Conference outputs

Permanent URI for this collection


Recent Submissions

Now showing 1 - 20 of 16453
  • Conference output
    Impact of System Strength and Control Parameters on the Small-Signal Stability of Grid Following and Grid Forming Inverters
    Cui, H; Bai, F; Cui, Y (2023 33rd Australasian Universities Power Engineering Conference (AUPEC), 2023)

    Inverter-driven small signal stability issue has been identified as a new stability class by global power industries, which are the bottleneck for high solar/wind power penetration. This new stability problem is mainly caused by the interactions between controllers of the converters and other power system elements. To eliminate the oscillation problems, it is crucial to develop fundamental theories to fully understand the IBR dynamic behaviours and identify the cause of oscillations. In this paper, the small-signal mathematic models are established for grid-following invertor (GFLI) and grid-forming invertor (GFMI), based on which the small signal analysis is performed on both GFLI and GFMI to study the impacts of system strength and control parameters on their small-signal stability. Small-signal analysis reveals that low system strength can lead to instability in GFLI, but the instability of GFMI happens when it connects to a grid with high system strength. It is also found that those instability issues can be resolved by tuning their control parameters. However, parameter tuning brings side effects to those systems. Therefore, a modified inner current controller is novelly developed for GFMI to maintain its stability without bringing other negative effects. The developed controller and theoretical analysis are verified in the EMT testbed built in PSCAD.

  • Conference output
    Predicting Phosphoglycerylation with Transformer Features and Deep Learning
    Chandra, A; Sharma, A; Dehzangi, I; Tsunoda, T; Sattar, A (2023 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE), 2024)

    Understanding protein sequences can advance treatments for various diseases. However experimentally obtaining this information is laborious, time-consuming, and expensive. Traditional machine learning techniques, like support vector machine, random forest and logistic regression, offer potential to fast-track this process but are sometimes limited by data complexity. Deep learning algorithms, in contrast, tend to yield higher performance. In this study, we employed a convolutional neural network to predict protein phosphoglycerylation. Features were extracted from pre-trained transformer models and compared with conventional features, such as evolutionary information and physicochemical/biochemical properties. Our results indicate significant performance improvements across all feature types, with the combination of transformerbased features and the convolutional neural network being especially effective. This methodology holds potential for other protein property prediction tasks. Our software and datasets used in this study are publicly available at

  • Conference output
    Medical decision making: Microsimulation versus cohort modelling for assessing type 2 diabetes treatments
    Lyu, Juntao; Kim, Hansoo (25th International Congress on Modelling and Simulation (MODSIM2023), 2023)

    Type 2 Diabetes Mellitus (T2DM) is associated with high rates of mortality and complications, making it challenging to evaluate new treatments. Health economic models, such as cohort models and microsimulations, are currently available to evaluate new medications. However, it remains unclear which type of modelling is more suitable for evaluating T2DM treatments. Tirzepatide is a new treatment that has demonstrated promising results for glycaemic control in diabetic patients requiring insulin. The aim of this study is to compare the differences between microsimulation and cohort modelling when assessing tirzepatide for treating type-2 diabetes. This study compared the outcomes from a microsimulation and a cohort model using the same baseline patient characteristics. The microsimulation used heterogeneous individual baseline data of 10,000 patients, while the cohort model only considered the average baseline values of the entire cohort. Both the cohort and microsimulation were based on published risk equations from the United Kingdom Prospective Diabetes Study (UKPDS OM2) (Hayes, Leal et al. 2013). These parametric proportional hazards equations predict each annual cycle’s events of mortality and diabetes-related complications based on patients’ characteristics. The microsimulation was implemented in R, and each patient’s clinical data, including glucose control and BMI values and event histories are updated and carried forward to the next annual cycle. The cohort model was implemented in Excel using time-varying annual transition probabilities which are calculated from the characteristics of the cohort. Clinical data from a recent trial comparing tirzepatide (5mg, once per week) versus insulin glargine was used to inform the glucose control and BMI changes of diabetic patients (Del Prato, Kahn et al. 2021). Results over 10-year time horizons were assessed. For the 10-year time horizon, compared to the cohort model, the microsimulation model resulted in statistically significant higher cumulative mortality rates (51.1% vs 28.9%, p<0.001) and higher cumulative rates of renal failure (14.3% vs 2.1%, p<0.001), heart failure (10% vs 3.7%, p<0.001), blindness (0.7% vs 0.02%, p < 0.001), and myocardial infarction (20.3% vs 15.3%, p < 0.001). However, the cohort model predicted a higher rate of ischemic heart disease (4.8% vs 8.3%, p<0.001) and amputation (4.2% vs 9.6%, p < 0.001). No difference in stroke incidence (6.8% vs 6.8%, p=0.978) was observed. Since the 1-year clinical trial outcomes only indicated the 5% mortality rate and the 6% cardiovascular events, this comparison between microsimulation and cohort simulation cannot conclude the exact accuracy but only the differences between the two model options. Overall, the microsimulation of health outcomes for tirzepatide resulted in significantly higher rates of mortality and most of the complications associated with type-2 diabetes when compared to the cohort model. The cohort model only predicted higher rates of ischemic heart disease and amputation. There are significant differences in estimated outcomes when comparing microsimulation and cohort modelling for the treatment of tirzapatide in T2DM. This research suggests researcher to be aware of these significant differences and the contexts of using different models, further validations with more longitudinal clinical outcomes for T2DM models are required for future research.

