Leveraging Synthetic Patient Data for Process Mining in Healthcare

Process mining in healthcare offers the potential to optimize workflows and ensure compliance with clinical guidelines, yet its application is often hindered by data privacy issues and restricted access to real-world patient data. This study presents a methodological infrastructure to explore the use of synthetic patient data, generated with Synthea, to advance healthcare process mining. By integrating HL7 FHIR AuditEvent Resources, we establish standardized audit logs that capture healthcare processes while maintaining interoperability. These logs are then converted into process mining formats such as XES and OCEL, enabling analysis with established process mining techniques. The evaluation was performed using FHIR Resources generated from four Synthea modules, each representing the treatment course of a distinct disease. It demonstrates the accuracy of process reconstruction and the influence of data volume. Despite encountering challenges like missing events and the limitations of FHIR R4 AuditEvent, our approach shows the capabilities of process reconstruction using synthetic healthcare data. Future work will apply this methodology to real-world scenarios to further validate its practical effectiveness in healthcare settings.