Critical infrastructure for training and innovation with synthetic health data

Start Date

4 October 2023

End Date

31 October 2025

Project Participants

University of Sydney

Status: Completed

Tags: artificial intelligence (AI), data analytics, data governance, data linkage, education and training, health system, privacy preserving, research infrastructure, synthetic data, tertiary education, workforce capability

Opportunity 

Access to realistic health data is essential for training Australia’s digital health workforce and developing better analytical methods but this is difficult while maintaining appropriate privacy and confidentiality controls. 

This project explored whether high-quality synthetic health data could bridge that gap, using existing open-source synthetic data tools and testing their usability in real world settings and projects with industry partners.  

Project Objectives  

  • Gap analysis of tools for data pipelines in production informatics, data science and health research with medical-device grade traceability and reproducibility. 
  • Trial capstone projects to support experiential learning with realistic data challenges for undergraduate, postgraduate and professional short course students. 
  • Trial datathon and industry projects using synthetic data supporting innovation over retrospective data for clinical and informatics research and development. 
  • Evaluation of algorithm accuracy when trained on synthetic vs real data. Utility assessment for research and innovation. Roadmap for tooling and platform. 

The project delivered synthetic cohorts, reproducible evaluation scaffolding, internships, student projects and a national datathon involving 35 students.  

It also established discovery collaborations with Australian health-sector partners and completed a detailed proof of concept using NSW Health’s Lumos data asset. In the Lumos demonstration, synthetic data closely reproduced the statistical and predictive characteristics of real data, while correlation patterns and key clinical relationships were strongly preserved. 

The project has laid the groundwork for more accessible and privacy-aware synthetic data education and innovation in Australia. Its findings support a model in which sensitive data remains securely within a health service environment while approved synthetic datasets are used for training, exploratory research and method development. 

The project gives health services, universities and technology partners a practical foundation for reducing barriers to health data collaboration. It identifies the requirements for responsible scale-up: national governance guidance, standard privacy tests, documented release thresholds, audit trails, educator resources and continuing maintenance of open-source evaluation tools. 

Integrity, Excellence,
Teamwork and Authenticity

Hand holding a smartphone against a colourful, defocused background