Smart Contract Based Sensitive Data Hashing for Security in Healthcare Environment

Abstract

A Big Data environment is a robust ecosystem in Healthcare data analysis to extract sensitive information from Medical Personal Healthcare Records (MPHR) to preserve privacy to protect sensitive information. From the data analysis context, the sensitive and non-sensitive information is identical in behavioral approach, so those sensitive items have a similar frequency of access in all security roles, leading to privacy and security especially becoming crucial issues. Thus, the existing problems are the most non-sensitive approach to dealing with the privacy standard in the form of low performance and lead time complexity. Information regarding exposure and risk-response relationships is crucial for estimating the burden of illness caused by environmental variables. The world's efforts to promote sound preventative measures through existing policies, methods, measures, methods, and knowledge can be bolstered by a better grasp of the extent to which disease and ill health is attributable to adjustable risks related to the environment. To tackle these issues, we propose an MPHR- Sensitive data prediction system based on a Pragmatic attribute Identifier Using Advance Blockchain security to Secure the Sensitive data in the big data healthcare environment. Initially, the proposed technique pre-processes the PHR information to eliminate potential errors. Then Sensitive Scaling Impact Rate (SSIR) method is used to identify the sensitive and non-sensitive marginal values. Based on the marginal values, the proposed Pragmatic Sensitive Feature Clustering Algorithm (PSFCA) is used to analyze the importance of sensitive feature relations. Next, the sensitive relation is fed into the Densenet Convolutional Neural Network (Densnet-CNN) method to classify the sensitive and non-sensitive attributes. Further, the security principle is applied based on Smart Contract Master Aggregation (SCMA) based blockchain health care security. The proposed approach outperforms existing privacy preservation in sensitive data prediction systems in terms of blockchain privacy and security, according to the findings of the experiments.