We are experiencing issues opening hdl.handle.net links on ANU campus. If you are experiencing issues, please contact the repository team repository.admin@anu.edu.au for assistance.

Open Access Theses

Permanent URI for this collectionhttps://hdl.handle.net/1885/3

To view all theses in this collection, select one of the 'Browse by' options (Issue Date, Author, Title, Subject, Title or Type (of thesis). You can also enter your keyword/s into the text box above and click on Search.

ANU theses are harvested by the National Library of Australia's Trove service and other search engines, making them fully discoverable online.

Find Australian theses.
Full instructions available here

Submit your thesis (Approved ANU research theses only)

Please note: The Abstracts displayed in item metadata are in many cases truncated. For the full Abstracts, see the thesis document files.

Browse

Now showing 1 - 20 of 15830

Open Access
V3O5 as a Functional Material for a Solid-State Neuron
(2025) Das, Sujan Kumar
Artificial intelligence (AI) is evident in every facet of our lives but traditional computing, based on CMOS technology and von Neumann architecture, is very inefficient for such tasks. Neuromorphic computing aims to address this shortcoming by emulating the computational architecture of the brain. This typically takes the form of a spiking neural network (SNN) comprising a highly interconnected network of solid-state synapses and neurons, where the synapses serve as distributed memory (synaptic weights) and the neurons function as simple processing units. This thesis is focused on the development of a suitable solid-state neuron based on threshold switching in V3O5. In the first part of this thesis, the structural, electrical, and thermal properties of V3O5 thin films are investigated. X-ray diffraction (XRD), Raman spectroscopy, and electron diffraction techniques show that the as-deposited V3O5 films are polycrystalline with a monoclinic structure. Electrical characterisation shows that the films undergo an insulator-metal transition (IMT) at 420 K. The IMT occurs over a broad temperature range due to the inhomogeneous nature of the V3O5 phase transition, as indicated by temperature-dependent C-AFM maps of through-film current distributions. The thermal conductivity of the V3O5 films was shown to increase slightly with temperature below the IMT temperature and significantly during the IMT due to the increasing electronic contribution. These properties show that V3O5 holds promise as a functional oxide for threshold switching applications. The second part of this thesis focuses on the threshold switching characteristics of Pt/V3O5/Pt memristors. These devices exhibit volatile threshold switching and negative differential resistance (NDR) from the first voltage-sweep and stable cycle-to-cycle operation, with a variation in switching parameters of less than 3 percent during 10000 cycles. The threshold and hold voltages decrease with increasing substrate temperature while still maintaining an NDR window up to the V3O5 IMT temperature. The physical mechanisms underpinning these characteristics are investigated using a combination of electrical measurements, in situ thermal imaging, and device modelling. This shows that conduction is confined to a narrow filamentary path due to the self-confinement of the current distribution. Additionally, it shows that the threshold switching and NDR response are initiated at temperatures well below the IMT temperature due to the temperature-dependent conductivity of the insulating phase. Variations and discontinuities in the NDR slope are further shown to be consistent with the averaging of the electrical oscillations. In the final part, the application of V3O5 memristors as solid-state neurons is investigated. This focuses on the dynamics of individual and coupled relaxation oscillators and leaky integrate-and-fire (LIF) neurons. It is observed that individual oscillators exhibit a fast-spiking pattern and maintain stable relaxation oscillation even at operating temperatures exceeding 400 K. Capacitively coupled oscillators are then shown to exhibit rich non-linear dynamics, including frequency and phase synchronisation, which is a key feature for oscillator-based computing. Effective LIF neuron's behaviour can be tuned through relative timing and amplitude of the external stimulus, and resistively coupled LIF neurons exhibit various neural functionalities (i.e. phasic, regular and adaptation, etc.) depending on the input voltage and circuit parameters. The behaviours of both individual and coupled neurons are shown to be well described by a physics-based lumped element circuit model, which, therefore, provides a solid foundation for exploring more complex systems. These results establish V3O5 as a potential functional material for volatile threshold switching and advance the development of robust solid-state neurons with low power consumption for neuromorphic computing.
Open Access
Hole-selective passivating contacts: from process development to application
(2025) Bartholazzi Lugao De Carvalho, Gabriel
Increasing energy demand and need to reduce greenhouse gas emissions are major concerns for this century. In this context, photovoltaic (PV) technology has risen as one of the primary solutions to overcome these challenges. Therefore, it is important to seek ways to further reduce costs and improve performance of c-Si solar panels. Current state-of-the-art solar panels utilize so-called passivating contacts, a key technology to enable high-efficiency solar cells. Passivating contacts comprise a single layer or stack of materials designed to induce carrier-selectivity and passivation between the c-Si and metal electrodes, minimizing recombination losses. Despite enabling high performances, current passivating contact technologies, based on amorphous or polycrystalline silicon, suffer from one or a combination of drawbacks, such as high parasitic absorption, high-thermal budget or thermal instability, and process complexity. In this context, transition metal oxides (TMOs) have emerged as alternative passivating contact materials mainly due to their improved transparency. In this thesis, the development and application of hole-selective passivating contacts based on TMOs is explored. The thesis can be divided into two sections. In the first part, the development of novel atomic layer deposition (ALD) processes for potential hole-selective contact materials based on copper oxide is reported. In the second, a novel ALD interlayer is employed together with existing hole-selective materials and the resulting structures are investigated as hole-selective contacts in c-Si solar cells. Novel ALD processes are developed for the binary oxides Cu2O and CuO using thermal and plasma processes, respectively. The complete process development is reported, including saturation curves and growth behaviour. Self-limiting reactions are achieved for both processes, producing uniform thin films with low roughness. In a third ALD work, a ternary oxide deposition process to form CuxCryOz is investigated using the super-cycle approach. A wide range of compositions is explored, ranging from pure CuOx to CrOx. An unexpected dependency of growth rate as a function of the CuOx/CrOx sub-cycle ratios is found. Through modelling, it is shown that CuOx enhances the growth of CrOx by a factor of 9, working as a catalyst to the reaction. The properties of the various CuxCryOz compositions are investigated both as-deposited and after annealing, and some tested as potential hole-selective contacts. As-deposited Cr-rich CuxCryOz thin films appear to be strong hole-selective candidates as they present a wide band gap and the lowest contact resistivity among different compositions. In separate works, Cu2O and MoOx hole-selective materials are investigated in novel stack structures containing an ALD-deposited AlyTiOx/TiO2 stack as a passivating interlayer. Both stacks are optimised to achieve the best trade-off between passivation and contact resistivity. Record-low recombination is achieved for both materials while maintaining low contact resistivity and high optical transparency. Additionally, the AlyTiOx/TiO2/MoOx stack demonstrates self-healing behaviour after sputtering of a transparent conductive oxide layer, without requiring an additional annealing step. The optimized stacks of AlyTiOx/TiO2/Cu2O and AlyTiOx/TiO2/MoOx are implemented as full-area contacts at the rear side of proof-of-concept c-Si homojunction solar cells. A significant improvement in open-circuit voltage confirms the boost in surface passivation provided by the interlayer. A record efficiency for devices incorporating full-area TMO-based contacts (including electron-selective materials) without an a-Si:H interlayer is achieved for the stack containing MoOx, providing an important stepping stone towards the development of high-efficiency cells employing TMO-based contacts fabricated using simple, low-temperature processes.
