Sydney eScholarship
Not a member yet
31878 research outputs found
Sort by
The Effect of Nocebo Education on Socially-Induced Nocebo Cybersickness in Virtual Reality
No full textThe nocebo effect occurs when aversive physiological symptoms are elicited psychologically, typically via negative health-related expectations. Social modelling is a key mode of inducing nocebo effects. Despite this, there is a paucity of research into interventions to reduce social modelling of negative symptoms. The current study investigated nocebo education as a potential intervention to attenuate socially-induced nocebo cybersickness in a VR paradigm. Nocebo education involves explaining the nocebo effect to participants prior to a treatment or procedure known to cause negative symptoms. Additionally, the effects of social modelling and nocebo education on two proposed mechanisms of the nocebo effect, expectancy and anxiety, were investigated. In line with predictions, social modelling increased cybersickness, an effect that was mediated by increases in expectancy and anxiety. However, contrary to hypotheses, nocebo education had no significant effect on cybersickness. Moreover, nocebo education increased participants’ expectancy and anxiety. This study makes several contributions to the literature. Firstly, the social modelling effect is replicated. Secondly, the role of expectancy and anxiety as key mechanisms underlying social modelling are demonstrated, an area where, despite evidence for these mechanisms in other modes of nocebo induction, research is sparse. Finally, it is demonstrated that nocebo education is not appropriate for inhibiting social modelling of negative symptoms, due both to its lack of effects on cybersickness and its expectancy and anxiety-inducing effects. This is not only empirically significant; it is also pertinent for clinical practice, where nocebo education has been promoted but negative social influences on health are widespread. Recommendations for future research are made. Specifically, other interventions that may be applied to social modelling, such as latent inhibition, are recommended. Moreover, the need to directly compare the relative strength of different modes of nocebo induction, and the respective effects of nocebo education on each, is emphasized
The effect of rotation speed and flow rate on evacuation of particles from a spinning dry powder inhaler capsule
No full textThis study investigated a capsule’s powder evacuation behaviour when rotating about its
minor axis in a cross flow and considering the effects of rotation speed and flow rate on powder
emission. The experimental platform, an optically accessible capsule chamber, was designed
to uncover the independent effects of these variables by enabling high-speed imaging of the
powder evacuation.
The capsules were rotated at three speeds (1500, 2500 and 3650 RPM) and two constant
flow rates, 30 SLPM and 60 SLPM (inlet velocity: 16.67 m/s and 33.33 m/s, respectively).
Two powders were selected: a lactose carrier, Respitose (SV010, D50 = 104 μm) and Mannitol
(D50 = 7 μm), the latter representing pure active pharmaceutical ingredient formulations that
form agglomerates. In addition to imaging, the capsule was weighed before and after each
device actuation to quantify powder emission.
Increasing the flow rate was found to have the largest impact on the mass emitted from the
capsule at all rotation speeds. The emitted mass for all cases was highly variable and influenced
by the cohesiveness of the powder and subsequent blockage of the capsule aperture.
The potential for blockage was more pronounced for mannitol at the high rotation speeds.
Emitted dose over time was modelled using a natural logarithm function to describe the rate
of emptying and demonstrate the advantage of increased flow rate and favourability of low/-
moderate rotation speeds. The study of powder size distribution during evacuation found no
significant difference between flow conditions for mannitol, as dispersion was dominated by
shearing at the capsule aperture
Optimising Thermal Comfort in The Classroom: High schools in the hot and humid tropical climate of Indonesia
No full textThermal comfort significantly influences students' health, cognitive abilities, and academic performance. This thesis explores strategies to optimise thermal comfort in Indonesian classrooms, where dissatisfaction with thermal conditions is prevalent. These classrooms rely on natural ventilation, offering limited control over the indoor environment. Clothing insulation is crucial for individual comfort, but mandatory school uniforms restrict adaptations. Female Muslim students face additional challenges due to modesty requirements, exacerbating discomfort in Indonesia's hot and humid climate.
The research was conducted in three phases. First, thermal manikin tests assessed the thermal performance of 11 uniform configurations. Results showed that a loose, untucked long-sleeve shirt paired with trousers and a hijab reduced clothing insulation by 0.39 clo compared to the standard uniform, significantly improving thermal comfort.
The second phase used simulations to analyse the impact of classroom design features, such as natural ventilation, on thermal performance. The results indicated that cross-ventilation in top-floor classrooms enhanced air movement in both highland and lowland coastal climates. In contrast, single-sided ventilation was more effective for first-floor classrooms in Surabaya's coastal climate. Orienting top-hung windows at 45° to the prevailing winds further improved natural ventilation performance.
