Physiological and Physics-Based Machine Learning for ECG/PPG Emergency Department Triage and Diagnosis

The MARCS Institute/ICNS, together with the School of Computer, Data and Mathematical Science is offering a research scholarship to a highly motivated PhD candidate to work within a research group on machine learning for medicine, data science and signal processing. This project will be a joint project together with South-Western Sydney Local Health District.

A hospital's Emergency Department (ED) needs to act quickly to identify patients at high-risk. Sometimes physiological indicators of danger, such as loss of blood pressure, occur after damage has been done. ECG and PPG are ubiquitous diagnostic tools in EDs. Under pressure and through looking at waveforms, it is hard for a human to ascertain the patient's status in a timely fashion. AI/Machine Learning (AI/ML) and Signal Processing techniques offer great promise to improve diagnosis quickly and improve outcomes. AI/ML on its own cannot, however, help with interpretation and generalisation. This project will investigate incorporating the physics of sensing and physiology into the acquisition and interpretation of the mass data available to assist triage and diagnosis.