In this thesis, we explore how process mining techniques can be used to gain insights into the healthcare domain. In this domain one has to cope with cross-functional and multi-disciplinary processes and these are characterized by the terms, dynamic and flexible domain. Today’s healthcare organizations are striving to provide timely, cost effective and quality medical services. The first step to achieve this is to analyze the present processes. The process mining research area aims at extracting useful and meaningful information from event logs. In this research project, we work on process mining techniques that automatically discover the process model underlying the existing processes. The performance of process mining techniques on less structured healthcare processes is investigated by using data from two healthcare organizations. A critical evaluation has been made for some of the existing algorithms in the Process Mining Framework (ProM). With a focus on their limitations a new plug-in for the ProM framework for discovering association rules was proposed and also implemented. These association rules in combination with the clustering technique can be further used for generating process models specific to a group of patients sharing some similar characteristic. The focus of this work is to use all these tools and techniques to gain information about the less structured and flexible processes in healthcare
Gupta, S. (2007). Workflow and process mining in healthcare. Master’s Thesis, Technische Universiteit Eindhoven, (May), 1–163.