Open Access
Dynamics of grassroots collectivism: natural resource conflicts and social movements in Thailand's Special Economic Zones
(2025) Arpornsilp, Ratchada
Open Access
Topics in Tax and Child Benefit Design
(2025) Tin, Darapheak
This thesis examines the interaction between means-tested child benefits and the progressive tax system in Australia, focusing on efficiency, welfare, and equity. It consists of three chapters. Chapter 2 provides the empirical foundation by documenting income dynamics and the role of government transfers in insuring parents against earnings shocks. The findings also reveal that public transfer responses to primary earners' shocks correlate with weak labour supply responses from secondary earners, suggesting potential work disincentives. Chapter 3 develops a dynamic general equilibrium overlapping-generations model with heterogeneous households, incorporating family structure, education, female human capital formation, uninsurable earnings shocks, child-related costs, and the exact structure of child benefit programs in Australia. The model is used to assess the current child benefit design and potential reforms. The results support universal child benefits for enhancing female labour supply, output, and overall welfare. However, they also underscore the importance of means-testing by demonstrating that while the universal scheme improves aggregate outcomes, it imposes a heavier tax burden that disproportionately harms single mothers, the intended beneficiaries. Chapter 4 extends the analysis by examining the broader tax-benefit interplay. It argues that optimizing the tax system alone can undermine child benefit objectives and proposes an optimal joint policy design featuring a less progressive tax schedule and a universal lump-sum child benefit set at 30% of the average income per child. While this system improves overall and parental welfare by enhancing consumption allocative efficiency, particularly for vulnerable groups such as single mothers, it comes at the expense of non-parents, who face higher tax burdens, raising important equity considerations. This thesis makes three key contributions: (i) From a policy-making perspective, it demonstrates that overall welfare (under the veil of ignorance) improves significantly when policies target vulnerable demographics, such as young single mothers. However, the distributional consequences for non-parent households must be carefully considered; (ii) In line with recent literature, it highlights the importance of modelling family structure in policy analysis, as different demographic groups face distinct economic constraints, including household size, earnings capacity, access to family insurance, and child-related costs; (iii) It underscores the critical role of policy interactions, particularly the trade-off between means-testing and tax distortions, in shaping welfare outcomes, providing a foundation for future research. Lastly, the findings suggest multiple policy options to improve both overall and parental welfare. If both tax and child benefit policies can be adjusted flexibly, an optimal policy mix can be designed to maximize welfare. Otherwise, relaxing the phase-out rates of the Child Care Subsidy emerges as a well-rounded reform that effectively balances improvements in female labour supply, output, overall welfare, and equity.
Open Access
Theoretical calculation and machine learning aided design of functional materials for energy conversion
(2025) Sun, Zhehao
This thesis investigates the integration of machine learning (ML) and theoretical calculations to design and optimize functional materials for photocatalytic applications. By combining experimental techniques with theoretical calculations, that is, finite-difference time-domain (FDTD) simulations, and density functional theory (DFT) calculations, this work aims to accelerate the discovery of efficient, selective, and scalable photocatalytic systems for CO2 reduction and seawater splitting. The central focus is on leveraging ML and advanced simulations into experiments to provide new insights into plasmonic photocatalysts and microenvironmental perturbations in photoreaction. The first study explores the development of Ag-TiO2 core-shell photocatalysts for the selective reduction of CO2 to methane (CH4). A significant contribution of this work is the use of FDTD simulations to model and optimize microenvironmental perturbations, thereby enhancing the catalytic activity of the plasmonic core-shell nanoparticles. Additionally, DFT simulations demonstrate that localized surface plasmon resonance (LSPR)-induced electric field enhancements lower the energy barriers for CO2 activation and methanation. Experimentally, this system achieves 100% selectivity for CH4 with a production rate of 75 umol/g/h. This study emphasizes the advantages of microenvironmental engineering in optimizing photocatalytic activity and selectivity, with FDTD and DFT simulations further elucidating the mechanisms of microenvironmental perturbations. The second study focuses on the design of Co-NC@Cu core-shell photocatalysts for solar-driven hydrogen production from seawater. By dispersing single Co atoms on a nitrogen-doped carbon (NC) shell surrounding a Cu core, this novel catalyst achieves a hydrogen production rate of 9080 umol/g/h and a solar-to-hydrogen (STH) conversion efficiency of 4.78%. A key highlight of this work is the detailed investigation of the local coordination environment of the single Co atoms, as well as the thermodynamic and kinetic effects of electric field perturbations on the catalytic process. DFT calculations reveal that the single Co atoms act as highly active sites for hydrogen evolution, exhibiting low energy barriers for the reaction. Furthermore, the electric field's role in enhancing the reaction thermodynamics and kinetics was elucidated, providing insights for further optimization of catalytic performance. Integrating single atoms, photothermal effects, and localized surface plasmon resonance (LSPR) demonstrates a robust and efficient design for seawater splitting. The third study showcases a comprehensive workflow combining ML and DFT calculations to accelerate the discovery and optimization of single-atom-based (SA) 2D photocatalysts. Using a dataset of Janus-TMD materials as a case study, ML models were trained to identify high-activity catalytic sites and screen potential substrates for photocatalytic CO2 reduction. The ML-driven predictions successfully prioritized optimal single-atom catalysts, with experimental validation confirming the activity and selectivity of two synthesized Janus substrates MoOSe with single-atom Pt. Photocatalytic experiments demonstrated the potential of the ML-guided design in delivering efficient and selective catalysts, underscoring the synergy between computational and experimental approaches. The growing dataset of atomic structures, intermediates, Janus configurations, and adsorption models provides a robust foundation for refining ML models and driving innovations in SA-based 2D materials discovery. In conclusion, this thesis demonstrates the successful integration of ML, FDTD, and DFT techniques with experimental approaches for the design of advanced functional materials, which contribute to the development of sustainable energy solutions through CO2 reduction and hydrogen production.
