48080 research outputs found
Sort by
The use of geometric morphometries to improve conservation outcomes of a ‘sugar glider’ Petaurus cryptic complex in Australia and New Guinea
No full textUnderstanding the geographic distribution and morphological variation of species is crucial for biodiversity conservation, particularly in regions with complex biogeographic histories such as Australia. Despite some recent research, the distribution and morphological characteristics of many Australian terrestrial species remain ambiguous, often due to unresolved taxonomy. This thesis focuses on the Petaurus genus, specifically Petaurus breviceps, Petaurus notatus and Petaurus papuanus to address gaps in their range boundaries by identifying morphological variation of the skull. The study employed geometric morphometries and linear measurements to analyze skull size and shape, providing new insights into species distribution and evolutionary responses to geographic and ecological pressures.The first chapter of this thesis examines the distribution of Petaurus breviceps and Petaurus notatus across three key regions: the eastern, middle, and western areas of the Great Dividing Range (GDR), as well as north and south of the Brisbane Valley Barrier (BVB). These biogeographical barriers have long been hypothesized to influence species distribution. Significant differences in skull size and shape were found between P. breviceps populations located in the mid-GDR and southern eastern regions. However, no significant shape differences were observed between mid- and west-GDR populations. Additionally, skull size and shape differed significantly between north and south eastern specimens, though northern populations showed no significant intra-regional shape differences. These findings suggest that P. breviceps is primarily restricted to the eastern coastal zone south of the BVB, with hybridization likely occurring in the northern regions. This study refines the distribution of P. breviceps, emphasizing the role of geographic barriers in shaping species ranges and informing conservation strategies.The second chapter focused on unresolved taxonomic classifications within the Petaurus genus, particularly for populations in Cape York and New Guinea. Historically, Petaurus breviceps from northeastern Queensland has been considered a distinct taxon. Recent revisions recognized three subspecies at the species level: P. notatus, P. ariel, and P. papuanus (the latter distributed across New Guinea). Using geometric morphometries, we compared known Australian species with specimens from Cape York and New Guinea. Our analysis revealed no significant shape differences between Cape York specimens and P. notatus, supporting their identification as P. notatus (formerly P. longicaudatus). Conversely, significant shape variation was found between Cape York and New Guinea populations, indicating no close association with P. papuanus. Further, island specimens from New Guinea exhibited significant morphological divergence from the mainland specimens. This variation likely reflects evolutionary responses to isolation and novel environmental pressures. Mainland New Guinea populations, however, showed no significant morphological differences, suggesting overlapping distributions and possible hybridization. These results contribute to the growing body of evidence for the complex evolutionary history of the Petaurus genus and highlight the need for continued taxonomic and biogeographic research.The third chapter of this thesis investigated the introduction of Petaurus notatus to Tasmania in the 1830s, an event that provides a unique opportunity to study phenotypic plasticity and its ecological implications. Translocated as pets, P. notatus established a wild population that has since been associated with predation on endemic and migratory birds. Using geometric morphometric techniques, we assessed whether the Tasmanian population exhibited morphological changes in response to the novel environment. Our findings reveal significant skull shape variation between Tasmanian and Victorian P. notatus specimens. Notably, Tasmanian specimens exhibited increased skull size, with a pronounced elongation of the rostrum and dorsal shift in the frontal bones. These changes were accompanied by an increase in the posterior angle of the coronoid process on the mandible. Interestingly, these morphological changes occurred without a corresponding increase in overall body size. The observed phenotypic changes in Tasmanian populations may be indicative of a dietary shift towards camivory. Similar morphological adaptations have been documented in other species undergoing dietary transitions. This case study underscores the importance of monitoring introduced species and their potential impact on native ecosystems.These findings also contribute to the growing body of evidence for the complex evolutionary history of the four species within the Petaurus genus and highlight the need for continued taxonomic and biogeographic research. It provides a comprehensive reassessment of the geographic distribution and morphological variation for ‘sugar gliders’. By integrating geometric morphometric analyses with biogeographical data, the research highlights the importance of geographic barriers in shaping species distributions and emphasizes the evolutionary responses of isolated populations. The findings have implications for conservation management, particularly in refining species distributions and identifying populations at risk of hybridization. Future research should focus on the genetic underpinnings of the observed morphological variation and explore the potential ecological consequences of dietary shifts in introduced populations. This thesis contributes to the broader understanding of species distribution, evolution, and conservation in the unique and diverse ecosystems of Australia and New Guinea.</p
