Anonymized Security Framework for Safeguarding and Distributing Clinical Data

Advancements in Industry 4.0 have significantly improved healthcare through enhanced treatments, communication, remote monitoring, and cost reduction. However, sharing sensitive healthcare data is challenging due to privacy and security concerns. This study presents an attribute-focused privacy-preserving data publishing approach combining fixed-interval methods for numerical attributes and enhanced l-diverse slicing for categorical data. Horizontal and vertical data partitioning ensures privacy without compromising utility. Experiments with real-world datasets show a 13% improvement in classification accuracy and a 12% reduction in information loss compared to existing methods, protecting against identity and attribute disclosures.