Carl A Gunter, Ph.D. - Hospital Access Audit Logs - Roles and Anomalies
Learning Roles and Anomalies in Hospital Access Audit Logs
Carl A. Gunter
University of Illinois at Urbana-Champaign
There is significant risk in denying access to a healthcare provider who seeks to examine a patient’s medical record. For this reason, hospital access control systems are generally based on optimistic security in which providers are given broad permissions; accountability through audit then becomes the primary means to detect and react to policy violations. Hospital audit logs are too large to audit manually so models and automation are needed to support accountability. In this talk I will present results of research on balancing access controls with audit in medical record systems and discuss broader implications for other areas. Key goals involve using audit logs to understand roles and workflows to develop reliable anomaly detection and safe access controls. This work is based on collaboration with Northwestern Memorial Hospital, Vanderbilt University Medical Center, the Johns Hopkins hospital system, the Illinois Office of Health Information Technology, and others. Topics include experience-based access management, role prediction, role evolution, topic models from machine learning, graph-based models from operations research, and formal models of policy.
Carl A. Gunter received his BA from the University of Chicago in 1979 and his PhD from the University of Wisconsin at Madison in 1985. He worked as a postdoctoral researcher at Carnegie-Mellon University and the University of Cambridge in England before joining the faculty of the University of Pennsylvania in 1987 and the University of Illinois in 2004 where he is now a professor in the Computer Science Department and a professor in the College of Medicine. He serves as the director of Illinois Security Lab, the Health Information Technology Center (HITC), and the Strategic Advanced Research Projects on Security (SHARPS). Professor Gunter has made research contributions in the semantics of programming languages, formal analysis of networks and security, and privacy. His recent research focuses on security and privacy issues for the electric power grid and healthcare information technologies.