Novel piezo-controlled ATR-FTIR microspectroscopy for in-situ monitoring of electrochemical reaction in battery models
No full textAttenuated total reflection Fourier transform infrared (ATR-FTIR) technique has become indispensable for surface-specific molecular analysis. At the Australian Synchrotron's Infrared Microspectroscopy (IRM) beamline, we have advanced this technique by developing a novel piezo-controlled ATR-FTIR device designed for in-situ monitoring of electrochemical reactions in battery models. This piezo-controlled ATR-FTIR system incorporates high-precision piezoelectric linear translation stages, enabling sub-micron positioning and a gentle approach to engage samples with step intervals as small as 50 nm. By capturing high-quality spectral data during charge–discharge cycles of zinc ion batteries (ZIBs) at a controlled 100 nm distance from the electrode surface, the system overcomes common spectral artifacts associated with traditional reflectance setups and provides genuine interfacial chemical information without disrupting ongoing reactions. Combining the unique ability to monitor interfacial chemistry with its precision and reproducibility, this piezo-controlled ATR-FTIR device expands the analytical potential of synchrotron-FTIR microspectroscopy, offering transformative insights into the formation of solid electrolyte interphase and solvation mechanisms. The applications in ZIBs demonstrated in this study highlight the capability of the piezo-controlled ATR-FTIR technique for understanding critical interfacial processes that underpin energy storage performance and catalysis research, setting a new standard for synchrotron-FTIR studies of dynamic interfacial phenomena.</p
Toward Adaptive and Robust Multi-Agent Collaboration in Open and Dynamic Environments
No full textThe abstract for this item has not been populated.</p
Enhancing Privacy and Security in Cloud-Based Machine Learning
No full textMachine Learning (ML) has been widely applied across various domains, including computer vision (CV), natural language processing (NLP), automatic speech recognition (ASR), and recommender systems (RS). The success of ML models largely depends on data availability and computational resources. For instance, ChatGPT , a widely used large language model, was trained on an extensive dataset derived from books, websites, and other text sources, encompassing approximately 570 GB of text data and containing a total of 175 billion parameters. Training such a model requires massive computational power, involving thousands of GPUs and consuming weeks or months of training time. Therefore, training a ML model is a challenging for users who lack access to sufficient data or computational resources. To address this issue, cloud-based ML - also known as ML as a Service (MLaaS) - has emerged as a scalable solution, enabling users to leverage pre-trained models or train custom models using cloud infrastructure.MLaaS offers two primary services: inference services (IS) and training services (TS). In IS, a client sends data to a cloud server hosting a pretrained model and receives prediction results. In TS, a client outsources model training to a cloud server with high computational capacity and obtains the trained model. Despite its advantages, MLaaS presents significant privacy and security challenges. In IS, client data often contains sensitive information; thus, exposing it to the server raises privacy concerns. In TS, the outsourced training process is vulnerable to backdoor attacks, where a malicious server implants hidden functionalities into the model, raising security concerns.While extensive research has been conducted on privacy and security in MLaaS, several research gaps remain. For privacy concerns in IS, prior works propose secure inference (also known as oblivious inference) methods, enabling clients to obtain predictions without exposing plain data to the server. Cryptographic techniques such as homomorphic encryption (HE) and secure multi-party computation (MPC) have been widely adopted in secure inference. However, these approaches suffer from high computational and communication overheads. This thesis introduces more efficient protocols that achieve a better balance between computational cost and communication efficiency for secure inference. For security concerns in TS, existing research extensively examines backdoor attacks in CV, NLP, and ASR, but their impact on RS remains unexplored. Furthermore, the recommender systems as a service (RaaS) paradigm is increasingly adopted, where e-commerce companies outsource RS model training to the cloud (e.g., Amazon Personalize). This thesis investigates backdoor attacks in RaaS and proposes robust mitigation strategies to enhance its security.By addressing both privacy and security challenges in MLaaS, this thesis contributes to a robust framework for mitigating privacy risks in IS and security threats in TS, making safer and more trustworthy MLaaS.</p
