Evaluating and developing the evidence-base and data mining approaches to strengthen the consumer product safety system in Australia. Consumer product-related injuries cause over 173,000 injuries per year though there is limited evidence about the causes and risks to enable early identification and warnings for consumers. This project will evaluate the evidence-base and develop new methods to support an early identification and surveillance system for product-related injuries.
A data science framework for modelling disease patterns from medical images. A data science framework for modelling disease patterns from medical images. This project aims to extract models of disease patterns from medical imaging data, using deep learning, smart image processing, machine learning, and statistical modelling to quantify and model patterns conventional methods cannot detect. These disease models are expected to improve understanding of particular diseases and enable precision medi ....A data science framework for modelling disease patterns from medical images. A data science framework for modelling disease patterns from medical images. This project aims to extract models of disease patterns from medical imaging data, using deep learning, smart image processing, machine learning, and statistical modelling to quantify and model patterns conventional methods cannot detect. These disease models are expected to improve understanding of particular diseases and enable precision medicine, which recognises that there are important differences between individuals with a particular disease, and that when patients are separated into sub-populations with similar disease patterns, treatment can be tailored to these sub-populations.Read moreRead less
Omni-modality medical image analysis and visualisation. The term ‘Omni’-modality imaging (OMI) has been coined to describe the integration of multiple, complementary medical imaging modalities. However, there is currently a lack of an appropriate means to assimilate and derive maximum benefit from these integrated data. This project aims to provide a new approach to OMI data analysis and visualisation, by deriving a novel ‘level of relevance’ from the overlapping anatomical and pathological stru ....Omni-modality medical image analysis and visualisation. The term ‘Omni’-modality imaging (OMI) has been coined to describe the integration of multiple, complementary medical imaging modalities. However, there is currently a lack of an appropriate means to assimilate and derive maximum benefit from these integrated data. This project aims to provide a new approach to OMI data analysis and visualisation, by deriving a novel ‘level of relevance’ from the overlapping anatomical and pathological structures in the data which will be used to suppress superfluous data and highlight the most relevant data to maximise the information gained from the OMI data. Further, OMI visualisation is proposed to efficiently navigate through the overlapping data.Read moreRead less
Assistive technologies for autism support harnessing social media. This project aims to tap social media to revolutionize early intervention therapy for children with autism. By creating open, extensible software for therapy delivery, and tools for parents to access high quality information and support, we will provide children a greater chance to achieve their potential and much-needed relief for parents and carers.
Discovery Early Career Researcher Award - Grant ID: DE150100104
Funder
Australian Research Council
Funding Amount
$330,000.00
Summary
Towards transforming data streams into real-time knowledge. This project aims to address a key problem of interpreting and providing meaningful information in real-time from large volumes of multivariate, noisy and incomplete data in fine-scale monitoring applications. Specifically, it targets air quality monitoring within a workplace. The project aims to significantly advance the current models for online data clustering and real-time anomaly detection in streaming data. The project aims to pro ....Towards transforming data streams into real-time knowledge. This project aims to address a key problem of interpreting and providing meaningful information in real-time from large volumes of multivariate, noisy and incomplete data in fine-scale monitoring applications. Specifically, it targets air quality monitoring within a workplace. The project aims to significantly advance the current models for online data clustering and real-time anomaly detection in streaming data. The project aims to produce computational models for the two aforementioned tasks and a complete system prototype for indoor air quality monitoring. This system has major health benefits for workers and the showcased computational models have various industrial potentials with significant socio-economic benefits to Australia.Read moreRead less
