Medical Image Analysis
The Medical Image Analysis group uses images to gain a better understanding of the living human body, to better diagnose diseases, to allow for better intervention, and better monitoring of disease state or the effect of treatment.
Image data includes:
- computed tomography (CT) scans
- positron emission tomography (PET) scans
- magnetic resonance imaging (MRI)
- ordinary visual images; and
- a variety of multidimensional data arrays that blur the boundary between images and signals.
Research within the Medical Device Research Institute (MDRI) includes theoretical work on processing signals and images such as:
- multi-modal and multi-channel signal processing
- image segmentation
- image understanding
- signal and image representation
- machine learning
- hardware design
- visualisation; and
- human-computer interaction.
Early detection of breast cancer
Computer programs are being developed to assist radiologists in interpreting screening mammograms with the objective of detecting breast cancer sooner, and hence, reduce the morbidity and mortality due to this disease.
Talking head, thinking head, teaching head
Speech recognition, language processing, computer lip reading, and computer facial expression understanding are being combined to develop a comprehensive understanding of human communication.
The control of devices by speech is being extended to control devices by thinking. This project, in collaboration with the Cognitive Neuroscience Laboratory at Flinders University, also strives to develop methods for measuring and monitoring skill levels, stress and fatigue.
We actively supervise and support many student projects at both Honours and Post-graduate levels and invite students who are interested in pursuing research in this program to contact us for further information.