Mechanical Behaviour of Strata under the Coupled Influence of Fluid Flow and Geo-Stress in Underground Mining
No full textGroundwater significantly affects the safety and stability of underground mining. It can activate strata fractures and weaken rock masses. Mining activities may trigger the release of water from high-pressure water-bearing strata, leading to stress in overburden strata, increasing risks of mining disasters. This study addresses the engineering challenges related to the coupling of high-pressure water and geo-stress in underground coal mines, focusing on the hydro-mechanical interactions that govern fracture development, water inflow, and strata instability. The effects of water on weakening coal and rock, along with the influence of pore water pressure on their mechanical behaviour, were investigated. The mechanisms of high-pressure water release in overburden aquifers during mining, and its interactions with the stress and fracture fields were examined. This research aims to enhance the understanding of groundwater and geo-stress coupling mechanisms and offers valuable insights for addressing real-world challenges in actual engineering cases. The study primarily focuses on the following aspects:Using three-dimensional Computed Tomography (3D-CT) scanning technology, the water distribution characteristics within coal specimens during the saturation process were studied. Uniaxial compression tests were then conducted to assess the mechanical properties and bump proneness of coal specimens at various degrees of water saturation. Additionally, overburden sandstone was subjected to water immersion and uniaxial compression tests. The results show that coal specimens exhibited greater strength and higher bump proneness in a dry state. Saturated coal retained significant bump proneness despite reductions in strength and Young’s modulus. Sandstone specimens show higher sensitivity to water, with pronounced changes in mechanical properties upon saturation.An improved discrete element numerical (DEM) method was developed to simulate the hydro-mechanical coupling process in fractured rock masses. This enhanced modelling framework incorporates fracture development, dynamic evolution of pore pressure, and two-way interactions between seepage and stress fields, enables more accurate simulations of internal seepage behaviour in rock masses with complex fracture networks. An engineering model based on one of the Australian coal mines was then constructed to investigate the effects of high-pressure water on the stress field and hydro-mechanical coupling behaviour of the overburden rock as the longwall coal face advances. The results indicate that high-pressure water in the water-bearing strata increases tensile stress in the lower overburden rock, potentially promoting coal roof instability and fracture development. However, water pressure influence decreases as confined water is progressively released during mining.Based on actual engineering cases, a mining model was developed for the longwall working face with a high-pressure water-bearing strata, and the evolution of the overburden stress field, flow field, fracture field, and water inflow during the advancement of the longwall face was studied. The results indicate that high-pressure water intensifies stress concentrations in the strata and accelerates fracture development. In turn, the formation of fractures facilitates the release of confined water, significantly lowering water pressure in the water-bearing strata near the fracture zone. Numerical trends closely matched field observations, with water inflow initially low but increasing significantly after periodic overburden failure.A 2D numerical model examined fault activation mechanisms and their impacts on strata stability and water inflow. The research reveals that when the working face nears the fault, fractures caused by high compressive stress concentration in front of the face interact with the fault, activating its water-conducting capacity. This creates a continuous water-discharge path between the fault and the overlying water-bearing strata, resulting in a surge in water inflow at the working face and forming the peak water inflow during the mining process.The outburst propensity of coal seams under coupled gas pressure and geo-stress was investigated using DEM simulations for both “normal” and “rapid” longwall drivage conditions in an Australian coal mine. The study incorporated variations in gas pressure, presence of gas pockets, and geological faults. Results indicate that elevated gas pressures significantly weakened coal-rock masses, increasing outburst risks, especially near faults and gas pockets. Rapid mining reduced roof deformation but still showed instabilities near high-pressure gas zones.This study advances critical knowledge on the coupling of high-pressure water and geo-stress in underground coal mines, their effects on strata stability, water release, and fracture development. The research offers valuable insights for improving mining safety, managing groundwater inflow, and optimising mining operations to mitigate potential mining hazards.</p
