Nowadays, e-healthcare is a main advancement and upcoming technology in healthcare industry that contributes to setting up automated and efficient healthcare infrastructures. Unfortunately, several security aspects remain as main challenges towards secure and privacy-preserving e-healthcare systems. From the access control perspective, e-healthcare systems face several issues due to the necessity of defining (at the same time) rigorous and flexible access control solutions. This delicate and irregular balance between flexibility and robustness has an immediate impact on the compliance of the deployed access control policy. To address this issue, the paper defines a general framework to organize thinking about verifying, validating and monitoring the compliance of access control policies in the context of e-healthcare databases. We study the problem of the conformity of low level policies within relational databases and we particularly focus on the case of a medical-records management database defined in the context of a Medical Information System. We propose an advanced solution for deploying reliable and efficient access control policies. Our solution extends the traditional lifecycle of an access control policy and allows mainly managing the compliance of the policy. We refer to an example to illustrate the relevance of our proposal.

Abstract. The majority of today's web-based applications are based on back-end databases to process and store business information. Containing valuable business information, these systems are highly interesting to attackers and special care needs to be taken to prevent them from malicious accesses. In this paper, we propose (RBAC + ), an extension of the NIST RBAC (Role-Based Access Control) standard with the notions of application, application profile and sub-application session to distinguish end users that execute the same application, providing them by only the needed roles and continuously monitoring them throughout a whole session. It is based on business application logic rather than primitive reads and writes to enhance the ability of detecting malicious transactions. Hence, attacks caused by malicious transactions can be detected and canceled timely before they succeed.