Project description
Emergency department (ED) crowding is a challenging problem faced by many EDs worldwide. The study will use readily available data of ED ambulance visits over one-year period, in two major Australian hospitals, to develop a novel predictive model for emergency department (ED) hourly ambulance occupancy. Methods: various time series analysis techniques will be employed. The figure below shows an example of 3-days predictions (in red) superimposed with ground truth (in green). A rigorous error analysis and comparison with literature will be conducted.
Assumed knowledge
The project requires familiarity with machine learning tools including neural networks as covered in Neural Networks and Machine Learning topic. Students are also expected to possess sound programming skills in Python.
Note: You need to register interest in projects from different supervisors (not a number of projects with the one supervisor).
You must also contact each supervisor directly to discuss both the project details and your suitability to undertake the project.