A Pharmacist-led Intervention to Support Type 2 Diabetes Self-Management in Community Pharmacy Settings of Pakistan
No full textBackground: The increasing prevalence of Type 2 diabetes mellitus, along with the growing number of patients with poor glycaemic control, is a significant public health concern, especially for developing countries such as Pakistan. The surge in T2DM cases in Pakistan has put the current resource-deficient healthcare system under pressure, urging researchers to seek alternative approaches for better disease management and improved quality of life. There is evidence that self-management educational support can facilitate T2DM patients to assume an active and participatory role in their health, thereby enhancing disease-related outcomes. Community pharmacists (CPs) are vital in supporting T2DM self-management. However, their role in developing countries is underexplored, and there needs to be more evidence on the effectiveness of and normalisation of service in routine practice once it is implemented in Pakistan's community pharmacy settings. To fill this knowledge gap, this study was conducted in three phases. A systematic review of RCTs assessing the effectiveness of community pharmacist-led interventions on glycaemic control and self-management behaviours was conducted to inform study design. The objective of phase I was to map any association between T2DM self-management behaviours and clinical and psychosocial outcomes. Phase II aimed to implement and evaluate the CP-led diabetes self-management educational intervention program, and phase III examined the feasibility, acceptability, and sustainability of the intervention program in community pharmacy settings of Pakistan.Methods: This was a multi-method study. Phase I was a cross-sectional study involving T2DM patients over 18 years old who could comprehend the Urdu language and were willing to participate. Participants were recruited from community pharmacies located in Lahore, Pakistan. Phase II was a pre- and post-interventional, quasi-experimental study without a control group. The diabetes self-management educational intervention was delivered by trained CPs practicing in community pharmacies in Lahore, Pakistan. The intervention was delivered via 04 face-to-face sessions and followed up through monthly telephone calls for 06 months. The effectiveness of the program was measured by changes in diabetes knowledge (DKQ), adherence to diabetes self-management behaviours (SDSCA), HbA1c level, patient-reported height and weight-based BMI, diabetes quality of life-brief clinical inventory (DQoL-BCI), and problem area in diabetes (PAID-5). The quantitative data of Phase I and Phase II were analysed using SPSS 27. Phase-III intervention program was further evaluated through semi-structured interviews with 20 participants (T2DM patients and CPs) who completed the phase-II. The questionnaire was used in both phases I and II. Phase I data were analysed using t-tests, one-way ANOVAs, and regression analyses. Phase II data were examined by paired-sample t-tests. In phase III, the audio-recorded interviews were professionally transcribed and thematically analysed.Results: Phase I involved a cohort of 260 patients; most of them were male (54.6%), aged 41-60 years old (63.1%), and from low-income families (53.8%).The mean HbA1c level was 7.96%, ranging from 6.2 to 12.3%, and most participants (78.8%) had poor glycaemic control. The participants had moderate diabetes-related knowledge, with a mean score of 13.6 (out of 24) and a standard deviation of 2.5. Most of the participants had mild to poor adherence to self-management behaviours, average quality of life, and high diabetes-related emotional distress. Regression analysis revealed that married individuals from high-income families with family support and high diabetes knowledge were more likely to adhere to T2DM self-management behaviours. We also found that for every unit, an increase in adherence to diabetes self-management behaviours (specific diet, medication adherence, general diet, and exercise) was significantly associated with a 1.26 to 1.43 times higher likelihood of having good glycaemic control and better quality of life. Similarly, higher adherence to general diet (OR = 0.71, 95% CI 0.56 – 0.89, p = 0.01), exercise (OR = 0.77, 95% CI 0.66 – 0.91, p = 0.01), medication uptake (OR = 0.73, 95% CI 0.61 – 0.86, p Conclusions: The findings from the cross-sectional study revealed significant gaps in diabetes knowledge, self-management behaviours, and psycho-social outcomes which established a need for diabetes self-management educational intervention. The results of the experimental study highlighted that the intervention program effectively improved adherence to diabetes self-management behaviours and diabetes-related outcomes. Qualitative exploration from patients and pharmacists during Phase III confirmed the acceptability of the intervention and highlighted contextual factors that affect its sustainability and scalability. Overall, our findings offer a solid base to support the formal inclusion of pharmacists in managing chronic diseases in low-resource countries like Pakistan. Future research should focus on economic analyses and large-scale, randomised controlled trials to implement the program nationally.</p
