With the increasing demand for health care, hospitals are looking for ways to optimize their care processes in order to increase efficiency, while guaranteeing the quality of the care. Process modeling is a crucial step for process improvement, since it provides a process model that can be analyzed and optimized. Process mining is a recent promising methodology to discover process models based on data from event logs. However, early applications of process mining to health care has produced overly complex models, which have been attributed to the complexity of the health care domain. In this paper, we argue that existing process mining methods fail to identify good process models, even for well-defined clinical processes. We identify a number of reasons for this shortcoming and discuss a few directions for extending process mining methods in order to make them more suitable for the clinical domain.
|Title of host publication||Proceedings of the 2012 IEEE International Conference on Systems, Man, and Cybernetics (Seoul, Korea, October 14-17, 2012)|
|Place of Publication||Piscataway|
|Publisher||Institute of Electrical and Electronics Engineers|
|Publication status||Published - 2012|