Open Access
The impact of undernutrition on tuberculosis prevalence and treatment outcomes in Ethiopia
(2025) Shiferaw, Fasil
Introduction: Tuberculosis (TB) is a leading infectious cause of death globally, with an estimated 10.6 million new cases and 1.3 million deaths annually. The emergence of multidrug-resistant tuberculosis (MDR-TB), characterized by resistance to isoniazid and rifampicin, further complicates the global TB control efforts. The progression from latent to active TB is influenced by various factors including undernutrition. While the individual-level association between TB and undernutrition has been documented, there has been a lack of studies exploring their geospatial co-distribution, particularly in high TB-burden countries like Ethiopia. There was also limited evidence on the impacts of undernutrition or failure to gain weight during follow-up on treatment outcomes among patients with TB and MDR-TB in Ethiopia. Furthermore, there was insufficient conclusive evidence to measure the effects of nutritional intervention on improving adherence to TB treatment and outcomes. This PhD thesis aimed to address these knowledge gaps by investigating the geospatial overlap of undernutrition and TB (objective 1), quantifying the effect of undernutrition and weight change on TB and MDR-TB treatment outcomes (objectives 2 and 3), and assessing the impact of nutritional intervention on TB treatment adherence, prognostic markers, and treatment outcomes (objective 4). Various study designs and statistical analyses were used to address the thesis objectives. A Bayesian model-based geostatistics was used to investigate the geospatial overlap of undernutrition and TB prevalence in Ethiopia (objective 1), using data from national prevalence surveys. A Cox proportional hazard model was performed to assess the effect of undernutrition on TB treatment outcomes using data from a retrospective cohort study in northwest Ethiopia (objective 2). Longitudinal data analysis using a joint model was applied to determine the effect of weight variation over time on MDR-TB treatment outcomes in patients with MDR-TB (objective 3). Systematic reviews and meta-analyses were performed to synthesize evidence on the impact of nutritional intervention on improving TB treatment adherence, prognostic markers, and treatment outcomes (objective 4). Results: The results of this thesis showed that spatial co-distribution of undernutrition and TB prevalence was found in some parts of Ethiopia. Population density was positively associated with TB and negatively associated with all forms of undernutrition. The distance to a health facility was positively associated with stunting and adult undernutrition. The overall successful treatment outcomes for patients with TB and MDR-TB in northwest Ethiopia were 89% and 67%, respectively. Undernutrition was positively associated with unsuccessful treatment outcomes, higher mortality, and a longer time to sputum culture conversion in patients with TB and MDR-TB. This thesis also indicated a significant relationship between weight variation over time and early unsuccessful MDR-TB treatment outcomes, suggesting that a one-unit increase in weight corresponds to a nearly 5% reduction in the risk of an unsuccessful outcome. Systematic reviews and meta-analyses demonstrated that the provision of nutritional support, as well as zinc and vitamin A supplementation, can improve adherence to TB treatment, increase early sputum smear conversion, and higher levels of serum zinc, retinol, and hemoglobin among people with TB. Conclusion: This thesis provides evidence that could guide decision-making and shape service delivery strategies for supporting nutritional aspects in the prevention and management of TB. The co-distribution of TB and undernutrition provides important evidence for targeted intervention strategies. This thesis has also made policy suggestions that could contribute to achieving the End-TB targets, emphasizing the relevance of strengthening the integration of nutritional interventions as an integral part of TB care.
Open Access
Development of a Modular Biosensor Platform for Direct, Sequence-Specific DNA Detection
(2025) Vernon, Arwen
Point of care (POC) diagnostic tests are useful as they provide fast time-to-results and are easy to use, minimising time to diagnosis and helping combat the spread of infectious diseases. Consequently, POC tests are increasingly in demand, however no existing POC test for DNA analytes is comparable in accuracy and sensitivity to lab-based methods. This thesis, comprised of six chapters, describes the development of a new biosensor for sequence-specific DNA detection by coupling a DNA-binding domain to a bioluminescence resonance energy transfer (BRET) signalling component. Chapter 1 highlights the need for POC diagnostic tests for genetic sequence detection, and the suitability of synthetic biology technologies to meet this demand. Biosensors are introduced as ideal candidates for POC testing devices, and the properties of DNA as a target analyte are discussed. Additionally, the utility of amplifying specific DNA sequences to improve detection thresholds is emphasised. Chapter 2 introduces the materials and methods used throughout the thesis. Reagent sources, equipment, and methods relevant to all chapters are presented. Methods and experimental design specific for individual chapters are described in each experimental chapter. Chapter 3 details the development and application of biosensors for sequence-specific DNA detection. Two DNA-binding proteins were used to produce several unique biosensor designs, which were tested to identify the best candidate. The chosen biosensor's performance was assessed, including measuring output under application-relevant conditions. Overall, a proof of principle bipartite biosensor was developed by combining the DNA-binding specificity of zinc finger proteins (ZFPs) with the ratiometric signal from BRET signal transduction. Importantly, this meant the biosensor was produced in a polypeptidic fashion, obviating the need for expensive synthesis steps using harsh chemicals. To modify the biosensor specificity to a new, clinically relevant, analyte was engineered to bind a DNA sequence specific to Mycobacterium tuberculosis, the causative agent of tuberculosis (TB). Chapter 4 introduces isothermal DNA amplification (IA) to improve the biosensor limit of detection without compromising its applicability to POC. The gold standard for DNA amplification is polymerase chain reaction, however this method is not suitable for application at POC. Instead, IA was used due to its amplification of nucleic acids without requiring expensive equipment and reagents, or skilled technicians. Novel primer sets for two different IA technologies, loop-mediated IA and recombinase polymerase amplification, were developed to amplify the biosensor's DNA targets. Following amplification of target DNA, the crude reaction product was successfully detected with the biosensor. Chapter 5 details the development of a screen to find DNA-binding proteins for specific DNA sequences. The screen facilitated rapid selection of the best-performing biosensors and removed the need for a modular assembly approach to ZFP programming, further speeding up the biosensor development process. Inclusion of cell-free protein synthesis (CFPS) in the screen reduced the time required for protein expression from a week to 3 hours. Furthermore, the crude CFPS product was compatible with the biosensor, obviating the need for protein purification. The screen was used to test different ZFPs for the development of new biosensors, and to create an alternate version of the TB biosensor described in Chapter 3. In the future, the screen could take DNA templates of biosensor designs and return a compatibility result for a DNA target within ~3 hours. Chapter 6 concludes, sets the findings in the context of the literature and provides future directions.
