DP2AS—Definitive Privacy-Preserving Analytical Scheme for Healthcare Data Processing

Abstract

Smart healthcare systems require secure and robust data computations for providing uninterrupted monitoring, recommendation, and assistance. Wearable sensor (WS) data sources serve as the prime aggregator for data handling. Considering the security demands in sensitive healthcare data, this article introduces a Definitive Privacy-Preserving Analytical Scheme (DP2AS). The proposed scheme exploits the data classification feature based on false positives and replication. The suggested method detects redundant data in healthcare by comparing open and secure aggregation scenarios. Classifying data features as either continuous or replicating helps prevent fraudulent data insertion. By employing tree classifiers, the data attributes are accounted for in different WS aggregation intervals preventing replications. The computations are independent of false data and application-specific computations, retaining the WS privacy. In this analysis process, the error-free/ false positive fewer data chunks are concealed with user adaptable security mechanism for preventing data poisonings. The analytical model considers the previous data state with the current processing data for avoiding erroneous interruptions. The state classffier’s maximum replication mitigation provides application-specific data transfers with fast computation possibility. The proposed scheme’s performance is analyzed using the metrics false rate, data utilization, and analysis time.