Blockchain-based Medical Data Preservation via Improved Association Rule Hiding with Optimal Key Generation

Abstract

Remote health data monitoring to achieve intelligent healthcare has recently drawn a lot of interest due to the Internet of Things’ (IoT) substantially increased implementation. However, due to the constrained processing and storage capabilities of IoT devices, users’ health data are typically stored in a centralized third party, such as a hospital database or cloud, and this causes users to lose control of their health data, which is easily the cause of privacy leakage and a single-point bottleneck. A medical data transmission and preservation strategy is proposed which is based on the hospital’s private block chain in order to enhance the electronic health system. The three major phases of the privacy-preserving medical data strategy initiated in this study are “data sanitization, optimal key generation, and data restoration”. Prior to being added to the block–blocks chains, the medical record is cleaned up. The sanitization phase will employ the improved association rule concealment technique. The cleaned content is recovered at the receiver’s end. More significantly, both methods heavily rely on optimal key creation, with the ideal key being selected using a new hybrid optimization model called Dragonfly-Updated Elephant Herding Optimization (DUEHO). Elephant Herding Optimization (EHO) and the traditional Dragonfly Algorithm (DA) are conceptually combined in the proposed DUEHO. At the end, proposed model’s performance is compared over existing techniques concerning various metrics. The convergence rate of the suggested model is 1.05%, 0.31%, 0.32%, 0.4%, 0.43%, 0.45% better than the existing models like FF, MFO, GWO, WOA, DA and EHO, respectively.