Effective Recommendations based on Multi-Source Data. Large-scale data collected from multiple sources such as the Web, sensor networks, academic publications, and social networks provide a new opportunity to exploit useful information for effective and efficient recommendations and decision making. The project will propose a new framework of recommender systems that is based on analysing relationships between different types of objects from multiple data sources. A graph model will be built to ....Effective Recommendations based on Multi-Source Data. Large-scale data collected from multiple sources such as the Web, sensor networks, academic publications, and social networks provide a new opportunity to exploit useful information for effective and efficient recommendations and decision making. The project will propose a new framework of recommender systems that is based on analysing relationships between different types of objects from multiple data sources. A graph model will be built to represent the extracted semantic relationships and novel linkage-analysis based algorithms will be developed for ranking objects. The results from this project will underpin many critical applications such as healthcare.Read moreRead less
Active and interactive analysis of prescription data for harm minimisation. Active and interactive analysis of prescription data for harm minimisation. This project aims to enhance prescription monitoring to reduce and prevent dangers to the public from inappropriate drug use. The project will develop a framework integrating active machine learning, interactive data mining, and data visualization into analysis of prescription data. The expected outcomes include online interactive analysis of lar ....Active and interactive analysis of prescription data for harm minimisation. Active and interactive analysis of prescription data for harm minimisation. This project aims to enhance prescription monitoring to reduce and prevent dangers to the public from inappropriate drug use. The project will develop a framework integrating active machine learning, interactive data mining, and data visualization into analysis of prescription data. The expected outcomes include online interactive analysis of large scale prescription data and a system that can interact with health professionals to provide high quality real time prescription monitoring, thereby improving patient outcomes and the efficiency of the healthcare system.Read moreRead less
The development of automated advanced data analysis techniques for the detection of aberrant patterns of prescribing controlled drugs. The state of the art in ICT for healthcare monitoring is rapidly advancing, however, the value of data depends on effective tools and techniques. This project will develop novel techniques for the detection of emerging patterns in the prescribing of controlled drugs, supporting Queensland Health’s role in patient harm minimisation.
Discovery Early Career Researcher Award - Grant ID: DE160101518
Funder
Australian Research Council
Funding Amount
$294,111.00
Summary
Multi-Object Recognition of Biomedical Images via Holistic Ontology. This project seeks to advance the development of new biomedical image recognition and analysis solutions by associating biomedical images with biomedical knowledge and personalised data. The provision of accurate and robust multi-object recognition and analysis from biomedical image data is a fundamental requirement for biomedical imaging applications. This project aims to improve the recognition and analysis of anatomical and ....Multi-Object Recognition of Biomedical Images via Holistic Ontology. This project seeks to advance the development of new biomedical image recognition and analysis solutions by associating biomedical images with biomedical knowledge and personalised data. The provision of accurate and robust multi-object recognition and analysis from biomedical image data is a fundamental requirement for biomedical imaging applications. This project aims to improve the recognition and analysis of anatomical and functional structures from biomedical images with ‘holistic ontology’ modelling that represents a multi-level biological, physiological, and anatomical knowledge base. The project will potentially have application in many health care areas, such as computer aided diagnosis, image-guided surgery planning, and image-based disease modelling.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE200101439
Funder
Australian Research Council
Funding Amount
$418,998.00
Summary
Towards a Reliable and Explainable Health Monitoring and Caring System. This project aims to unleash the power of deep learning on health monitoring and caring domain through a safe, reliable and explainable way. Its innovations lie on 1) developing a set of robust and explainable deep learning models that are guaranteed to be safe to complex environmental uncertainty; 2) designing an intelligent health monitoring and caring platform, powered by robust deep learning models, to better support the ....Towards a Reliable and Explainable Health Monitoring and Caring System. This project aims to unleash the power of deep learning on health monitoring and caring domain through a safe, reliable and explainable way. Its innovations lie on 1) developing a set of robust and explainable deep learning models that are guaranteed to be safe to complex environmental uncertainty; 2) designing an intelligent health monitoring and caring platform, powered by robust deep learning models, to better support the home-based health monitoring and caring for the elderly. The result will enable end-users to trust the decisions of deep learning models in safety-critical systems and significantly contribute to Australian aging society and national healthcare economy.Read moreRead less