Genus <i>Gobiodon</i>: Taxonomy, Group structure and Relationship with host <i>Acropora</i> coral Habitat Complexity
No full textGobiodon are a genus of obligate coral dwelling cryptic fish with a mutualistic relationship with scleractinian corals, making them an integral component of coral reef functioning. Despite this, there are significant gaps in knowledge regarding their taxonomy, identification and the relationship between sociality and host coral characteristics. The thesis aimed to address these knowledge gaps by providing formal descriptions of two species of Gobiodon, a cohesive field identification guide to the coral-dwelling fish genus Gobiodon and two studies on the relationship between sociality and the structural complexity of their host corals using experimental and observational studies.The literature review provides the background regarding the role of habitat structural complexity on the sociality of animals, in particular, coral dwelling fishes. The first chapter contains a revised description of the genus Gobiodon, formal descriptions of two undescribed species of Gobiodon and taxonomic notes regarding the Southern Pacific Ocean population of the newly described species Gobiodon spadix. This first chapter makes use of morphological and genetic techniques, including morphometrics, meristics, x-ray photography and phylogenetic analysis, to provide supporting evidence for species descriptions. The second chapter aims to enable the efficient identification via inclusion of reviews of the known species of Gobiodon, summarised in a field guide format for coral reef surveys and studies. The guide uses the diagnostic features of each species, known coral hosts, known reef habitats, geography, sociality and known genetic relationships to provide a description of each species alongside photographs and illustrations. Additionally, a field identification key, occurrence table with accompanying map and colour plate of all species illustrations is provided. The third chapter involves the quantification of host Acropora coral complexity characteristics and their relationships with Gobiodon social structure (solitary individual, pair or group). Findings suggest that there is a significant difference between the corals occupied by solitary individuals/pairs and groups. Solitary individuals and pairs were observed in smaller corals with higher levels of structural complexity. However, groups were observed occupying larger corals with lower levels of structural complexity. The fourth chapter focuses on investigating the relationship between Gobiodon quinquestrigatus and the structural complexity of the coral host using a manipulative aquarium experiment. The study utilised 3D printed coral structures to accurately manipulate coral complexity and size to investigate the differences in coral occupancy across individuals of differing social status. The experiment highlighted the importance of social status and coral complexity on coral choice, which has important ramifications for the formation and persistence of sociality under altered environmental conditions.Overall, the thesis fills the gaps in knowledge that surround the genus Gobiodon and their relationship with the mutualistic host corals of the genus Acropora. Coral reefs are predicted to be exposed to increased severity and frequency of catastrophic events, such as tropical storms and mass coral bleaching, therefore understanding how Gobiodon and their coral host interaction has never had greater significance than it does in the current day. The thesis adds to a growing body of work on the genus that aids in their continued use as a model species for sociality and potential barometer for coral reef ecosystem health, helping us understand coral reef ecosystems before they are irreversibly altered by the changing nature of the natural world due to human impact.</p
Factors Influencing Bullying of Adolescents on the Autism Spectrum in Mainstream Secondary Schools
