Research Projects

Data Science in Health

Learning as We Go: An Examination of the Statistical Accuracy of COVID19 Daily Death Count Predictions

This research provides a formal evaluation of the predictive performance of a model (and its various updates) developed by the Institute for Health Metrics and Evaluation (IHME) for predicting daily deaths attributed to COVID19 for each state in the United States.

The IHME models have received extensive attention in social and mass media, and have influenced policy makers at the highest levels of the United States government. For effective policy making the accurate assessment of uncertainty, as well as accurate point predictions, are necessary because the risks inherent in a decision must be taken into account, especially in the present setting of a novel disease affecting millions of lives.

Graphical Models on Childhood Obesity

Childhood obesity is recognised as a serious public health issue worldwide. The project aims to detect direct/indirect casual factors to childhood obesity resulting in a better understanding of childhood obesity. Based on more knowledge, intervention strategies to prevent childhood obesity can then be developed.

The project is approaching its end. Bayesian networks are adopted to model the joint distribution of multiple random variables, including children’s Body Mass Index. An efficient Markov Chain Monte Carlo algorithm has been adopted to draw samples from its posterior distribution. We have achieved a better understanding of childhood obesity.

This research is supported by the Paul Ramsay Foundation.

Developing an age-agnostic sepsis screening tool for patients presenting to Emergency Department.

Sepsis, defined as organ dysfunction due to an infection, is a major cause of morbidity and mortality worldwide. Around 150,00 patients in Australia are diagnosed with sepsis each year with approximately 3000 resulting in death.

Early accurate identification is challenging and crucial. Delays in treatment are associated with an increase in mortality of 4% per hour in adults and in paediatrics with a 2-fold increased odds of mortality. The identification should possess a high level of specificity as to reduce ‘alert fatigue’ among ED clinical staff as well as to not promote the unwarranted use of broad spectrum antibiotics.


Characteristics, management and outcomes of patients presenting with chest pain and acute coronary syndrome using data extracted from electronic records from Sydney Health Partners

Chest pain is one of the leading medical reasons for presentation to Emergency Departments. If related to the heart, it is commonly due to an acute coronary syndrome, a leading cause of mortality, morbidity and health care costs in Australia.

This project aims to use electronic medical records to optimise the patient journey and outcomes across Sydney Health Partner local health districts in Sydney. The application of novel data analytic techniques will enable the identification of contributing factors to adverse outcomes and improve clinical care.