First Steps Towards Process Mining in Distributed Health Information Systems

Abstract

Business Intelligence approaches such as process mining can be applied to the healthcare domain in order to gain insight into the complex procedures taking place. Disclosing as-is processes helps identify room for improvement and answers questions from medical professionals. Existing approaches are based on proprietary log data as input for mining algorithms. IHE defines in its ATNA profile how real-world events must be recorded. Since IHE is used by many healthcare providers throughout the world, an extensive amount of log data is produced. In our research we investigate if audit trails, generated from an IHE test system, carry enough content to successfully apply process mining techniques. Furthermore we assess the quality of the recorded events in accordance with the Maturity level scoring system. A sample audit message, recorded while creating a patient document, is presented. Furthermore it is evaluated whether the information is valuable for process mining.