  • Conference output
    Integrating hip exosuit and FES for lower limb rehabilitation in a simulation environment
    de Sousa, Ana Carolina C; Freire, Joao Pedro CD; Bo, Antonio PL (IFAC-PapersOnLine, 2nd IFAC Conference on Cyber-Physical and Human Systems CPHS 2018, 2019)

    Lower-limb rehabilitation for spinal cord injury (SCI) and other motor disorders is often a lengthy process for the patient. The combination of active orthoses and functional electrical stimulation (FES) promises to accelerate therapy outcome, while simultaneously reducing the physical burden of the therapist. In this work, we propose a controller to a hybrid neuroprosthesis (HNP) composed of a hip orthosis and FES-controlled knee motion. In our simulation analysis using a detailed musculoskeletal model, we use experimental data from an able-bodied subject during slow-speed walking to compare the performance provided by such a system. Furthermore, we analyzed the obtained results in comparison to gait data collected from experiments where we used an active hip orthosis. Although the knee stimulation controller still oscillated during gait, we acquired control results with errors smaller than five degrees. Besides, we were able to examine the performance at very slow speeds.

  • Conference output
    Ethical risk in the Offshore Blue Economy Integrity System
    Breakey, Hugh; Sampford, Charles (ASME 2023 42nd International Conference on Ocean, Offshore and Arctic Engineering, 2023)

    This paper explores the ethical risks at play in the Blue Economy integrity system, specifically the Australian offshore Blue Economy—including aquaculture and renewable energy industries. ‘Ethical risk’ refers to the risk that unethical actions will be done, or unethical consequences will result. The paper describes six ethical principles at work in Blue Economy activities. These principles are then applied to relevant stakeholders to describe standing ‘ethical concerns’. Like other industries, Blue Economy operations are housed in a larger ‘integrity system’— a mutually supportive network of ethical standards, legal norms, institutions and incentive structures. Integrity systems can facilitate publicly stated goals, and limit temptations and opportunities for wrongdoing and corruption. While different integrity system elements and institutions can be important in confronting and mitigating different ethical concerns, each depend on the proper functioning of the larger system. Ethical risks may thus be understood as the standing ethical concerns that are not well addressed by the existing integrity system. This process allows ethical risks in the offshore Blue Economy to be conceptualised, while pointing practically to integrity system reforms that might mitigate these risks.

  • Conference output
    Energy Management of AC Residential Microgrids Using Advanced Fuzzy Inference Method
    Chowdhury, NA; Yang, F; Bai, F (2023 IEEE International Conference on Energy Technologies for Future Grids (ETFG), 2023)

    This article presents an innovative Energy management control technique using Fuzzy Inference System (FIS) for an AC microgrid to reduce grid dependency. The primary focus of this article is to reduce grid power fluctuations in an AC microgrid to improve stability and resilience of the system. This microgrid considering a typical residential system that consists of a Photovoltaic (PV) system, Battery Energy Storage System, inverters, and household Load. The cost function is based on imported power from the utility grid and only considering the peak electricity rate for optimum solution. The Fuzzy analysis method shows the benefit to operate and optimise the household microgrid in peak loading condition and fully enhance the battery storage system state of charge.

  • Conference output
    Cybersecurity Defence of Synchrophasors in Distribution Systems: A Deep Learning Approach
    Zhang, G; Cui, Y; Zhang, R; Bai, Feifei (2023 IEEE International Conference on Energy Technologies for Future Grids (ETFG), 2023)

    Phasor Measurement Unit (PMU) has become a critical component for the modern distribution network, as it records high-resolution synchrophasor data which contain abundant static and dynamic information of the system. However, PMUs are vulnerable to potential cyberattacks, for example, data spoofing attacks. A deliberate PMU spoofing attack can confuse the existing data source authentication models, especially when the models are used for identifying multiple PMUs at the same time. This paper proposes a data-driven cybersecurity defence model which can identify the source information of a large group of PMUs with high accuracy. The model utilizes the inherent correlations among PMUs with a deep neural network to enhance the data source authentication performance. The effectiveness of the proposed model is examined by the PMU data collected from a real distribution network with different error metrics. Through comprehensive numerical experiments, the proposed model provides consistent superior performance in comparison with other state-of-the-art data source identification approaches.