Open Access
Detection and Analysis of Calcitonin Species Produced by Small Cell Lung Cancer
(2025) Howes, Joanna
The work described in this thesis covers the development of an HPLC method that enables the simultaneous detection of CT, CTG, CTGK, and CTGKK from DMS-53 cell growth medium and cell lysate. This method was used to discover that DMS-53 cells produced the novel glycopeptides gCT and gCTG. Isolation of these compounds enabled analysis using various forms of mass spectroscopy to completely characterise both gCT and gCTG. The detection method was utilised along with PHM and CPE inhibitors to investigate CT biosynthesis with the intent to limit CT production. It was found that gCT biosynthesis takes place via a parallel biosynthetic pathway that allows for CT to bypass inhibitory effects by altering the regulation between the two pathways. Additionally it was found that, in the face of interrupted CT biosynthesis, the maintenance of intracellular CT is the first priority of DMS-53 cells. Combined, this helps CT remain insulated from the effects of outside regulation. The CPE inhibitor GEMSA was shown to successfully lower extracellular CTG and gCTG levels however dosed samples exhibited an increase in extracellular CT and gCT.
Open Access
A transformative blue economy? Discourse, hegemony and passive revolution in the Pacific
(2025) Louey, Philippa
Spurred by the advocacy of ocean-reliant communities, the blue economy has emerged as a charismatic vision for sustainable ocean development over the last fifteen years. Central to its charm rests a promise to transform human interactions with the ocean wherein extractive, inequitable and polluting ocean practices are transitioned-out in favour of activities that advance ecological sustainability, social justice and economic efficiency. A "paradigm shift," to echo the World Bank and United Nations Department of Economic and Social Affairs, from business-as-usual with an underlying commitment to systemic change. Reflecting on the blue economy encounters of Pacific ocean development actors, this thesis interrogates the political function of this widely-referenced ocean development vision. I raise the provocation: does the blue economy work to support a transformative approach to ocean development, and if not, what is it doing? Three key case studies provide the empirical base of this research: Pacific regional organisations interfacing with ocean issues, the Cook Islands Government with regards to seabed minerals activities and the grassroots non-government actor, the Locally-Managed Marine Area Network. I engage argumentative discourse analysis to examine both the public performance (document analysis) and private perceptions (key stakeholder interviews) of case study actors. These distinct cases provide insight into the varied strategies Pacific actors engage to (re)negotiate the sustainable ocean visions set out under the blue economy, and the complex political calculations that shape these interactions. This doctoral research indicates that far from transforming humanity's relationships with the ocean, the blue economy has provided a powerful political tool for defending, reproducing and extending neoliberal capitalism across maritime spaces and communities. Contrary to its transformative promise, it is a conservative discourse that enables neoliberal capitalism to perform commitments of change in the face of deepening societal unease; acting to restore consent in neoliberal capitalism's social authority and reassemble it anew for legitimacy in evolving contexts. In this way, the blue economy appears to function in service of passive revolution, which I (re)frame as a process of revolution-recursion as consistent with the work of Antonio Gramsci detailed in Prison Notebooks. While grounded in the Pacific, this thesis contributes to a larger conversation about the role discourse plays processes of political change; specifically, the impact promises of transformation can have in (re)ordering social relations at moments of rupture. An equally important finding of my research is the agency of Pacific actors to unsettle the blue economy, be that through direct critique, creative reworking or careful discursive bounding. Rarely have mainstream visions of the blue economy been adopted wholesale, nor endorsed unconditionally. I observe how the politicisation of blue economy understandings and articulations has enabled Pacific ocean development actors to resist external development (and environmentalist) logics. This has not always resulted in a rejection of the blue economy, but such discursive play has provided an avenue for dissent. 'Holding the political' is a key strategy for refusing the absorption of discourses promising transformation into processes of passive revolution, and an important insight I take from this doctoral project.
