Royal Melbourne Hospital trials machine learning to predict ED stays

The models are designed to adapt across various healthcare settings.

The Royal Melbourne Hospital (RMH), in partnership with the University of Melbourne, is trialling machine learning models to predict the length of stay for patients in the emergency department and their next destination.

The models are designed to adapt across various healthcare settings, including other hospitals.

The project has produced 12 models to forecast four outcomes at three time points, using historic, de-identified hospital presentation data.

The technology’s usefulness is yet to be determined, but work is underway to explore its potential in clinical environments, said Mark Putland, RMH Director of Emergency Medicine.

“These tools are about supporting our clinicians, not replacing them, and ensuring patients continue to receive timely, high-quality care,” Putland added.