Scholarlink

Healthcare data management in mobile applications requires robust privacy-preserving mechanisms, particularly under evolving regulations like the General Data Protection Regulation (GDPR). This paper presents a novel approach combining AES-256-CBC symmetric encryption and differential privacy techniques to protect sensitive clinical data while enabling meaningful analytics. We developed and evaluated a dual-interface mobile application serving both patients and healthcare providers, processing real-world clinical data from a multi-specialty medical center. Our architecture implements three key innovations: (1) a hybrid encryption system utilizing AES-256-CBC with HMAC for data confidentiality and integrity verification, (2) a native differential privacy framework that adds controlled noise to statistical outputs while maintaining analytical utility, and (3) a role-based access control system that enforces strict data handling policies. The system allows patients to securely manage appointments and access records, while providing healthcare providers with privacy-preserved analytics through an intuitive dashboard. Experimental results demonstrate the effectiveness of our approach, with encryption operations averaging 150 ms for sensitive data protection and query response times under 200 ms even with privacy preservation enabled. User evaluation (n \(=\) 50) showed high satisfaction scores (4.5/5.0 for patients, 4.3/5.0 for providers) despite the added security overhead. Performance analysis revealed that our hybrid approach achieves a 40% improvement in data utility compared to traditional differential privacy implementations while maintaining epsilon-differential privacy guarantees ( \(\epsilon = 1.0\) ). This research contributes to the growing field of privacy-preserving analytics by demonstrating that robust privacy protections can coexist with usable healthcare applications. Our findings provide practical insights for implementing GDPR-compliant mobile health solutions that balance data utility with patient privacy.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Privacy-Preserving Healthcare Analytics: A Hybrid Approach Using AES-256-CBC Encryption and Differential Privacy in Mobile Applications

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Patrique Santos,
  • Luis Cunha,
  • Adelino Soares,
  • Paulo Váz,
  • José Silva,
  • Pedro Martins,
  • Maryam Abbasi

摘要

Healthcare data management in mobile applications requires robust privacy-preserving mechanisms, particularly under evolving regulations like the General Data Protection Regulation (GDPR). This paper presents a novel approach combining AES-256-CBC symmetric encryption and differential privacy techniques to protect sensitive clinical data while enabling meaningful analytics. We developed and evaluated a dual-interface mobile application serving both patients and healthcare providers, processing real-world clinical data from a multi-specialty medical center. Our architecture implements three key innovations: (1) a hybrid encryption system utilizing AES-256-CBC with HMAC for data confidentiality and integrity verification, (2) a native differential privacy framework that adds controlled noise to statistical outputs while maintaining analytical utility, and (3) a role-based access control system that enforces strict data handling policies. The system allows patients to securely manage appointments and access records, while providing healthcare providers with privacy-preserved analytics through an intuitive dashboard. Experimental results demonstrate the effectiveness of our approach, with encryption operations averaging 150 ms for sensitive data protection and query response times under 200 ms even with privacy preservation enabled. User evaluation (n \(=\) 50) showed high satisfaction scores (4.5/5.0 for patients, 4.3/5.0 for providers) despite the added security overhead. Performance analysis revealed that our hybrid approach achieves a 40% improvement in data utility compared to traditional differential privacy implementations while maintaining epsilon-differential privacy guarantees ( \(\epsilon = 1.0\) ). This research contributes to the growing field of privacy-preserving analytics by demonstrating that robust privacy protections can coexist with usable healthcare applications. Our findings provide practical insights for implementing GDPR-compliant mobile health solutions that balance data utility with patient privacy.