Open Access
Bilateral Integration in the Mouse Whisker System
(2025) Singh, Taylor
To generate a coherent percept of the external world, the brain needs to combine and integrate sensory signals processed in the left and right hemispheres. In humans, bilateral integration is an essential and ubiquitous process. Common tasks such as tying shoelaces or typing on a keyboard, require one hand to be aware of what the other is doing, or rather, the brain to be aware of what both hands are doing. Although an essential exercise since the origin of separate cerebral hemispheres 500 million years ago, little is known about the neuronal mechanisms underlying bilateral integration. The primary aim of my PhD project is to elucidate the underlying circuitry involved in the transfer of information across hemispheres and its integration in cortical neurons of the primary somatosensory cortex. By designing a behavioural paradigm which requires bilateral integration, I was able to observe the transformation of sensory signals between the two vS1 hemispheres and propose a basic mechanism for the comparison of sensory stimuli between cortical sides. Here, I investigated bilateral integration of somatosensory inputs in mouse primary vibrissal (vS1) cortex under urethane anaesthesia, awake and passively receiving stimuli and during an active sensory integration task. For the behavioural task, the mice were trained to compare the stimuli between the two whisker pads during head-fixation. In Experiment 1 (Go/No-Go), mice were trained to respond to uni-lateral (Go) but not to bilateral stimulation (No-Go) of whiskers. In Experiment 2 (2-Alternative-Forced-Choice), mice were trained to choose one of two reward spouts that corresponded to the stronger stimulus. Mice learned to perform the task in both behavioural paradigms. In the case of the Go/ No: Go paradigm, mice were required to make comparisons that were more complex than simple associations. Here mice were required to respond to left or right stimulus presentations, but to withhold responses when the left and right were presented simultaneously. This is a complex task, akin to an Exclusive/Or operation in logic, and similar to negative patterning paradigms in learning theory. Surprisingly, mice were capable of solving this task albeit after extensive training. During the behaviour, I visualised vS1 neuronal activity by using NeuroNexus and Neuropixels probes. Under anaesthesia, vS1 neurons were predominantly responsive to contralateral whisker stimulation with no detectable modulation by the ipsi-lateral stimulation. When mice were awake and passively receiving whisker stimuli, the neuronal profile was as expected, with predominantly contralateral and bilateral stimulus responsive neurons, and some responsive to the ipsilateral stimulus presentation. This response profile changed significantly during the Go/No-Go paradigm, where responses to ipsilateral (Go) stimuli became more pronounced and the response to bilateral (No-Go) stimuli was distinct from contralateral (Go) stimuli. The proportion of stimulus responsive neurons also shifted after training, with a distinct population only responsive to the No-Go stimuli. Our results thus showed that neuronal activity in the primary sensory cortex can be functionally reshaped to reflect the ecological relevance of the incoming bilateral stimuli. Future experiments could investigate how interactions with other cortical areas, such as the secondary somatosensory cortex or the medial posterior complex, could underlie this finding. In summary, this thesis presents a novel behavioural paradigm which can be successfully implemented in head-fixed mice, allowing for the investigation of bilateral integration in vS1. The paradigm itself also lends the ability to study complex logical operations in mice. The neuronal findings implicate the involvement of the primary somatosensory cortices in bilateral integration, and support the literature regarding task-relevant plasticity in the barrel cortex.
Embargo
Biophysical and Functional Characterisation of IRF4 DNA Binding Domain Mutations Associated with Lymphoid Malignancies
(2025) Seneviratne, Sandali
Interferon Regulatory Factor 4 (IRF4) is a transcription factor that is uniquely restricted in its expression to lymphocytes. IRF4 gene regulation is documented to play a critical role in many lymphocyte subsets, including in the development and terminal differentiation of B and T cells. Mutations in this transcription factor have been associated with a slew of haematological malignancies including B cell-related chronic lymphocytic leukemia and adult T cell leukemia. Interestingly, a large proportion of disease-causing mutations appear to localise in the DNA binding domain (DBD) of IRF4. This suggests the observed disease outcomes may stem from altered interactions of IRF4 with regulatory DNA sequences. This thesis used a two-prong approach, combining immunophenotyping mouse models and biophysical characterisations to investigate the impacts of three select IRF4 DBD mutations (K59R T95R and L116R). For these mutations, CRISPR generated mouse models were generated and immunophenotyping of major lymphoid organs (spleen, lymph nodes and peritoneal cavity) was conducted. With the primary focus on naive B cells populations, we observed unique effects for each mutation with regards to B1 cell and marginal zone populations. To further understand these in vivo effects, biophysical characterisation of the mutants was conducted. Interestingly, differential scanning fluorimetry and surface plasmon resonance studies found all mutations were impacted similarly compared to wild type. Specifically, mutant DBD's were observed to be less stable in apo forms but increased affinity for known IRF4 DNA binding motifs. Using X-ray crystallography, we were able to investigate the structural basis of these interactions. In the process, we crystallised IRF4WT and IRF4K59R DBDs with a canonical IRF DNA motif which both found to form homodimers. This allowed us to draw comparisons and investigate the structural basis for the observed increase in DNA binding affinities in the mutants. Comparing the interactions of the IRF4WT DBD with that of IRF4K59R, it is clear that not only do these mutations result in localised structural alterations involving the mutated residue, but they also cause global conformational changes to the interacting DNA-protein complex. From these findings, we concluded that, on a molecular level, each IRF4 mutation acts in a gain of function manner. As such, we attempted to design a peptide that would disrupt the increased DNA affinity. IRF4 is known to be critical for the differentiation of plasma cells and the survival of multiple myeloma cells. In order to test the efficacy of this peptide, in vitro trials on plasma cell differentiation and multiple myeloma cell lines were conducted. Although the peptide effectively blocked plasma cell differentiation, it was less successful at inhibiting multiple myeloma cell survival. From these studies, it was clear that although our peptide has potential as a lead compound, more work is needed to improve its efficacy and bioavailability. To conclude, IRF4 is a transcription factor that is involved in many critical immunological pathways however, the wide presentation of IRF4 mutations is still poorly understood. This thesis, primarily focused on naive B cell immunology, highlights the complex nature in which select IRF4 mutations that act very similarly at a molecular level have vastly differing in vivo presentations.
Open Access
Understanding and Improving Generative Adversarial Networks in Low-Data Settings
(2025) Ni, Yao
Generative Adversarial Networks (GANs) have emerged as a powerful tool in machine learning, excelling at generating realistic data through adversarial training between a generator and a discriminator. Despite their success, training GANs remains challenging, particularly in low-data settings, where issues such as discriminator overfitting and unstable convergence arise. These challenges are amplified by the scarcity of available training data, which hinders generalization and leads to poor performance on unseen data. While extensive research has been conducted to improve GANs in limited data regimes, existing approaches often fail to provide a deeper theoretical understanding of the underlying causes of these challenges, particularly from the perspective of gradient control. In this thesis, we address this gap by focusing on how controlling gradients during training can improve both stability and generalization of GANs in data-efficient scenarios. Specifically, we explore how different techniques for gradient control can mitigate overfitting and improve stability, providing novel insights into the relationship between gradient behavior and GAN performance. This thesis is organized into five main chapters. The first four chapters tackle different aspects of gradient control in GANs and the final chapter applies gradient insights from the preceding four to effectively fine-tune large foundation models. Chapter 1 identifies the challenges of discriminator overfitting and unstable training in Generative Adversarial Networks (GANs) under limited data conditions, highlighting the need for effective solutions to these issues. Chapter 2 introduces a manifold learning approach that constrains the discriminator's capacity, helping to control gradient magnitudes and improve GAN stability. Chapter 3 presents a novel regularization technique using noise modulation on discriminator feature to prevent discriminator overfitting, and using consistency regularization to reduce gradient explosion and stabilize training. Chapter 4 proposes a Lipschitz-constrained normalization technique that harnesses generalization benefits from Batch Normalization (BN), while also prevents gradient explosion issues of BN during training, demonstrating its effectiveness in enhancing generalization while maintaining computational efficiency. Chapter 5 extends theories from Chapters 3 and 4, introducing a novel consistency regularization that significantly enhances generalization and prevents knowledge forgetting when fine-tuning large foundation models. In conclusion, this thesis demonstrates that gradient control is pivotal for improving GAN stability and generalization, as well as enhancing transfer learning generalization, particularly in limited-data settings. Moreover, the insights gained from this research have broader implications for other machine learning methods, highlighting the importance of managing gradients to enhance generalization and convergence across various domains.
