Skip to 0 minutes and 0 seconds Health care has changed significantly in the past decades. Through advances in medicine and science in general, more and more diseases can be treated more effectively. However, this has also increased the complexity of the health care system. Medical staff needs more training, machines are becoming more and more complex, and multiple solutions exist to cure a particular disease. This all increased the cost of health care significantly. 10% of the gross domestic product of almost any country is being spent on health, and only growth is predicted. But in this day and age, we have the data. Data can help us analyse what’s going on, and suggest improvements.
Skip to 0 minutes and 0 seconds In this course, we will teach you how to use event data and process mining techniques to analyse the data that’s already there to get insights and suggest improvements for the processes. Consider this very simple process of a patient treatment. I hope you can see that this generates a lot of data that can be used to analyse the process flow, and provide insights and suggestions for improvement. Using process mining techniques, applied on this data, we can automatically discover a process model. A process model describes which activities are performed, and in which order. And in the top model, you see yellow dots, moving through the model indicating where patients are at a particular time and date in history.
Skip to 0 minutes and 0 seconds in the bottom model, we are projecting time performance on this process model. We automatically discover this model, and time is now being projected on the activities, showing where queues exist. This allows you to analyse the process flow in great detail. When you follow this course, you will learn how to get this data, how to apply existing techniques to get these models. But also, how you can analyse conformance and compliance issues with current rules and regulations, how users are working together, and much more. If you follow this MOOC, then you will be able to do all this on the data in your health care organisation.
Skip to 0 minutes and 0 seconds I’m Joos Buijs from Eindhoven University of Technology, and I hope to see you again in this course.
Process mining in healthcare
In this course you will learn how you can analyze healthcare data by applying process mining techniques and the process mining tool ProM.
The course is divided in four weeks, where in each week we combine theory and concepts, with practical exercises and real-world examples. Throughout the course we will show how certain analysis can be done in the free and open source process mining tool ProM.
Quizzes and tests
Each activity will have at least one quiz to test what you’ve learned. We distinguish two types of quizzes: one that will test concepts and whether you understood the lecture, and a second type that will walk you through the use of the tool ProM and test whether you understood. Similarly, each week will be closed by a test to verify what you have learned that week. At the end of the course there is an additional test that walks you through a process mining project, and asks you for some key results and insights.
Concretely, we will cover the following topics in this course (among others):
- Process mining in general
- Event data in the healthcare environment
- Event data handling and conversion
- Event data exploration
- Process discovery and conformance checking
- Process improvement and challenges in healthcare
By the end of this course you will be able to:
- Explain how process mining can help in analysing and improving healthcare processes
- Identify opportunities for process mining in a healthcare organisation
- Describe the data requirements in order to apply process mining
- Interpret the results of various process mining techniques in the ProM lite tool
- Apply ProM lite on real data to obtain process mining results
This course is a collaborative effort between several partners, giving you different insights in the field. In the next step each of the educators will introduce themselves briefly.
- Joos Buijs, Eindhoven University of Technology
© Eindhoven University of Technology