A NOVEL PRIVACY PRESERVATION OF HEALTHCARE DATA WITH INFORMATION ENTROPY-BASED MULTI-SCHEME FULLY HOMOMORPHIC ENCRYPTION AND RIVEST SHAMIR ADLEMAN TECHNIQUES

Problem background: The biggest problem in the medical field is data security and privacy. Enhanced protection of data is required as a result of the implementation of increasing laws and regulations, the utilization of electronic patient records, the reorganization of providers, and the growing necessitate for data across patients, taxpayers, and physicians. Several standard methods of anonymization are more effective in sanitizing data, yet they are unsuccessful in restoring the data. Although a few privacy preservation algorithms have been established in recent days, the accuracy of maintaining both private and insensitive information seems to be relatively low. Objective: The aim is to develop a novel system for privacy preservation. Here, the encryption process is done using the Information Entropy-based Adaptive Encryption Technique (IEAET). The health data are gathered from appropriate online sources. The sensitive and insensitive attributes are distinguished with the help of Information Entropy. From these data, the sensitive attributes are analyzed by the information entropy optimally selected using the Enhanced Walrus Optimization Algorithm (EWaOA). Then the sensitive attributes are encrypted using the Multi-scheme Fully Homomorphic Encryption (Multi-scheme FHE) approach. Here, the key required for encryption is optimally generated using the proposed EWaOA, and this key is utilized for the encryption of sensitive data with high integrity. Finally, the insensitive attributes are enciphered using the Rivest–Shamir–Adleman (RSA) algorithm. Therefore, the health data are preserved with more security without any information loss. Result: The outcome of the proposed method provided the key variation of restoration efficiency is 98.5%. Discussion: The developed model provides better restoration efficiency by using an optimization algorithm to optimally select the attributes. It provides more effective outcomes and security, and is well performed in healthcare data preservation to enhance the privacy of data. The developed model is used for reducing the risk and saving the cost. Thus, it proved that the developed model significantly outperformed conventional methods and its reliability was also improved.