Open Access
X-ray Scatter In 3D Microscopes
Chen, Yizhou
X-ray computed tomography (XCT) is a non-destructive imaging technique that enables 3D internal structure examination of objects. This is achieved by employing X-rays to penetrate the object of interest, and these X-rays are measured by a detector to form images. XCT is very useful in medicine and industrial inspection \cite{ou2021}, it can be used to examine an aircraft's part to ensure that there is no internal fracture. However, such an object would introduce \textbf{object scatter}, reducing image contrast, resulting blurry images that prevent us from seeing the small fractures. This is because some X-rays straight from the X-ray source are deflected or scattered by the object to become secondary X-rays that reach detector regions not initially aligned with these primary X-rays. In this work, we demonstrate a method to measure the spatial distribution of object scatter and subsequently remove it. We used an array of small tungsten spheres to block X-rays from reaching the detector, creating shadow regions on the detector. An array of these spheres is called a Beam-Stop-Array (BSA). By design, only X-rays scattered from the object can reach these shadow regions, thereby contributing extra counts in the shadow regions. Assuming object scatter distribution is spatially slowly variant, we can obtain object scatter distribution with extra counts in shadow regions by interpolation. During the project, we found that X-rays can also scatter: (i) within the source and from the cabinet wall of the XCT system; (ii) within the detector. We called these confounding factors as they can also contribute extra counts in the shadow regions, thereby reducing the accuracy of the object scatter measurement. After addressing these confounding factors, the extra counts in the shadow regions are solely due to object scatter. In our experiment, after measuring and correcting for object scatter, the root-mean-square-error (RMSE) of the attenuation profile of the image of an acrylic cylinder and the theoretical profile drops from $2.84\times10^{-2}$ to $2.07\times10^{-2}$. This drop in RMSE indicates that our method successfully increased the accuracy of images.
Open Access
Revisiting ‘Merantau’: A Study on Gender Dynamics among Indonesian Diasporas
Zatalini, Annisa Shabrina
Merantau, traditionally a practice for the Minangkabau men to seek opportunities beyond their matrilineal society, represents a deeply gendered migration experience. Rooted in the enculturation process of Minangkabau males, merantau embodies hegemonic masculinity, where men are expected to assert independence, ambition, and economic success. While historically focused on male mobility, modern interpretations of merantau encompass a broader range of migration experiences, including those of women, challenging traditional gender roles within diaspora communities. This study adopts a qualitative approach, drawing on semi-structured interviews with Indonesian diasporas in Canberra and analysis of diaspora-produced literature. Fieldwork conducted in October 2024 highlights how Indonesian men and women in Australia negotiate and renegotiate their gender roles within both professional and domestic spheres. The findings reveal that Indonesian women are increasingly embracing leadership and decision-making roles, reflecting shifts in gender dynamics. However, the persistence of hegemonic masculinity is evident, as men’s domestic contributions are continuously perceived as sacrifices, and they strive to maintain their identity as primary breadwinners. Enculturation within the Indonesian diaspora ensures the preservation of traditional values, yet it also fosters an environment where gender roles are continuously negotiated. This study underscores the complexity of gender negotiations within the Indonesian diaspora, as women assert greater agency and men adapt to evolving domestic expectations. The research amplifies the voices of Indonesian migrants in Australia, contributing to a nuanced understanding of migration, identity, and gender in a transnational context, while highlighting the ongoing influence of cultural heritage in shaping gendered experiences of migration.
Open Access
Artificial Intelligence Generated Content on Spatio-Temporal Modeling
(2025) Tang, Bowen
Spatio-temporal relationships are fundamental to many real-world phenomena. Spatio-temporal modeling is also integral to Artificial Intelligence Generated Content (AIGC). Spatio-temporal modeling is essential for various applications, such as motion prediction and video restoration. In this thesis, we address two distinct yet deeply interconnected tasks: hand motion prediction and turbulence-degraded video restoration. To address these tasks, we aim to investigate the following challenges: 1) how to develop efficient and effective models for spatio-temporal data while extracting and disentangling meaningful information across time and space, 2) how to leverage the decoupled motion patterns to predict diverse and physically plausible motion sequences, and 3) how to maintain visual structure consistency throughout the entire video. In the first part of this thesis, we establish a robust spatio-temporal modeling framework for observed motion sequences. This framework is designed to predict diverse yet plausible future motions. A key challenge in motion prediction lies in decouplin motion patterns from observed sequences and consistently leveraging these patterns for accurate prediction. To address this, we propose PromptFDDM, which utilizes future motion data to enhance spatio-temporal modeling. The training process consists of two stages. In the first stage, real future motion data guides the Spatio-Temporal Extractor Network (STEN ), a framework designed to extract spatio-temporal dependencies and motion patterns. In the second stage, Diffusion Models (DM), a class of generative models that iteratively refine data distributions through a denoising process, generate reference data that sulbstitutes real future motion during inference, further refining the STEN's predictions. Additionally, prompt learning is introduced in the second stage to help the DM better capture and utiize spatio-temporal patterns from observed motion sequences, thereby improving the quality of the generated inference data. In the second part of this thesis, we focus on extracting and disentangling the stable visual structure from turbulence-degraded video. The goal is to restore the visual structure and mitigate the degradation effects. A critical challenge in turbulence mitigation tasks lies in maintaining temporal consistency to prevent unstable and blurred visual artifacts. To address this challenge, we propose TMamba, a novel framework that capitalizes on the linear complexity of state-space models to process spatio-temporal data efficiently. The core component of TMamba is the Spatial/Channel-Temporal (SCT) block, a core architectural component designed to process spatio-temporal data. This block comprises two specialized modules: the Spatio-Temporal State-Space (STSS) module, which captures long-range dependencies in spatio-temporal data, and the Channel-Temporal Fusion Attention (CTFA) module, which models interactions across temporal and channel dimensions. These modules ensure comprehensive modeling of spatio-temporal dependencies and channel-temporal interactions. Furthermore, a weight-sharing technique is implemented to reduce model complexity and enhance scalability. This approach provides an effective solution to the challenges associated with processing spatio-temporal data in turbulence mitigation tasks. In conclusion, this thesis addresses two critical tasks in spatio-temporal modeling: PromptFDDM for hand motion prediction and TMamba for turbulence mitigation. Despite their distinct applications, both methods share a common foundation in spatio-temporal modeling, which is efficient and effective in extracting and disentangling meaningful information from spatio-temporal data. They demonstrate significant advancements in addressing spatio-temporal modeling challenges within their respective domains. Furthermore, they provide valuable insights and contribute to the progress of both motion prediction and video restoration.
