#machine learning

Prediction of molecular interactions with RNA using AI
This Project will develop deep-learning models to predict interactions of ribonucleic acid (RNA) with other molecules. The expected outcomes are to improve prediction capabilities to decode RNA interactions in disease mechanisms, identify novel therapeutic modalities, and improve existing therapies for targeting RNA. This could result in enhanced capacity to design new therapies and potential to optimise RNA targeting molecules for therapeutic applications.

Applying imaging methods and data analytics to explore the listening brain
This Project aims to understand brain circuits and processes supporting communication in individuals with hearing problems, including those who use devices such as hearing aids and cochlear implants. The potential benefits are that individualised strategies based on real-time brain states estimate algorithms to empower listening and support effective communication. The Project will use brain-imaging techniques‚ including those compatible with listening technologies, including electroencephalogram (EEG), to explore the listening brain. The Project will explore brain changes that arise from hearing loss, how changes in brain function – within and beyond the auditory brain – arise to support listening when hearing is impaired, and how these findings can be used as a part of devices such as cochlear implants that engage the rest of the brain to support an individual's listening.

Synthetic CT via Generative AI for MR-guided Radiotherapy Planning in the Abdomen and Lungs
This Project will leverage artificial intelligence to develop and validate synthetic computed tomography (CT) from magnetic resonance imaging (MRI) in the abdominal and lung regions. The expected outcomes are an AI-based synthetic CT model, thorough technical and clinical validation and potential patent/licensing opportunities. This may reduce unnecessary ionising radiation of CT in patients and improve treatment efficiency during radiotherapy planning.

Clinical lab automation with AI human robot interface
This Project will develop an AI-based robotic programming interface based on large-language model that allows practitioners, regardless of their technical expertise, to efficiently program and control robots. The expected outcomes are to improve efficiency in designing and deploying clinical lab automation and to expand the use of robotics within laboratories. This may lead to improvements with workflow for clinical lab automation, particularly during high-demand situations like pandemic outbreaks.

Cyber security risk mitigation for sensor data integrity
This Project will investigate cyber security risk mitigation approaches to secure the integrity of sensor information feeding into critical infrastructure operational systems and digital twins. The expected outcome is the development of guidelines for implementing robust cyber security measures. This may enhance resilience against cyber threats and ensure the integrity of decision-making processes.

Predictive farming
This Project will investigate artificial intelligence (AI) to improve weather forecasts and use crop models for making better farming decisions. The expected outcome is protocols for integrating AI-enhanced weather forecasts and crop models into farm management and planning tools. This is expected to result in farmers making more informed decisions that optimise resource use, boost crop yields, and enhance overall farm productivity and profitability while minimising environmental impact.

Satellite-Based Methane Detection
Methane is a potent greenhouse gas and an important contributor to climate change. This project will develop neural network-based methods to detect anthropogenic methane plumes in satellite imagery and quantify emission rates. The expected outcomes are better detection and monitoring of methane emissions in Australia compared to current methods, with enhanced temporal and spatial coverage. These advancements will enhance Australia’s capability to efficiently identify, quantify and mitigate methane emissions.

Evaluating Robotic Medical Surgery with Multimodal and Responsible AI
This Project aims to develop multimodal and responsible artificial intelligence (AI) for automated robotic surgery assessment. The expected outcome is to develop multimodal and responsible AI for automated robotic surgery assessment. The potential benefit is enhanced surgical training, improved patient outcomes, reduced training costs, and increased transparency.