Sydney eScholarship
Not a member yet
31878 research outputs found
Sort by
The computations of buoyant fires and fire suppression
No full textThis thesis presents a comprehensive computational study of buoyant turbulent diffusion flames and water-based fire suppression, using a novel burner configuration developed at the University of Sydney. The research aims to advance current fire modelling practices by capturing the full progression of fire development from turbulence-enhanced flame formation to suppression dynamics. A key focus is placed on resolving the complex interplay between turbulence, combustion, and suppression effectiveness, particularly under varying injection strategies.
To validate the modelling framework, benchmark simulations of Sandia Flame D were performed using the ReactingFoam solver. The results demonstrated good agreement with experimental data and established the baseline performance of the numerical approach. The study then introduces the USYD Burner, a newly developed platform featuring a recessed perforated plate that enables controlled modulation of fuel-side turbulence without altering heat release rate. Numerical simulations using ANSYS Fluent and FireFOAM were conducted under both laminar and enhanced turbulence conditions, revealing clear changes in flame structure, mixing intensity, and velocity fields
The development and piloting of a mobile health (mHealth) intervention to support patients with chronic pain to taper prescription opioid medications
No full textBackground: Clinical guidelines advise against long-term opioid therapy for chronic
noncancer pain (CNCP), yet tapering opioids is difficult without adequate support. Mobile
health (mHealth) interventions may meet a need for scalable, accessible tapering support.
Objective: This thesis describes the co-design and piloting an mHealth intervention to
support opioid tapering in CNCP.
Methods: The thesis includes (1) a scoping review of mHealth interventions for CNCP and
opioid tapering, (2) a qualitative study of patient attitudes and recommendations to mHealth
support, (3) a co-design study with 12 patients and 12 clinicians to develop the intervention,
(4) a protocol detailing study objectives and methods, and (5) a pilot randomised controlled
trial (RCT) to meet the objectives and inform improvements.
Results: The scoping review revealed scarce research on mHealth interventions for opioid
tapering in CNCP. Patient feedback supported the acceptability and potential benefit of text
message mHealth supports. The co-design study developed a video and 28-day, twice-daily
text intervention, which was considered useful, appropriate, and likely to be effective in
supporting tapering. The pilot RCT recruited 28 participants. The mHealth intervention
demonstrated acceptability and feasibility but had recruitment challenges. After four weeks,
improvements in opioid tapering self-efficacy, pain self-efficacy, and pain intensity in the
intervention group, versus the control group, demonstrated preliminary efficacy.
Implementation analysis informed recommendations for a future definitive RCT.
Conclusion: The thesis introduces one of the first mHealth interventions to support opioid
tapering in CNCP. The mHealth intervention’s acceptability and preliminary efficacy were
demonstrated, though addressing recruitment challenges and other feasibility issues
should support scaling the intervention
Novel deep learning-based methods for improved prediction and feature-learning in high-throughput proteomic and transcriptomic data
No full textThe rise of high-throughput Omics technologies has allowed researchers to measure biomolecular
species of interest en masse at the sample or individual cell level. These technologies, including bulk
and single cell transcriptomics, mass spectrometry (MS) proteomics, and other MS techniques
capable of quantifying post-translational modifications (PTMs) of proteins, produce extremely large
datasets, presenting new opportunities and challenges for data analysis. These datasets may
capture complex relationships in the regulation of genes, proteins and PTMs. However, the
development of sophisticated techniques is required both to extract this information and to overcome
pathologies and challenges that arise. Issues such as missingness, biological noise, the curse of
dimensionality, and others make these datasets non-trivial to analyse.
This thesis explores different approaches to analysing high-throughput datasets, extracting useful
information and addressing some of the challenges involved.
Chapter 2 introduces Thunderbolt, a traditional analysis pipeline which provides tools for diagnosis
and remedy of pathologies inherent to specific MS proteomics datasets; differential expression
analysis; and downstream analysis tools. The chapter demonstrates a full analysis workflow to
address a specific hypothesis and discusses approaches to dealing with dataset pathologies.
Chapter 3 introduces scCCESS, a flexible autoencoder-based framework for improving the
performance of clustering methods when applied to single-cell RNA-seq datasets by diversifying and
simplifying inputs to the chosen clustering algorithm.
Chapter 4 introduces ConGregatE-PPI, a predictive ensemble artificial neural network model which
leverages complementary information from multiple datasets to improve prediction of protein-protein
interactions in a specific biological context
Childhood respiratory and neurodevelopmental outcomes of extreme premature infants – risk and protective factors and impact on the family
No full textBackground: As survival of extremely preterm infants improves, longer term outcomes become increasingly important. Respiratory and neurodevelopmental morbidities have the most impact for later life. Beyond measuring outcomes for the individual, monitoring effects of premature birth on the family are important to be able to describe the full impact of premature birth.
