Future postgraduate student scholarships

Biomedical science projects

Development of a Potential Diagnostic Test for Bipolar Disorder (BD) with EEG

BD, also called manic depression, is currently diagnosed in over six million people worldwide with about three million in the USA alone. The condition has phases of mania and depression and periods of remittance. Full cycles of BD can occur as many as three times a year and for many patients this is a lifelong condition. BD is effectively treated, once it is diagnosed. It is estimated that 20% of sufferers go undiagnosed and many more are misdiagnosed. BD has a hereditary component, with the slow hemispheric switch rate being an indicator of the genetic trait. This phenomenon allows for an objective test for BD, even if an individual has not had an episode of BD. The slow switch rate allows relatively easy separation of BD sufferers from those exhibiting symptoms that may have other causes. Current projects aim to establish clinical EEG measurements for comparing the inter-hemispheric functional rivalry in normal subjects and BD patients to assist diagnosis of BD. A unique test that monitors the rate of switching between the hemispheres of the brain in response to visual stimuli has been devised. Our research will apply this technique to measure the switching rate between the hemispheres, in terms of the EEG waveform variation and investigate how such measurements can be applied as a means to diagnose bipolar disorder (BD).

Supervisor

Dr Guang Bin Liu

Influence of inhomogeneous head tissue conductivity in forward and inverse electroencephalograms(EEG) using realistic head models

The electroencephalogram (EEG) represents electrical activity manifested by the ensemble of a great number of neurons within the brain. While the primary regions of activity are well located, the accurate estimating of the location and distribution of the underlying neuronal generators based on the EEG is highly sensitive to uncertainty in the conductivities of head tissue. In this project we investigate the influence of head tissue conductivity on EEG by comparing the forward calculated electrical potential at 128 detectors and the inverse localization accuracy of realistic head model using dipolar sources and containing inhomogeneous scalp, skull cerebrospinal fluid, gray, and white matter conductivities to the results obtained using a FEM realistic head model with the same dipolar sources but containing published baseline conductivity values.

Supervisors

Dr Yan Li is a senior lecturer (Computing) in the Department of Mathematics and Computing and has been with the University of Southern Queensland (USQ) since July 2001. Her qualifications include a BSc and MSc from Huazhong University of Science and Technology (China) and a PhD from the Flinders University of South Australia, Australia.

Her research interests include a wide range area of Bio-medical Engineering, Artificial Intelligence, Blind Signal Processing and Independent Component Analysis. She is the chairperson for the research division of Medical Informatics and Neural Technologies in the Centre of Systems Biology (CSBi) at USQ, leading a research group to do head Modelling and analyse EEG brain signals. The interested application areas of this research are in helping patients with brain diseases, such as epilepsy, dementia etc. She is happy to supervise postgraduate students and assist them to achieve their academic goals.

Dr Peng (Paul) Wen is a senior lecturer in the Faculty of Engineering and Surveying and a member of the Centre for Systems Biology (CSBi) at University of Southern Queensland, Australia. He is interested in the areas of biomedical engineering, artificial intelligent, modelling and simulation and networked control system. He obtained about A$1.93 million in 17 research grants, and published 66 papers and one book.

Dr Wen received his Bachelor of Engineering, Master of Engineering and PhD degrees from the Department of Control and Computer Engineering, Huazhong University of Science and Technology (HUST) in China, in 1983, 1986 and 1994 respectively. He also obtained a PhD degree in the biomedical engineering area from The Flinders University of South Australia, Australia, in 2001.

Analysing EEG Brain Signals using Wavelet based Independent Component Analysis

This project aims to develop a software tool to automatic detect problematic signals from a specific brain disorder disease (such as epilepsy and dementia etc) from EEG recordings and help neurologists to diagnose the disease using wavelet based independent component analysis (ICA). The approach merges the advantages of wavelet decomposition and ICA. Wavelet decomposition projects EEG signals into a high-dimensional orthogonal basis where the ICA performance is significantly improved. This project will improve the quality of life of brain disorder patients through accurate diagnoses and early intervention.

