Differential Privacy for Secure Machine Learning in Healthcare IoT-Cloud Systems

2025-12-11

AI Summary

Addressing the challenge of simultaneously ensuring privacy preservation, real-time responsiveness, and data utility in medical IoT-Cloud systems, this paper proposes a hierarchical collaborative privacy-preserving computing framework. It establishes an IoT-Edge-Cloud three-tier architecture and introduces urgency- and persistence-aware task scheduling tailored for emergency care scenarios. The work pioneers a Laplace-Gaussian hybrid noise mechanism coupled with an adaptive privacy budget allocation strategy, significantly improving the privacy–utility trade-off for both low- and high-dimensional healthcare data. Furthermore, it integrates edge-accelerated inference with blockchain-based auditability to ensure traceable and tamper-proof analytical provenance. Experimental evaluation demonstrates that, at a practical privacy budget ε = 5.0, supervised models achieve 82–84% accuracy; attribute inference attack success rates decrease by 18%; data reconstruction correlation drops by 70%; and edge-level emergency response latency is reduced by 8×.

Technology Category

• Machine Learning: Privacy
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust
• Computer Vision: Bias, Fairness & Privacy

Application Category

• Security and Privacy: Data transparency and provenance
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems

Abstract

Healthcare has become exceptionally sophisticated, as wearables and connected medical devices are revolutionising remote patient monitoring, emergency response, medication management, diagnosis, and predictive and prescriptive analytics. Internet of Things and Cloud computing integrated systems (IoT-Cloud) facilitate sensing, automation, and processing for these healthcare applications. While real-time response is crucial for alleviating patient emergencies, protecting patient privacy is extremely important in data-driven healthcare. In this paper, we propose a multi-layer IoT, Edge and Cloud architecture to enhance the speed of response for emergency healthcare by distributing tasks based on response criticality and permanence of storage. Privacy of patient data is assured by proposing a Differential Privacy framework across several machine learning models such as K-means, Logistic Regression, Random Forest and Naive Bayes. We establish a comprehensive threat model identifying three adversary classes and evaluate Laplace, Gaussian, and hybrid noise mechanisms across varying privacy budgets, with supervised algorithms achieving up to 86% accuracy. The proposed hybrid Laplace-Gaussian noise mechanism with adaptive budget allocation provides a balanced approach, offering moderate tails and better privacy-utility trade-offs for both low and high dimension datasets. At the practical threshold of $varepsilon = 5.0$, supervised algorithms achieve 82-84% accuracy while reducing attribute inference attacks by up to 18% and data reconstruction correlation by 70%. Blockchain security further ensures trusted communication through time-stamping, traceability, and immutability for analytics applications. Edge computing demonstrates 8$ imes$ latency reduction for emergency scenarios, validating the hierarchical architecture for time-critical operations.

Problem

• Ensuring patient data privacy in IoT-Cloud healthcare systems
• Improving emergency response speed through distributed edge-cloud architecture
• Balancing privacy and utility in machine learning with differential privacy

Innovation

• Multi-layer IoT-Edge-Cloud architecture for emergency response
• Differential Privacy with hybrid Laplace-Gaussian noise mechanism
• Blockchain security for trusted communication and data integrity