The University of Queensland is leading the National Infrastructure for Federated Learning in Digital Health (NINA) project, which aims to unlock siloed data on chronic diseases using machine learning. NINA received funding of A$6 million from the Australian government’s Medical Research Future Fund and an additional A$7.7 million from various universities and organizations. The project will harmonize and protect privacy while analyzing and sharing data to develop innovative solutions for managing chronic conditions and addressing the challenges of data access and interoperability in healthcare.
The University of Queensland is leading a groundbreaking project aimed at harnessing the power of machine learning to unlock siloed data and address the challenges associated with managing chronic diseases. The initiative, known as the National Infrastructure for Federated Learning in Digital Health (NINA), has received substantial funding from the Australian government’s Medical Research Future Fund, along with contributions from UQ, Monash and Macquarie universities, and the Queensland Cyber Infrastructure Foundation.
Over five years, NINA will enable researchers to access fragmented information on debilitating chronic conditions, including diabetes, rheumatoid arthritis, and osteoarthritis. The project will ensure that the data is prepared and standardized to global standards while prioritizing individual privacy. By generating and sharing analyses across health organizations and states, NINA aims to facilitate collaboration and drive the development of innovative solutions for managing chronic diseases.
To achieve its goals, UQ will collaborate with 23 Australian and global partners to co-design the conceptual framework for NINA and expedite the adoption of the data model at a national scale. This collaborative effort seeks to address the existing challenges faced by researchers in accessing health databases and advancing digital health research in Australia.
According to Clair Sullivan, associate professor at UQ’s Queensland Digital Health Centre, Australia possesses excellent digital health records. However, these records are siloed across different health systems, preventing researchers from leveraging millions of treatment records and trends related to chronic conditions. The lack of data connectivity between primary, secondary, and tertiary care has created significant barriers.
The NINA project aims to bridge this gap by “bringing machine learning to the data” instead of attempting to merge disparate datasets. By doing so, it aims to find effective solutions for managing chronic conditions. Chris Bain, Professor of Practice in Digital Health at Monash University Faculty of IT, highlights that current privacy and data sharing restrictions hinder the meaningful utilization of health databases. The NINA project intends to overcome these challenges and leverage the potential of data to improve healthcare outcomes.
This project aligns with the larger trend of healthcare organizations in Australia and New Zealand grappling with data sharing and interoperability issues. A recent study commissioned by InterSystems reveals that while there is a desire for digital transformation using data and analytics, most healthcare organizations cannot share real-time data and integrate disparate systems. Despite having vast amounts of data, healthcare providers face limitations due to the lack of interoperability between multiple datasets.
To address this issue, the Australian Digital Health Agency has launched the National Healthcare Interoperability Plan, which aims to create a more connected Australian health system by 2027. The agency has partnered with Health Level Seven Australia to promote the consistent adoption of Fast Healthcare Interoperability Resources (FHIR) standards in the country. Additionally, it has collaborated with CSIRO’s Australian e-Health Research Centre to establish the National Clinical Terminology Service, which facilitates connectivity across the health system by providing terminology services and tools.
The NINA project, alongside these broader initiatives, represents a crucial step toward leveraging machine learning and data-driven approaches to unlock the potential of siloed health data and drive advancements in chronic disease management. By breaking down barriers and facilitating collaboration, these efforts have the potential to revolutionize healthcare outcomes in Australia.
