Lightweight Bidirectional Differential Privacy Protection Method for Privacy-Sensitive Data in Medical Internet of Things

Puning Zhang,Tingting Wu,Jing Yang,Jing Wang,Zhen Zhang
DOI: https://doi.org/10.1109/healthcom56612.2023.10472379
2023-01-01
Abstract:The sharing of medical data nowadays plays a pivotal role in advancing medical research, fostering innovation, enhancing clinical decision-making, improving healthcare quality, and propelling public health monitoring and personalized medicine development. Its significance in the field of medicine cannot be overstated. However, medical data encompasses a wealth of sensitive information pertaining to patients, possessing highly pronounced privacy characteristics. Consequently, while promoting the sharing of medical data, privacy protection must be accorded paramount importance. Existing approaches pre-dominantly focus on hierarchical categorization from the user side, aiming to achieve user-level differentiation, yet disregarding the diversity inherent in data sources. Moreover, these approaches suffer from excessive computational complexity. To address these issues, this paper presents a lightweight bidirectional differential privacy protection method tailored for privacy-sensitive data in Medical Internet of Things (MIoT). It encompasses the design of a protection architecture that combines symmetric encryption and attribute fusion, incorporating the principles of differential privacy protection. A lightweight attribute-based encryption scheme is devised to cater to differentiated user requirements, ensuring data encryption while simultaneously reducing computational overhead and storage space consumption. Additionally, a fine grained privacy protection method, targeting the distinctive characteristics of differentiated data, is introduced to facilitate granular privacy protection for individual sensitive data, ensuring that only authorized personnel can access, utilize, and process the corresponding data. Simulation experiments validate that this proposed approach significantly mitigates computational complexity and communication overhead while demonstrating enhanced data availability.
What problem does this paper attempt to address?