Keywords: Simulation, ProcessMining, Tabular
Need: Process mining techniques have the potential to unlock significant insights in healthcare event driven data. A recent publication of research challenges in this area can be found in Enhancing the usability and understandability of process mining in healthcare.
Ten recommendations were made including RC-5 - “Consider healthcare specificities” which highlights the need to understand underlying indirect conditions on the process, the robustness for temporal changes and the ability to focus on infrequent behaviour. This project would seek to apply process mining techniques (especially process discovery) to healthcare data in order to build care pathways that reflect both the average and the rare cases appropriately.
Current Knowledge/Examples & Possible Techniques/Approaches:
Related Previous Internship Projects: N/A
Enables Future Work: Optimisation and Learning algorithms on patient pathways
Outcome/Learning Objectives: Code and report demonstrating the range of considerations that need to be addressed when applying process mining to healthcare data. Preferably, outcome care pathways would be in a form that could be fed into Synthea or similar tool.
Datasets: Possibilities for using event logs from ambulance data, open data, or the emergency care dataset
Desired skill set: When applying please highlight any experience around process mining, tool development, coding experience (including any coding in the open), any other data science experience you feel relevant.
Return to list of all available projects.