Finally, students' thermal comfort responses revealed that uniform modifications had a greater impact than building adjustments. Uniform changes reduced the Standard Effective Temperature (SET) by 3°C, increasing annual comfortable hours to 83%. Building modifications, such as optimising ventilation, provided smaller improvements. The study recommends climate-responsive uniforms, enhanced natural ventilation, and fan integration as essential strategies for creating thermally comfortable learning environments in Indonesia
DEEP LEARNING FOR REMOTE STATE ESTIMATION UNDER PHYSICAL LAYER ATTACKS
No full textWith the rapid developments in sensing and communication technologies, cyber-physical systems (CPSs) have been widely applied in various engineering fields, playing a crucial role in the era of Industry 4.0. CPSs typically integrate spatially distributed plants, sensors, machines, and controllers to achieve the desired performance of physical processes. However, the integration of communication networks in CPSs makes them susceptible to malicious attacks, such as denial of service (DoS) attacks. These attackers not only disrupt communication networks but also pose significant threats to physical systems. Therefore, it is essential to design efficient control approaches to ensure the security of the CPSs in the presence of such attacks. Conventional security systems often lack effectiveness against these sophisticated attackers, whereas machine learning (ML) techniques have shown great promise in various cyber-security applications. This thesis focuses on addressing remote state estimation problems under physical layer DoS attacks using deep learning algorithms
Towards therapeutic modulation of the epigenome in females with X-linked liver disease using Adeno-Associated Viral vectors
No full textAllele-specific reactivation of healthy gene copies on the inactive X chromosome (Xi) has the potential to restore expression of absent proteins for therapeutic benefit as a method of treating X-linked disorders. Elimination of the master regulator XIST RNA may be required for success, however subsequent chromosome-wide Xi reactivation may lead to dosage imbalance of X-linked genes. Other approaches such as the recruitment of transcriptional activators using CRISPRa at the target loci offers the advantage of targeted effect. Here, we utilised our recently published novel model system, chimeric mouse-human FRG livers produced from hepatocytes derived from a heavily skewed female patient liver explant with the X-linked metabolic liver disorder, OTC deficiency, to investigate the effects of XIST RNA elimination and recruitment of a transcriptional activator in vivo. We observed that deletion of XIST in female patient hepatocytes in vivo, led to reactivation of the healthy OTC gene from the Xi. However, we also observe that elimination of XIST RNA causes potentially abnormal consequences on the epigenome environment of the human X-chromosome. To overcome the challenges and limitations associated with targeting the entire X chromosome, a locus-specific strategy was also explored to target the OTC promoter through the recruitment of a transcriptional activator. This strategy of modulating the epigenome induced expression of OTC in vitro and proved that the locus is amenable to editing, however no transcriptional activation was detected in the patient derived primary human hepatocytes. Additional analysis elucidated challenges inherent in delivering epigenome editing reagents efficiently to primary human cells in an in vivo setting. This thesis lays early foundations for overcoming hurdles associated with translating liver-targeted epigenome editing strategies from bench side to bedside, through bridging the gap between the gene therapy and the epigenome editing fields
Guitars not Greeks: The early chitarrone and its use in the 1589 La Pellegrina intermedii
No full textToday, the early chitarrone is assumed to be the same in form and tuning as the later designated instrument, the tiorba, a fundamental part of which is a neck extension to a second pegbox to allow for an extended bass string range. Foundational research into the nascence of the chitarrone has been dominated by related research into the origins of monody, with a focus on the influence of contemporaneous research into Ancient Greek practice; an influence which has since been disputed. Early chitarrone research is hampered by a complete lack of organological or iconographical evidence; however, the centrality of the chitarrone within the development of monody and basso continuo demands closer attention.
Clear descriptions by Alessandro Piccinini on the application of his arcilituto neck extension to the chitarrone sometime after 1595 have been contested or ignored but are here reexamined in the context of the chitarrone’s first documented appearance in the 1589 La pellegrina intermedii.
This thesis aims to critique research that retrospectively applies to the early chitarrone the characteristics of a chronologically later codified form in the tiorba. A framework of practice for the early chitarrone will be formulated, with particular attention being given to the underlying impetus for the creation of the new instrument. We contextualise the vocalists and instrumentalists that participated in the intermedii with contemporaneous collaborative instrumental performance practices, with a focus on the guitar. This new approach to the origins of the instrument forms the basis for speculation on the form and tuning of the early chitarrone.
