New Frontier Federated Machine Learning Network Launched to Transform Health Research in Australia

The Australian Research Data Commons (ARDC), in collaboration with leading Australian universities and digital research partners, has launched the Frontier Federated Machine Learning Capacity Building for Australia Project. It is a national initiative designed to make it safer and easier to securely analyse sensitive health data without moving it from where it is stored.

The two-year project brings together three of Australia’s major federated learning initiatives: the Australian Cancer Data Network (UNSW), the Australian Imaging Service (University of Sydney) and the National Infrastructure for federated learNing in DigitAl health (NINA, led by the University of Queensland and QCIF). The project will build shared tools, governance frameworks and technical architectures that will accelerate the use of federated learning in health and medical research.

Federated machine learning enables models to be trained directly where data resides, rather than centralising sensitive datasets. This privacy-preserving approach is increasingly important for analysing clinical, imaging and registry data, which are often too sensitive or too complex to move across institutional or jurisdictional boundaries.

Prof Lois Holloway (UNSW and Ingham Institute), the project lead, said the initiative addresses a major gap in Australia’s digital health research capability.

“Researchers want to work together nationally and internationally to understand health issues that affect people across Australia, but many datasets cannot be combined because of privacy and governance requirements. By building a trusted foundation for federated learning, we are making it possible to collaborate safely while protecting patient privacy at every step,” said Prof Holloway.

What the Network Will Deliver

The project team will create practical tools, shared standards, training and a new national community to help Australian researchers use federated machine learning safely and effectively. This includes technical frameworks, governance guidance and a secure architecture for running federated learning across multiple organisations.

The Federated Machine Learning Network is a co-investment partnership between the ARDC and partners across UNSW, the University of Sydney, the University of Queensland and QCIF Ltd.

Dr Adrian Burton, Director of the People Research Data Commons at ARDC, said the investment supports the ARDC’s mission to enable national-scale infrastructure for health research.

“Federated analysis is becoming more and more popular to avoid moving sensitive data around.  This project will promote national consistency and best practice to make it simpler for researchers and data custodians to use these frontier methods,” said Dr Burton.

By strengthening national capability in secure analytics, the project is expected to benefit clinicians, researchers, data custodians, industry innovators and, ultimately, patients. This will help generate insights that improve health outcomes and service delivery.

Learn more about Frontier Federated Machine Learning Capacity Building for Australia.

The ARDC is enabled by the National Collaborative Research Infrastructure Strategy (NCRIS) to support national digital research infrastructure for Australian researchers.