Methods: Single-centre implementation studies on preventative strategies to avoid lung disease and how to treat it in early childhood are included in this thesis. A summary of changes in early clinical management on a statewide scale is described. A retrospective study on the impact of intraventricular haemorrhage [IVH] on developmental outcomes was conducted. Population data was used for data linkage to describe primary school outcomes of extremely born infants and risk factors for poor performance. A review of current literature on parental mental health and family functioning was conducted.
Results: Less invasive respiratory resuscitation may lead to improved survival. Protocolised monitoring of oxygen requirement for infants with established chronic neonatal lung disease can lead to a decrease in length of stay. Mild IVH does not impact long-term neurodevelopmental outcomes, and most extremely preterm infants sit school exams and perform above national minimum standard. Risk factors for poor school performance include in-hospital morbidity and socioeconomic vulnerability along with parental capacity. Parental mental health after hospital discharge following premature birth is mostly affected in the first few years and improves over time.
Conclusion: The neonatal community continues to explore strategies to improve in-hospital outcomes. Avoiding major complications means most patients will attend school. However, optimal care in hospital does not predict school performance of extremely preterm infants, rather parental education and capacity are the major contributors to longer term outcome
A Beta Cauchy-Cauchy (BECCA) prior for sparse signal recovery in regression and graphical models.
No full textThis PhD thesis introduces a novel Bayesian approach for variable selection in high- dimensional regression settings, along with its potential extension to learning the structure of an undirected graphical model. Our proposed method, which we call the Beta Cauchy-Cauchy (BECCA) prior, replaces the indicator variables in the traditional spike and slab prior with continuous, Beta-distributed random variable and places half-Cauchy priors over the para-meters of the Beta distribution, which significantly improves the predictive and inferential performance of the technique. Similar to shrinkage methods, our continuous analog of the Spike-and-Slab (SS) prior enables posterior exploration using gradient-based methods, such as Hamiltonian Monte Carlo (HMC), while at the same time explicitly allowing for variable selection in a principled Bayesian framework. Building on the strong performance of the proposed approach in linear regression, we apply it to logistic regression context and further extend it to structure learning in Gaussian Graphical Models (GGMs) using a regression based framework. We evaluate the frequentist properties of our model through simulations and demonstrate that our technique not only outperforms the latest Bayesian variable selection methods in linear regression, but also performs comparably or better than existing methods for variable selection in logistic regression and structure learning in graphical models. The efficacy, applicability and performance of our approach, are further underscored through its implementation on real datasets
Combining Participatory Budget and Cost Benefit Analysis: A hybrid project evaluation framework
No full textThis paper introduces a novel hybrid methodology proposes
integrating Cost-Benefit Analysis (CBA) with Participatory
Budgeting (PB) to enhance public sector decision-making. To
highlight the approach, we assume that CBA has been
conducted to evaluate twelve independent infrastructure and
service projects, establishing their economic viability.
Subsequently, PB is employed to elicit community preferences
regarding the same projects, ensuring alignment with public
values. The combined framework aims to reconcile technical
efficiency with democratic legitimacy, promoting government
outcomes that reflect both expert analysis and citizen priorities.
Our findings reveal a strong community preference for projects
that improve health outcomes and utility infrastructure, while
initiatives focused on transport electrification received
comparatively low support. These results underscore the
importance of incorporating public sentiment into investment
decisions, particularly in sectors where societal benefits may
be perceived differently than economic returns
Understanding and preventing mental ill-health, substance use, and their co-occurrence among gender and sexuality diverse young people in Australia
No full textThe primary objective of this thesis was to advance the understanding and prevention of mental ill-health, substance use, and their co-occurrence, among LGBTQA+ young people in Australia. There are significant gaps in knowledge in this area, including lack of contemporary estimates of the prevalence mental ill-health and substance use among LGBTQA+ youth and limited evidence-based prevention approaches for this population. To address these gaps, this thesis proposed four research aims: firstly, to investigate the epidemiology of mental ill-health, substance use, and their co-occurrence among LGBTQA+ young people, with a focus on population-based data (Chapters 3 and 6) and data specific to trans young people (Chapters 4 and 5); secondly, to examine associations between minority stressors and traumatic events with mental ill-health, substance use, and their co-occurrence among LGBTQA+ young people (Chapters 4, 5, and 7); thirdly, to identify school-level protective factors for mental ill-health and substance use among LGBTQA+ young people (Chapters 4 to 8); and lastly, to advance current approaches toward preventing mental ill-health and substance use among LGBTQA+ young people (Chapters 9 to 11). A range of rigorous research methodologies were employed, including sophisticated cross-sectional and longitudinal analysis of unique epidemiological datasets, systematic review and qualitative inquiry and interpretation. Key findings included highlighting the magnitude of disparities in mental ill-health and substance use experiences by LGBTQA+ young people, compared to their cisgender, heterosexual peers, and identifying promising preventative interventions that are tailored to the unique experiences of LGBTQA+ youth. Suggestions for future research, particularly relating to future epidemiological work in this field, and the opportunity for researchers to explore the preventative utility offered by school peers and staff for LGBTQA+ young people, are also discussed
