Stillbirth remains a significant public health challenge in Australia and globally, with persistent disparities in outcomes across different communities. As predictive analytics and machine learning models are increasingly used to identify pregnancies at risk, it is essential that these tools perform equitably for all population groups. This PhD project will critically evaluate risk prediction tools, focusing on accuracy and fairness across diverse population groups, including culturally and linguistically diverse communities, and those living in rural or remote areas.
Given the complexity of advancing equity in health prediction, the project is designed to be adaptable – welcoming students from a wide range of backgrounds, including but not limited to public health, epidemiology, qualitative research, health economics, psychology, ethics, and sociology.
Aim
The aim of this project is to advance fairness in stillbirth risk prediction by evaluating disparities in model performance across different population groups. The research will focus on ensuring that fairness is embedded in every stage of model validation, so that all communities – regardless of cultural, linguistic, or geographic background – benefit equally from predictive advances in care.
Objectives
This project will explore fairness in stillbirth risk prediction from multiple perspectives, allowing candidates to tailor their research focus according to their disciplinary background and interests. Possible objectives include:
- Investigating how fairness is defined, measured, and understood in the context of health prediction and maternity care.
- Assessing the presence and impact of disparities or bias in stillbirth risk prediction models across different population groups.
- Exploring the social, ethical, economic, or psychological implications of fairness (or lack thereof) in predictive health tools.
- Developing, applying, or critiquing methods to enhance fairness – such as statistical adjustments, qualitative inquiry, stakeholder engagement, or policy analysis.
- Generating practical recommendations or frameworks to support the implementation of fair and equitable prediction practices in clinical or public health settings.
The objectives of the project are intended to be adaptable, enabling students to shape the project in line with their expertise – whether that involves quantitative analysis, qualitative research, policy evaluation, or interdisciplinary approaches.
Significance
Fairness in health prediction is essential for reducing preventable stillbirths and ensuring that all families receive the benefits of modern healthcare innovations. By placing fairness at the centre of model validation and deployment, this project addresses a critical gap in the responsible use of predictive analytics. The outcomes will inform best practices for equitable healthcare, influence policy and clinical guidelines, and contribute to safer, more inclusive maternity care. Graduates of this project will be equipped with expertise in health equity, data science, and implementation science, positioning them for impactful roles in research, healthcare, and policy.
Ideal Candidate
This opportunity will provide a full-time on-campus PhD scholarship in the Curtin School of Population Health. This project can be adapted to the student’s background, such as epidemiology, population health, public health, psychology, health economics, qualitative research, and sociology.
This project is open to domestic only applicants.
Scholarship
This research is supported by a Commonwealth of Australia Medical Research Future Fund grant. One scholarship is available to support a successful candidate with a living stipend, up to the value of $38500 p.a. pro rata indexed, based on full-time enrolment, for a maximum period of 3.5 years.
Applications Closing date: 31 March 2026
Enquiries
For enquires please contact Professor Gavin Pereira at Gavin.F.Pereira@curtin.edu.au
To apply please submit an Expression of Interest to Prof Gavin Pereira.