Embargo
Birds and Bugs and Bites (oh my!): Applied epidemiology in One Health
(2025) Gilbert, Michaela
I completed my Master of Philosophy in Applied Epidemiology (MAE) at the Commonwealth Department of Agriculture, Fisheries and Forestry (DAFF) in Canberra, ACT. The Department of Agriculture, Fisheries and Forestry focuses on a range of issues of importance to Australian primary producers and the public, including biosecurity, agricultural production, and trade. Along with State and Territory Governments, DAFF has strong connections with industry partners, ranging from industry representatives such as the Meat and Livestock Association to commercial producers such as SunPork, one of the largest pork suppliers in Australia. The Department of Agriculture, Fisheries and Forestry also has strong relationships with advisory bodies such as Wildlife Health Australia (WHA) and Animal Health Australia (AHA), and with university research partners such as the University of Melbourne. My projects focused on animal diseases and diseases of One Health significance. I described the effects of Japanese encephalitis virus on Australian piggeries, I evaluated the Northern Australian Quarantine Strategy component of the National Avian Influenza Wild Birds surveillance program, and I investigated an outbreak of very virulent infectious bursal disease in Solomon Islands chickens. This thesis also describes other experiences I had during the MAE, including two field trips I went on to support two of my projects, my teaching experiences, my involvement in working groups and my role coordinating the Veterinary Epidemiology and Surveillance Throughout Australia network.
Open Access
Linking Alpha-Synuclein and the Immune System in the Aetiology of Parkinson's Disease
(2025) Parkinson, Rebekah
Parkinson's disease, a motor-related neurodegenerative disorder, is characterised by the significant role that a-synuclein and immune activation play in driving disease progression. a-synuclein, an inherent protein prone to misfolding and aggregation is identifed in pathological forms in the central and peripheral nervous system of patients with Parkinson's disease. The immunogenicity of a-synuclein-derived epitopes has been discovered to drive both innate and adaptive immunity in prodromal patients. Despite these findings, the precise mechanistic role of the immune system in contributing to Parkinson's disease remains unclear, largely due to limitations with current animal models. The present study hypothesises that the relationship of a-synuclein and immune activation, as observed in preclinical cases, plays a causal role in Parkinson's disease. To confirm this, Chapter 1 presents the establishment of the first immune-induced murine model of Parkinson's disease whereby immune activation to a-synuclein triggers dopaminergic cell loss and motor deficits. By means of immunising mice to a-synuclein, the model consists of a peripheral injection of a-synuclein protein in adjuvants in wildtype mice. This effectively triggered immune activity identified by raised white blood cell populations, including monocytes, neutrophils, and lymphocytes. alpha-synuclein immunised mice exhibited significant behavioural and neurological alterations compared to control conditions (i.e., sham, and non-injected mice). To discern the trigger of these symptoms, the induction procedure was repeated by injection of a specific alpha-synuclein epitope known to elicit an antigenic response to CD4+ T cells in Parkinson's patients. These mice similarly developed neurological changes in the substantia nigra, including significant decrease of dopaminergic cell density, greater colocalisation of a-synuclein aggregates and increase in proinflammatory cells such as microglia and astrocytes. These changes were supported by levodopa-sensitive deficits in locomotion and gait kinematics. As a dopamine replacement agent, restoration by levodopa supports a Parkinsonian phenotype specific to dopamine loss. The mechanisms underpinning the phenotype of the model are explored in Chapter 2. Specialised analysis of the immunological features in alpha-synuclein immunised mice revealed that CD4+ T cells and neutrophils play a key role in pathogenesis. Induction in a-syn-/- and Rag1-/- (lymphocyte-deficient) mice failed to induce Parkinsons-like symptoms, suggesting that the modelled phenotype is reliant on endogenous a-synuclein and adaptive immune activity. Furthermore, this study posits that an optimal threshold of immune response is required, as neither muted nor boosted immune activation was sufficient to induce or exacerbate symptoms, respectively. In Chapter 3, findings from a cross-sequential analysis of Parkinson's disease patient data supported the hypothesised pathological interplay of immune activation and a-synuclein. Positive correlations were found between a-synuclein seed amplification assays and inflammatory markers in the proteomics and blood chemistry of prodromal Parkinson's disease patients. These immune signatures were observed throughout the disease course and correlated with symptom severity. However, these signatures were uniquely expressed in patients prior to diagnosis suggesting that immune activation is involved in precipitating disease onset. The present study uncovers a causal link between immune cells and a-synuclein in driving Parkinson's disease pathogenesis. This provides the basis for exploring preventative and therapeutic interventions in the first murine immunological model of Parkinson's disease. Ultimately, the model and it's mechanistic insights in early Parkinson's, equips future research with new avenues for the identification of non-invasive Parkinson's disease biomarkers during prodromal stages.
