Secure and Efficient Protocols for Providing Personalized Privacy Protection over High-Dimensional Healthcare Data

Abstract

According to the modern rule released by Health and Human Services (HHS), healthcare information will be outsourced to cloud computing services for medical studies. A significant concern concerning outsourcing attention information is its associated privacy problems. However, previous solutions have centered on cryptographically techniques that introduce important value once applied to attention information with high-dimensional sensitive attributes. To deal with these challenges, we tend to propose a privacy-preserving framework to transit insensitive knowledge to commercial public cloud and therefore the rest to trustworthy personal cloud. Under the framework, we tend to design two protocols to produce customized privacy protections and defend against potential collusion between the general public cloud service supplier and therefore the information users. We tend to derive obvious privacy guarantees and finite information distortion to validate the projected protocols. Intensive experiments over real-world datasets are conducted to demonstrate that the projected protocols maintain high usability and scale well to massive datasets.