The implications of the influence of the chitarrone as a collaborative instrument on vocal performance practice and notational practices, such as monody and basso continuo, will be explored
Musculoskeletal Pain in Children and Adolescents: Insights into Risk and Prognosis
No full textMusculoskeletal (MSK) pain is common in children and adolescents, with many experiencing persistent symptoms into adulthood. Despite advances in MSK research in adults, understanding of risk and prognosis in younger populations remains limited, impeding effective intervention development. Spinal pain, a subset of MSK pain, is thought to arise from interactions between biomechanical, psychological, and social factors. This thesis investigated sedentary behaviour as a potential predictor of spinal pain through three studies: a systematic review with meta-analysis, a cross-sectional analysis of adolescents with spinal pain, and a longitudinal causal analysis of adolescents who developed chronic spinal pain. These studies found only weak associations and no causal relationship between sedentary behaviour and spinal pain, with low certainty of evidence. The findings challenge the assumption that sedentary behaviour is a major contributor to spinal pain in adolescents and suggest that interventions targeting sedentary time alone are unlikely to be effective. Effective interventions require a thorough understanding of MSK pain prognosis to identify and target children and adolescents at higher risk of developing chronic, high-impact pain. This thesis also conducted a Cochrane review on MSK pain prognosis and a feasibility study tracking the clinical course of spinal pain in adolescents. The prognosis of MSK pain in youth is poorly understood, with substantial variability in recovery and persistence. Limited high-quality evidence highlights the need for robust longitudinal research. Weekly tracking of spinal pain progression in adolescents via SMS is highly feasible, with strong response and retention rates. However, recruitment challenges suggest future studies should expand clinician participation and explore alternative recruitment strategies. Future research should prioritise longitudinal studies addressing the complex biopsychosocial contributors to MSK pain in youth
A habit persistence model of multiple discrete/continuous demand for evaluating charging behaviour of Australian electric vehicle owners
No full textThis paper introduces a novel habit persistence model of discrete/continuous demand that allows the joint evaluation of the spatial (i.e., location) and temporal (i.e., time of day) dimensions of the charging decision-making process. The
model’s habit persistence structure further captures established recharging routines that influence both when and where charging occurs. The proposed model is applied to data capturing weekly charging activities collected using an online survey disseminated to a sample of EV owners recruited from across Australia between February and March 2024. Results show that charging at home is the most prevalent behaviour, with a strong tendency towards daytime charging largely driven by households with access to residential solar panels.
Workplace charging emerges as a viable alternative to home charging when employers provide free charging and
commuting frequency is high. The model also reveals the presence of state dependencies in charging behaviour,
indicating that past choices are likely to influence current charging patterns. The empirical findings are subsequently
used to demonstrate how changes in electricity prices can shift charging demand and impact grid load, corroborating the
importance of targeted policy interventions to manage the growing energy demand for EVs
Robust Phishing URL Detection Through Deep Learning and Domain Shift Mitigation
No full textThis thesis addresses the challenge of phishing URL detection by focusing on domain shift to improve performance. We show that state-of-the-art classifiers struggle when URLs differ from their training datasets and identify features with distribution shifts through statistical analysis. To address this, we propose an Unsupervised Domain Adaptation (UDA) framework that aligns features between source and target datasets, enhancing detection accuracy. Additionally, we leverage the reasoning capabilities of large language models (LLMs) to develop a one-shot phishing URL classification framework, demonstrating improved performance under domain shifts. Finally, we integrate these advancements into a federated learning framework, enabling secure, distributed training on private datasets from multiple organisations while overcoming domain shift and leveraging LLMs' contextual understanding
An Intact Tissue Analysis Pipeline for Quantifying Brain Pathology in Preclinical Research
No full textHistology is a foundational technique for assessing disease phenotypes and therapeutic interventions. In mouse models of Alzheimer’s disease (AD) and Parkinson’s disease (PD), it is used to visualise and quantify hallmark pathologies such as amyloid plaques in AD, and dopaminergic neuron degeneration in PD. Accurate quantification is essential for deriving meaningful biological insights and enabling clinical translation. While stereology is considered the gold standard for quantification, its reliance on thin tissue sections that incompletely represent 3D structures can lead to inaccurate estimates of pathology.
This project developed an optimised pipeline for the visualisation and absolute quantification of pathological features in intact mouse brain tissue, using amyloid plaques and dopaminergic neurons as exemplars. Two tissue clearing methods, CLARITY and iDISCO+, were compared, and iDISCO+ was selected for subsequent studies. Amyloid plaques and dopaminergic neurons were successfully visualised in cleared specimens. A deep learning-based 3D analysis pipeline integrated with brain atlas registration was optimised to segment amyloid plaques from volumetric light-sheet imaging data. Plaque load measurements from the 3D pipeline were validated against 2D stereology and showed strong correlation. Notably, 3D analysis revealed regional differences in plaque load not detected in 2D, and offered advantages in time, cost, and scalability. The utility of the 3D pipeline was demonstrated in a study evaluating the neuroprotective potential of saffron in a mouse model of AD. Saffron-treated mice exhibited significantly lower hippocampal plaque load (p < 0.01) compared to controls.
In summary, this intact tissue analysis pipeline enables the absolute quantification of amyloid plaques in the intact mouse brain, overcomes limitations of 2D histology, and provides a framework for 3D quantification of other pathologies in preclinical research