Open Access
Living with chronic illness and managing well : implications for primary health care policy and practice / Carmen Lee Pearce-Brown
(2015) Pearce-Brown, Carmen Lee
The prevalence of chronic conditions, such as chronic obstructive pulmonary disease (COPD), is predicted to increase as the population ages, and this creates significant challenges for health systems. Primary healthcare plays a key role in maintaining the success of Australian health outcomes. The Chronic Care Model has led redesign of healthcare, including primary healthcare, to meet the needs of people living with chronic conditions and has informed the Australian health system reform agenda. At the current time, the main Commonwealth policy lever to reduce the social and economic burden of chronic conditions such as COPD through health service delivery is assumed to reside in the general practice setting. However, there is limited evidence to support this assumption. There is a need for evidence of the role of general practice to inform health policy to strengthen primary healthcare policy and practice into the future. This thesis used a mixed methods design to investigate how people living withCOPD interact with general practice and other services to manage their illness. It used a systematic review of international literature to establish how evidence portrays the role of general practice in the healthcare of people living with COPD. Presence of this role was then sought in the construction of meaning for 'managing well' by people living with COPD. A critical interpretative lens exposed the social structures, including general practice, embedded in the reality of the everyday world of people living with COPD. The study demonstrates that spontaneous descriptions of strategies that might be understood as reflecting the Chronic Care Model are present in interviews conducted with people living with COPD and multiple other chronic conditions. However, the strategies are focused on wellness, are numerous and are developed following support from various people including healthcare professionals in general practice. Study participants presented themselves as confident, capable people who are in control of their total wellbeing. Living with a chronic condition was one aspect of their lives that was controlled in order to keep well. Confidence mediates access to healthcare because healthcare professionals are placed in the role of 'assistants7. General practice was one of many social structures influencing the way people manage their chronic condition and 'manage well'. Thus general practice plays a minor role in how people living with COPD 'manage well' At the present time Commonwealth health policy to strengthen primary healthcare is reliant, for the most part, on people attending general practice. This study indicates that opportunities to strengthen the role of general practice could be found in strengthening the way general practice links with other services. Furthermore, these results suggest there is also a lot to be gained from strengthening the primary healthcare system beyond general practice.
Open Access
Women in the Imperial Art World: Anglo-Australian Women Artists and their Transnational Careers, 1885 - 1907
(2025) Nicholls, Lara
Upon receiving a folio of drawings from the Anglo-Australian artist Sophia Sinnett in 1875, the British art critic John Ruskin replied to her with the advice, 'you must resolve to be quite a great paintress; the feminine termination does not exist, there never having been such a being...as a lady who could paint. Try to be the first'. Almost a century later in Australia, the art historian Bernard Smith opined in his foundational text, 'Australian Painting 1788-1960', 'But women, perhaps because of their status in Victorian and Edwardian society and other reasons that need not be discussed here, did not figure prominently in Australian art prior to the First World War.' Were there so few 'great' women artists in Britain in the late nineteenth century as Ruskin supposed? Were women artists invisible in Australian colonial society, as concluded by Bernard Smith? This thesis is concerned with these questions because the preconception that both British and Australian professional women artists were absent in the imperial art world clouds many accounts of nineteenth-century art history. Furthermore, ideas of 'greatness' and 'invisibility' are foundational concepts in key feminist texts which revise such assumptions. Through careful examination of both quantitative data from exhibition and museum catalogues and contextual evidence from the period, I argue that British-Australian women artists in the late nineteenth century were more visible in the imperial art world than has hitherto been recognised in accounts of British or Australian art. The lives and works of the Anglo-Australian women artists discussed in this thesis do not correspond to the notion expressed by Ruskin that the nineteenth-century woman artist was merely a 'lady who could paint'. Nor do they correspond to Smith's description of a late nineteenth-century art world devoid of women artists. Rather, the research in this thesis reveals that women in the period created considerable works of art and competed in the same imperial art world arenas as their male counterparts. Indeed, as women of empire, they were influential global art citizens.
Open Access
Going to the movies: effective leadership in the public interest in times of crisis.
(2025) Sullivan, Tim
Prominent scholars on the practice of leadership have attributed high-profile failures of leadership that have damaged public and private interests to moral and ethical failures in leadership decision-making. They have pointed to shortcomings in leadership development (LD) programs that pay insufficient attention to engaging with the humanities, particularly in ethical and moral decision making, and in developing a concept of the public interest. This dissertation demonstrates a stronger relationship between theorising on leadership effectiveness and awareness of the public interest as a philosophical concept and a political commitment than is evident in the literature. In contrast to recent work in the field, I argue that understanding leadership effectiveness is derived from closer attention to the function of leadership than to contrasting styles of leadership. Focusing on the function of leadership reveals the crucial nexus between purpose and power. Power is delegated to leaders to perform tasks that are purposeful, that is, protecting the interests of the organisations they lead and promoting its purposes. I contend that in contemporary society, leaders are accountable not only for the interests of those who are intended beneficiaries of leadership action, but also to the publics outside the organisations they lead and whose interests are affected by the leadership action whether its consequences are intended or unintended. Consequently, effective leadership in the public interest must have a working concept of the public interest, and the obligations embedded in it. I further contend that leadership is more deeply institutionalised to protect the public interest than is recognised in the extant literature, that is, the power of leaders is constrained by the system of bureaucratic and political checks and balances intended to prevent the abuse and misuse of power. That institutionalising of leadership in the public interest exists in two distinct domains: the bureaucratic and the political. The bureaucratic comprises the mechanisms used by government agencies and of governance structures that test the compliance of leadership decisions with legal statutes, codes of conduct and best practice, together with nuanced judgements of societal values and attitudes. The political comprises the institutions that represent the collective interests of the people, and the power exercised by political players over the bureaucratic processes and institutions. The framework of ethical and moral decision-making in this dissertation is based on Arendt's idea of representation, that is, of consistently and consciously bringing to mind the interests of others in decision-making. The principles presented in this dissertation are illustrated in a selection of movies interpreting leadership action relating to the US presidency in moments of crisis across four different domains of power: institutional, executive, corporate, and popular movement. The significance of this work is in establishing principles that forge a stronger understanding of the relationship between the function of leadership and effective decision-making and the concept of the public interest as a primary concern of leadership ahead of sectional or private interests.