A Discussion on Process Mining in Healthcare
In recent years, process mining has been studied for a number of applications in the healthcare domain. With the increasing need to bring healthcare processes better under control, such applications have significant potential to increase the share of process-oriented care delivery. Despite the potential, the usability of the method has been limited, mainly due to the overly complex models, which have been obtained in the healthcare settings. The complexity of these models is often attributed to the complexity of the health care domain. In this presentation, we will discuss whether this is sufficient explanation and argue that many process mining methods fail to identify good process models, even for well-defined clinical environments. We identify a number of reasons for this shortcoming and discuss a few challenges for future research in process mining in healthcare.