No full textBullying of adolescents on the autism spectrum in mainstream secondary schools is an international, complex, and pervasive problem. Adolescents on the autism spectrum are at greater risk of bullying than either their neurotypical peers, or students with other diagnoses. This is particularly problematic considering the growing numbers of students on the spectrum who are attending mainstream schools, and the detrimental short and long-term impacts of bullying. Despite the frequent occurrence of bullying in schools, limited research has been conducted to examine the ways that school environments and systems may influence bullying of adolescents on the autism spectrum.The aim of this study was to provide greater understanding regarding the ways that student-specific and contextual factors within schools, and within ecological systems around schools, influence bullying of adolescents on the autism spectrum in secondary school contexts. Questions posed related to the ways that these factors connected and interacted to place students at greater risk of bullying, or to help protect them from bullying when it occurred. A single case study design, drawing on multiple perspectives and observations, was used to explore participants’ experiences and perceptions of bullying, students’ characteristics, and most importantly, school environments and systems. Bronfenbrenner’s ecological systems theory served as the theoretical framework for the study, enabling an examination not just of students, but of the interactions between students on the autism spectrum, school environments, and the greater ecological systems surrounding students and schools.</p
Learn to Live Again: A Pilot Study to Support Women Experiencing Domestic Violence
No full textPurpose: The prevalence of domestic violence is increasing, and it is becoming more common for women who have experienced domestic violence to access support programs in their community. Learn to Live Again (L2LA) is an eight-week therapeutic program facilitated by Barnardos, which is provided through community support services in NSW, Australia. The program is designed for women who have experienced or continue to experience the traumatic effects of family and domestic violence. Methods: The pilot study involved collecting survey data from former participants and semi-structured interviews with current participants and facilitators. Data collection occurred between June and August 2023. Data collection included both face-to-face and online versions of the program. Results: All participants reported positive experiences of L2LA. The main benefits of the program for participants included connecting with women in similar situations, learning skills and strategies to cope with their experiences, sharing their lived experiences of domestic violence, and reconnecting with their children. Facilitators also had positive experiences of facilitating the program and observing the positive changes women experienced. Although, challenges were also identified and included managing the group dynamics, small group sizes, and managing trauma experiences in a group setting. Conclusion: The results indicate participants as well as facilitators had positive experiences of L2LA. Participants valued the program and felt that it helped them to begin the healing process and improve their overall wellbeing. Women were very passionate about recommending the program to other women. Facilitators observed many positive changes in the women throughout the program. However, L2LA challenges included the small group size, managing the dynamics of the group, and the range of trauma within the group as well as children being present. A larger evaluation of this program is required to confirm these findings.</p
Real-time lithology identification while drilling based on drilling parameters analysis with machine learning
No full textAccurate formation lithology information is crucial for addressing post-mining issues. Artificial intelligence is increasingly vital for lithology identification but faces challenges in underground coal mines, especially in accurately interpreting lithology from drilling parameters. These challenges include: the influence of drill string friction, difficulties in extracting valuable data from large datasets, insufficient real-time performance to guide drilling operations, and the limited adaptability of individual machine learning algorithm. These issues hinder the practical application of machine learning-based lithology identification methods in coal mines. To address these challenges, this study developed a smart drilling rig and established an automatic data acquisition system. A novel method for data acquisition, retrieval, and cleaning was proposed to enable real-time data collection, rapid retrieval, and standardized data processing workflows. An automatic drill string friction sensing method was introduced to acquire real-time friction data during drilling, which corrected the drilling parameters and enhanced their correlation with stratigraphic information. Furthermore, an ensemble learning lithology identification model based on soft voting was constructed, integrating SVM, decision tree, KNN, and neural network classifiers. Field tests were conducted in an underground coal mine in China. During the tests, the system completed four boreholes and collected over 100,000 sets of drilling data. The lithology identification model was established, achieving an accuracy of 98.79%. The results demonstrate that the drilling parameters, eliminating the influence of drill string friction, showed a stronger correlation with stratigraphic lithology. The proposed lithology identification model exhibited high accuracy and generalization capability.</p