Novel Edge Computing-Based Privacy-Preserving Approach for Smart Healthcare Systems in the Internet of Medical Things

This paper presents the Edge-Based Privacy Preserving Approach (EBPPA) for healthcare applications that store private user data on cloud servers, and perform computation operations for patient diagnoses. The increasing cyber-attacks on hospital systems and the exposure of mathematical operations on cloud-stored data to untrusted entities pose significant data privacy and security risks. To address these challenges, our proposed approach incorporates two key components such as Homomorphic Encryption and the XGBoost algorithm. Homomorphic Encryption protects medical plaintext data, by enabling computations on encrypted data without decryption. This ensures that sensitive patient data remains secure even if attackers access it. Additionally, we employ Secret Sharing, which distributes computations to multiple virtual nodes located on edge. By doing so, all arithmetic operations are masked, preventing untrusted cloud servers from gaining knowledge about the specific tasks performed on the encrypted patient data. This enhances the confidentiality of the data and safeguards against potential privacy breaches. The virtual edge nodes (VENs) in our approach leverage the computational resources of the cloud to handle computationally intensive mathematical functions efficiently. This enables accurate and efficient healthcare data processing and reduces latency in data transmission between devices and edge nodes. To validate the effectiveness of our approach, we conducted a comparative analysis with existing studies. The results demonstrate that storing homomorphically encrypted data at the edge preserves data privacy and integrity while ensuring the confidentiality of the data through secret sharing-based multi-node computation. This approach offers a robust solution for protecting sensitive patient data from unauthorized access and maintaining data confidentiality in untrusted cloud networks. Our proposed Edge Based Privacy Preserving Approach (EBPPA) combines Homomorphic Encryption, which secures medical data, and the XGBoost algorithm, which enables accurate computations, to address healthcare applications' challenges. By leveraging secret sharing and virtual edge nodes, our approach provides enhanced data privacy, integrity, and confidentiality, making it a promising solution for securing sensitive patient data in the healthcare domain.