Role-based access control through on-demand classification of electronic health record

Area of Science:

• Health Informatics
• Computer Science
• Data Security

Background:

• Electronic health records (EHR) facilitate seamless patient information exchange among healthcare providers.
• Existing access control mechanisms, including Role-Based Access Control (RBAC), are crucial for managing EHR access.
• Limited research exists on on-demand RBAC, highlighting a gap in secure and flexible EHR data management.

Purpose of the Study:

• To propose a novel algorithm for on-demand role-based access control (RBAC) in electronic health records (EHR).
• To enhance the security and robustness of EHR access control through physical data isolation.
• To leverage data mining techniques for efficient user role classification and authorization.

Main Methods:

• Developed a new algorithm for selective policy combination tailored to individual EHR database users.
• Extended the data mining classification technique to group EHRs based on defined user roles.
• Utilized feature vectors derived from EHR data to represent user roles for classification and authorization.

Main Results:

• Achieved robust and secure access control via physical data isolation.
• Successfully grouped EHRs according to specified roles using a classification-based algorithm.
• Demonstrated an effective method for defining user authority through feature vector classification.

Conclusions:

• The proposed on-demand RBAC algorithm enhances EHR security and data management.
• Physical data isolation offers a more robust security approach for sensitive health information.
• Data mining classification provides an effective framework for managing user roles and access permissions in EHR systems.