Data Science Centre to drive research innovation and discovery
The University of Sydney is making a major commitment to using its valuable data to improve lives and benefit society with the launch today of the Centre for Translational Data Science.
Crime prevention, reducing youth unemployment, advancing medical treatment and industry efficiency are among the centre’s goals.
The centre is one of only a few worldwide that is using data for ‘translation’ – interpreting the data to solve real world problems including the creation of tools and software that can address economic and social challenges.
Professor Hugh Durrant-Whyte, Director of the Centre for Translational Data Science.
”Data is the currency of the digital age and is changing how we live. It will be an age of wonder if we succeed in exploiting information to our best advantage including excelling at analysing relatively small data sets. The centre positions the University to provide leadership in this area including training a new generation in data-driven science,” said Professor Hugh Durrant-Whyte, the director of the centre.
The University already uses data scientists, especially in the areas of engineering, physics and health research. The creation of the centre will enable data expertise to leverage research across a much wider range of disciplines from criminology to nursing, psychology to government, from veterinary science to geoscience.
“This is the latest in our whole-of-University multi-disciplinary initiatives, building on the success of the approach exemplified by the Charles Perkins Centre, Brain and Mind Centre, Australian Institute for Nanoscale Science and Technology, the China Studies Centre and the Sydney Southeast Asia Centre”, said Dr Michael Spence, Vice-Chancellor of the University of Sydney.
“It supports the University of Sydney’s commitment to delivering the highest impact from our research whether social, scientific or economic and includes commercial developments for national benefit.”
Current and planned projects
- The University of Sydney has funded five data scientists to work in health-related projects at the Westmead Institute of Medical Research where many of the University’s leading medical researchers are based.
- Their focus is to work with clinicians on projects that will provide insights within months and be translational in their respective fields, using routinely collected clinical data. The projects are drawn from various areas and hospitals.
- These Westmead projects include the development of 'dashboards' to incorporate data to provide clear and comprehensive overviews of patient data that update in real time at the Breast Cancer Institute at Westmead; predicting whether a patient will be re-admitted to hospital within a month using the results of various routinely performed pathology tests; predicting health outcomes for pre-term babies from the continuously monitored signals such as oxygen levels in the brain and breathing.
- In mental health the focus is on the reasons why people who have had a severe mental issue are admitted to Emergency Departments and Hospitals within Western Sydney, in order to inform preventive strategies.
- The Brain and Mind Centre is working with a number of public sector, non-government and corporate organisations to understand what enables young people with problems to stay in or rejoin employment, education or training.
- The better we can understand how complex factors such as family, mental health or socioeconomic issues and their interactions lead to disengagement, the earlier measures can be taken to prevent what may become a lifetime of lost opportunity.
- The Charles Perkins Centre is collaborating with the CTDS to develop a strategic initiative in precision medicine. Considerable amounts of data including information about the genome, the proteome (proteins expressed by a cell) and the metabolome (metabolites present in a cell or tissue) of many individuals will be assembled to identify data patterns that predict long term health outcomes.
- These approaches will revolutionise health care and provide novel approaches for preventing metabolic diseases such as type 2 diabetes.
- The University's Institute of Criminology together with CTDS is working with the NSW Police Force to understand and predict the occurrence of domestic violence related crime in Sydney’s Central Metropolitan Region.
- In order to achieve this, new statistical models of criminal behaviour are being built based on historical data, demographics, health records and victim surveys. Other projects are assessing the role of Police Citizens Youth Clubs and exploring the effect of rapid population growth and urban development on specific criminal offences.
- Professor Yun-Hee Jeon from Sydney Nursing School is leading the Aged Care Quality Indicator Program in partnership with CTDS. The program aims to develop and test an evidence-based quality framework for residential aged care and services.
- It will use data on over 100 indicators of care quality such as nutrition, medication and injury, as well as staffing and policies, from over 400 residential aged care facilities. This closely aligns with a national aged care policy agenda and its findings will play a crucial role in building evidence for what constitutes good quality residential aged care and how it can be measured and interpreted.
- While data analysis has long been used to study political behaviour, researchers at the US Studies Centre are applying data science to learn about segments of society that are hard to reach and under-represented in traditional methods of data collection and analysis, such as the poor, the young and the itinerant.
- Tools from data science are also being used to gather and analyse massive collections of political speech. Researchers are using these tools to investigate the prominence of the United States and Australia in political discourse in both countries, contrasting this with the prominence of either country in mainstream media and social media and the way sentiment towards either country responds to events.
The centre will have crucial support from the Sydney Informatics Hub to build skills, capacity, community and collaboration in data science across the University.
The hub, a new multi-million research facility, employs a team of 15 dedicated data science engineers, working with the University’s academics and clinicians to extract information from data, working under the guidance of skilled data researchers. It provides high-performance computing, data storage, data privacy and software engineering support.
The Hub and CTDS rely on the availability of high performance computing and to support both initiatives the University has this week tripled the size of its flagship high performance computer Artemis.
The launch of the Centre for Translational Data Science (CTDS) takes place during the University’s Innovation Week.
More current and planned projects
- At Royal North Shore Hospital, the Kolling Institute is investigating the appropriate use of CT scans of the brain. The project involves using language processing to classify the diagnosis associated with a particular scan.
- The Northern Sydney Cancer Centre at Royal North Shore Hospital aims to revise the testing of prostate cancer patients following treatment. Various statistical methods are being applied to data to determine who should return early and who could safely delay their test. This will provide better patient outcomes and major savings to the health system by avoiding unnecessary testing.
- Machine learning replicates in machines the information processing capabilities of humans to make sense of data, and is another important focus of the Centre. These projects include developing a cognitive aid for people with dementia and using machine learning algorithms to predict driver behavior and therefore risk.
- As part of the Uncover Project led by the Australian Academy of Science the Centre, and collaborators, will address the challenge of revealing the unknown geology of what lies beneath the weathered rock and sedimentary basins covering approximately 80 percent of Australia.
- Ocean sediments cover 70 percent of the planet’s surface and underpin our understanding of how the deep ocean will respond to global warming. The University’s work in this area builds on the creation of digital maps created from over 200,000 seafloor sediment samples.
- Data from the University’s dairy farm operations, based on research from the Faculty of Veterinary Science, can suggest the optimal use of pasture, feeding and breeding to create major savings for the industry.