Supervisors

Monitoring the depth of anaesthesia using raw EEG signals

The use of clinical signs for assessing depth of anaesthesia (DoA), although universally employed, is notoriously unreliable. Changes in middle latency auditory evoked potentials have been shown to reflect reliably the level of anaesthesia with a wide range of anaesthesia drugs and to detect awareness. However, auditory evoked potential (AEP) waves are difficult to analyse in clinical situation, and are not available for auditory impaired patients. Bispectral index (BIS), derived from the EEG bispectrum, has been shown to predict movement in response to surgery and to detect consciousness when using a variety of anaesthetic drugs. Further study shows that BIS was unable to detect the transition from unconsciousness to consciousness. Nevertheless, there is not a reliable method to evaluate the depth of anesthesia to minimize the chance of awareness and overdosing. This study is to combine the AEP and BIS to develop an accurate monitoring system for DoA.

Supervisors

Applications of Independent Component Analysis to Microarray Data

Microarray techniques have revolutionized molecular biology research by allowing the parallel measurements of genes. This project aims to investigate the applications of independent component analysis (ICA) to gene microarray data. ICA is a statistical method used to estimate underlying sources from observed data. We apply ICA methods for decomposing microarray data into independent components. Each component represents a gene expression pattern of a putative biological process. Genes that exhibit significant up regulation or down regulation within each component are grouped into clusters, and putative biological meaning is assigned to each component. The ICA performance will be evaluated with other existing techniques.

Supervisors

Advanced role-based access control architecture for enterprise-wide applications

Information sharing in a distributed enterprise usually occurs in broad, highly dynamic network-based environments, and access control in a secure manner poses a difficult challenge. This project develops an architecture for secure information exchange and secure information access in Internet-based collaborative environments. We will design a new role-based access model to identify the issue of selective information sharing in large health-related enterprise systems; and to address and manage conflicting roles and permissions; and to provide a set of accesses to group members for a variety of digital data in highly dynamic networks. The feasibility of our architecture will be demonstrated through policy specification, enforcement and industry implementation.

Supervisor

Dr Hua Wang

Development and testing of novel drugs for cardiac disease and Duchenne Muscular Dystrophy

Muscular dystrophy is a genetic condition resulting in a progressive loss of muscle function. There are many different types of muscular dystrophy that vary in their effects and severity. The major focus of the Muscle Research Laboratory is Duchenne Muscular Dystrophy (DMD). Boys with DMD gradually lose muscle function resulting in a loss of the capacity to walk and then die from either respiratory failure or cardiac failure. The focus of our research is to improve the quality and quantity of life for boys with DMD, by improving the function of both their skeletal and cardiac muscles. Our research is seeking to understand the basis of why the skeletal and cardiac muscles are failing and then testing new treatment strategies ranging from restoring synthesis of dystrophin to restoring\prolonging muscle function while preventing fibrosis. We are using a range of novel and known drugs that are not currently being utilised in DMD as well as oligonucleotides to induce exon skipping. By examining the effects on both skeletal and cardiac muscles concurrently, it is possible to ascertain which treatment will be truly beneficial for both muscle types and not selectively beneficial for one and detrimental to the other. Our strategy of determining treatment effects on the whole body, isolated tissue, single cellular and subcellular levels provides a comprehensive picture as to the potential clinical usefulness and mechanisms of actions of pharmacological treatments.

Supervisors

Prof Andrew Hoey, Dr Mike Watson

The Auto-Reconstruction of Realistic Head Modelling of EEG

The objective of this project is to reconstruct a geometry model (3D) of the head from Magnetic Resonance Images (MRI). This work is a part of a wide research program that aims to develop better diagnostic tools for predicting the electroencephalograph (EEG) signal at any point on the scalp as a function of the source locations and blood flow within the cortex. It is envisioned that this will help to achieve a better understanding of the structure and function of the brain.