Research projects
Our large-scale research initiatives are applying digital solutions to target chronic and infectious diseases.
DIFFERENCE
The Digital Infrastructure For improving First Nations Maternal and Child Health (DIFFERENCE) project is an innovative initiative aimed at bridging the digital divide and enhancing health outcomes for Aboriginal and Torres Strait Islander mothers and babies. This Australian-first project, led by First Nations community leaders in collaboration with University of Queensland researchers, focuses on developing a digital system that integrates primary and secondary healthcare information.
By leveraging advanced technology, the project supports the creation and implementation of First Nations-led care models in clinical practice. A key component of this initiative is the development of the necessary digital infrastructure to support the Birthing in Our Community (BiOC) program, which was first introduced in urban Brisbane in 2013.
The DIFFERENCE project will not only build this infrastructure but also demonstrate and evaluate the impact of a standardized, harmonized First Nations health record, ensuring that culturally appropriate care models are effectively integrated into healthcare systems.
Partners
The DIFFERENCE project is built on strong partnerships in digital health research and clinical applications.
- The Institute for Urban Indigenous Health (IUIH) Aboriginal Community Controlled Health Network
- Mater Health
- University of Queensland (UQ)
- Poche Centre for Indigenous Health (UQ)
- The Commonwealth Scientific and Industrial Research Organisation (CSIRO)
- Queensland Cyber Infrastructure Foundation (QCIF)
- Queensland University of Technology.
In the media: Digital Infrastructure For improving First Nations Maternal and Child Health (DIFFERENCE) project
NINA
The NINA (National Infrastructure for federated learNing in DigitAl health) project aims to revolutionise chronic disease management in Australia by addressing the country's fragmented health data landscape. Traditional methods of centralising data for Artificial Intelligence/Machine Learning applications face significant legislative and privacy challenges. NINA proposes a novel approach using Federated Learning (FL), which allows data to remain at its source while enabling collaborative analysis. This method respects privacy laws and ethical standards, making it possible to harness vast amounts of health data across different jurisdictions without compromising data security.
NINA is a five-year initiative funded by the Medical Research Future Fund (MRFF) and various partners, totalling $13.7 million. The project is led by The University of Queensland and involves over 20 academic, health service, and industry partners. By leveraging FL, NINA aims to create new models of care for chronic diseases such as diabetes, rheumatoid arthritis, osteoarthritis, and cancer. The project will develop scalable ethics and governance pathways, establish necessary technology, and implement FL to enhance chronic disease research and treatment strategies. For instance, the diabetes use case focuses on predicting the risk of developing Type 2 diabetes in women with gestational diabetes, while the cancer use case aims to identify predictors of breast cancer recurrence.
Partners
UQ and the below collaborators aim to create a robust digital health ecosystem across public and private sectors that can significantly improve health outcomes for Australians suffering from chronic diseases.
- Monash University
- Macquarie University
- Queensland University of Technology
- Queensland Cyber Infrastructure Foundation
- Industry partners include Amazon, Google Cloud, Stryker, BioGrid, and the Advanced Robotics for Manufacturing (ARM) Hub.
- Health services partners include Queensland Health, Monash Health, and NSW Health.
In the media: A ‘digital health revolution’ to tackle chronic diseases - UQ News - The University of Queensland, Australia
NASCENT
The sustainability of Australian healthcare is dependent upon the adoption of Artificial Intelligence (AI) technologies and yet Australia is not ready to implement AI in healthcare. Further, a barrier for AI research to be translated into clinical practice in Australia is the lack of infrastructure for real-time AI evaluation capability within our healthcare system.
The National Infrastructure for Real-time Clinical AI Trials (NASCENT) will address these gaps by delivering the cutting-edge AI evaluation infrastructure, AI workforce capability and capacity, and best-practice consumer partnership required for clinical AI applications to meet expected standards for implementation within Australia’s healthcare system, and deliver on the quintuple aim of healthcare.
NASCENT will be Australia’s first attempt to provide a clinical AI prospective evaluation research infrastructure. It will be the critical foundation to the development of AI solutions, and the first component of translating predictive AI research into the real-world clinical environment. NASCENT will address the three identified gaps by developing the infrastructure to enable prospective evaluation of clinical AI algorithms - demonstrated in two areas of unmet clinical need: acute deterioration and sepsis.
Partners
Led by the University of Queensland (UQ) NASCENT has 20+ academic, health service and industry partners.
LACE
The Life and Health after Childhood Cancer (LACE) project aims to better understand the psychosocial and long-term health effects of cancer treatment by creating a national population-based data platform of childhood cancer survivors that will track survivors longitudinally. This important project is funded by the Medical Research Future Fund (MRFF) National Critical Research Infrastructure scheme. Read more about the LACE project.
NEURODESK
The Neurodesk project is an open-source software platform for neuroimaging research. Neurodesk lowers the barrier for researchers to engage in open science practices and advances neuroimaging research by facilitating access to open software, open data, and the development of educational resources. Neurodesk enables scientists worldwide to access neuroimaging data analysis tools and apply them in a reproducible manner. Neurodesk is also instrumental in bridging research and clinical domains by enabling containerised neuroimaging tools to be integrated into clinical environments, including MRI scanner consoles.
Neurodesk cultivates innovation, collaboration, and the implementation of best research practices. It offers a repertoire of 269 tools across various software versions, encompassing a wide range of neuroimaging data analysis domains. The project has been widely adopted, with a user base exceeding 1,300 active users from over 60 different countries each month. It is also a core teaching and training resource at institutions such as the University of Queensland (Australia), the University of Wollongong (Australia), Clemson University (US), the Athinoula Martinos Centre at Harvard (US), the Technion (Israel), and the University of South Carolina (US).
Neurodesk’s applicability has recently extended into clinical settings. By leveraging Siemens’ OpenRecon and community-driven efforts, we are integrating containerized workflows into MRI scanner environments to enable real-time, reproducible image reconstruction and analysis at the point of acquisition. This enables the translation of science directly into clinical practice, allowing scientific innovation to benefit patient care as fast as possible.
The project is co-led by Early-Career Researchers and actively contributes to Open Science guidelines and community standards through public GitHub development, documentation, and participation in hackathons.
Partners
This work is supported by the Wellcome Trust with a Discretionary Award as part of the Chan Zuckerberg Initiative (CZI), The Kavli Foundation, and Wellcome’s Essential Open Source Software for Science (Cycle 6) Program (Grant Ref: [313306/Z/24/Z])
This project is also supported by the Australian Research Data Commons (ARDC) via the AEDAPT platform project:
Australian Electrophysiology Data Analytics Platform.