Outsourcing high-dimensional healthcare data to cloud with personalized privacy preservation

According to the recent rule released by Health and Human Services (HHS), healthcare data can be outsourced to cloud computing services for medical studies. A major concern about outsourcing healthcare data is its associated privacy issues. However, previous solutions have focused on cryptographic techniques which introduce significant cost when applied to healthcare data with high-dimensional sensitive attributes. To address these challenges, we propose a privacy-preserving framework to transit insensitive data to commercial public cloud and the rest to trusted private cloud. Under the framework, we design two protocols to provide personalized privacy protections and defend against potential collusion between the public cloud service provider and the data users. We derive provable privacy guarantees and bounded data distortion to validate the proposed protocols. Extensive experiments over real-world datasets are conducted to demonstrate that the proposed protocols maintain high usability and scale well to large datasets.