CRISPR activation screens for functional genomics discovery
No full textClustered regularly interspaced short palindromic repeats (CRISPR) activation (CRISPRa) has emerged as a powerful tool in molecular biology, enabling high-throughput identification of genes whose upregulation drives specific phenotypes. This thesis explores the versatility of CRISPRa genetic screens by integrating recent advances in CRISPRa technology with a strategic approach that combines parallel and complementary screening methods. By conducting multiple screens with distinct objectives, this work highlights the advantages of diverse screening perspectives to improve our understanding of methodologies and outcomes. The primary objectives were to identify novel receptors for angiotensin II and bradykinin, uncover genetic modulators of cell survival under drug-induced stress, and discover genes that enhance lipid nanoparticle (LNP) efficacy. These aims were pursued through a series of CRISPRa screens. Key findings include the identification and validation of previously unrecognized modulators of ligand binding, protective factors against drug toxicity, and enhancers of LNP-mediated mRNA expression. This study demonstrates that performing multiple CRISPRa screens in parallel not only strengthens dataset robustness but also offers opportunities to refine screening approaches and assess variability. These findings underscore the value of multifaceted screens in receptor discovery, drug-response pathway analysis, and LNP efficacy optimization. By broadening both scope and depth, this research highlights CRISPRa’s potential to address complex biological questions and drive innovation in molecular biology
Technology and Mental Health: Exploring Attentional Bias from an experimental psychology perspective using virtual reality
No full textAttentional bias refers to preferences for specific stimuli, which can be driven by emotional states,
lifestyle, or traumatic experiences. Traditionally, research was almost exclusively monopolised by
clinical psychologists however minimal progress has occurred to develop a reliable measure for
attentional bias. This thesis used a psychophysical approach to gain insight into attentional bias with
a focus on the role of a stimulus feature, known as saliency.
Chapter 2 aimed to determine the perceived salience for images using psychophysical and
computational methods. The SV L2-norm colour model was found to be the most reliable determinant
of stimulus salience. Chapter 3 investigated whether there are perceptual differences between
alcohol or non-alcohol related images using noise paradigms. Each paradigm resulted in a different stimulus order which was unrelated to the order as found in the participants’ judgement task. Chapter
4 explored biases using free-viewing tasks, measuring eye-movements in virtual reality. Dwell times
at certain positions were found to be dependent on content (alcohol or non-alcohol). Interestingly,
spatial location biases was revealed which was not determined by semantic stimulus information.
Chapter 5 determined the efficacy of dynamic free-viewing paradigms which were presented as two
imaginary conveyer belts in motion. On average, participants with higher alcohol dependency
preferred alcohol-related stimuli. These findings could inform the development of an individualised
measure for substance use disorders
Cosmic ray probe survey data of soil surface neutron counts and associated soil analysis data across grain growing regions in NSW and QLD
No full textThis dataset includes cosmic ray probe survey data that details soil neutron counts detected using CSIRO's CosmOz Rover. Cosmic ray probes measure the flux of fast neutrons which is inversely proportional to the amount of hydrogen atoms. Water molecules are the dominant source of hydrogen atoms in soil so they can be measured close to the Earth's surface and used to infer soil moisture content.
In addition to the neutron count, the survey data also contains barometric pressure (mb), temperature (℃) and relative humidity (%). These measurements are needed to correct the raw neutron count data by removing influences from atmospheric pressure and water content above the soil surface.
Surveys were carried out across the grain growing regions in NSW and QLD, particularly Northern NSW, South-Eastern NSW, and South-Eastern QLD, and were completed over a period of 4 years from 2020-2023. Surveys were completed between 1 and 4 times within this 4 year period for each of the farms.
The mobile system has 16 capsules that each measure neutron count at 1-minute intervals and log the GPS locations of each measurement.
Survey data is in .csv format with 20 surveys completed across 8 farms.
Soil sampling campaigns were completed concurrently with the CosmOz surveys and the data is available in this collection.
Soil moisture, bulk density, soil organic carbon, and clay content was measured from each sample and was used to calibrate the neutron count from the surveys. 20 cores were taken at each farm to a depth of 30 cm.
Sampling sites were defined using a stratified random sampling scheme based on k-means clustering, where strata were defined using farm characteristics.
The available soil data for each farm is in .csv format.
The datasets are stored on the USYD-RDS at \\shared.sydney.edu.au\research-data\PRJ-soilwaternowarchive. This data has restricted access, third-parties need to contact the University of Sydney and the GRDC to access the data according to agreed terms and/or with a data supply and licence agreement. Please contact Dr Patrick Filippi ([email protected]) to request access to the data