  • Conference output
    Transition Towards Inverter-based Generation with VSG Control: Low Frequency Instability Prospective
    Feng, J; Bai, F; Nadarajah, M; Ma, H (2023 IEEE International Conference on Energy Technologies for Future Grids (ETFG), 2023)

    The proliferation of inverter-based resources (IBRs) with conventional control methods can result in significant power oscillations following disturbances. This is due to insufficient inertia and lack of damping with IBRs. To accommodate more IBRs into the system without compromising the system stability, the virtual synchronous generator (VSG) control may provide a promising solution. However, the active power control can emulate the SG rotor mechanism, the existing VSG may affect the electromechnical oscillations of power systems. This paper examines the issue with different IBR penetration scenarios on the low-frequency oscillation modes in multi-machine systems. The investigation starts with the comparisons of the impacts of IBR with VSG control integration at various locations, using eigenvalue, participation factor and nonlinear time domain simulations. Further, a comprehensive small-signal analysis is conducted to examine the changes in inter-area and local oscillation modes with an increase in IBR penetration level.

  • Conference output
    Unscented Predictive Control for Battery Energy Storage Systems in Networked Microgrids
    Wu, J; Guo, F; Yang, F; Boem, F (2024 UKACC 14th International Conference on Control (CONTROL), 2024)

    Controlling batteries State of Charge (SoC) within operational constraints, while minimising the power exchange among microgrids and with the grid, is an important problem to maximise microgrids performance and extend batteries lives. To address this problem, this paper adopts an Unscented Predictive Control (UPC) to optimise the SoC control under uncertain conditions. Based on the model of the NMG and the principle of model predictive control, the design of the SoC control strategy is formulated as an Optimisation Problem (OP) with probability operation conditions. To deal with the latter, the unscented transformation is integrated with predictive control to derive the mean value and variance of system states. A tractable OP for NMGs is then obtained and the effectiveness of the proposed UPC-based SoC control strategy is verified by simulations with different NMG frameworks.

  • Conference output
    Investigation on the Strain Effect on the Behaviour of Laminated Veneer Lumber Timber Connections
    Cheng, Xinyi; Gilbert, Benoit P; Lyu, Chunhao; Guan, Hong (7th International Conference on Geotechnics, Civil Engineering and Structures, 2024)

    This paper experimentally investigates the influence of the strain rate experienced during earthquake and progressive collapse events on mechanical properties affecting the design of softwood Laminated Veneer Lumbers timber connections. The parallel and perpendicular to grain embedment stiffness and strength, and the Mode I and II fracture energies were examined under four levels of strain rates. Results showed that the embedment stiffness and strength increased by up to 35% from the quasi-static to dynamic strain rates while the embedment ductility decreased by up to 17%. The fracture energies were found to be mostly insensitive to the investigated range of strain rates. Furthermore, the influence of the strain rate on the behaviour of timber connections is further analysed by quasi-statically and dynamically testing one connection type with two different fastener spacings. Results showed that for the connections investigated, the dynamic strength can be up to 30% higher than the quasi-static one, however, the dynamic ductility of the connections can be reduced substantially by up to 32.5%.

  • Conference output
    Conditional Backdoor Attack via JPEG Compression
    Duan, Q; Hua, Z; Liao, Q; Zhang, Y; Zhang, LY (Thirty-Eighth AAAI Conference on Artificial Intelligence, 2024)

    Deep neural network (DNN) models have been proven vulnerable to backdoor attacks. One trend of backdoor attacks is developing more invisible and dynamic triggers to make attacks stealthier. However, these invisible and dynamic triggers can be inadvertently mitigated by some widely used passive denoising operations, such as image compression, making the efforts under this trend questionable. Another trend is to exploit the full potential of backdoor attacks by proposing new triggering paradigms, such as hibernated or opportunistic backdoors. In line with these trends, our work investigates the first conditional backdoor attack, where the backdoor is activated by a specific condition rather than pre-defined triggers. Specifically, we take the JPEG compression as our condition and jointly optimize the compression operator and the target model’s loss function, which can force the target model to accurately learn the JPEG compression behavior as the triggering condition. In this case, besides the conditional triggering feature, our attack is also stealthy and robust to denoising operations. Extensive experiments on the MNIST, GTSRB and CelebA verify our attack’s effectiveness, stealthiness and resistance to existing backdoor defenses and denoising operations. As a new triggering paradigm, the conditional backdoor attack brings a new angle for assessing the vulnerability of DNN models, and conditioned over JPEG compression magnifies its threat due to the universal usage of JPEG.

  • Conference output
    Preserving Privacy of Input Features Across All Stages of Collaborative Learning
    Lu, J; Xue, L; Wan, W; Li, M; Zhang, LY; Hu, S (2023 IEEE Intl Conf on Parallel & Distributed Processing with Applications, Big Data & Cloud Computing, Sustainable Computing & Communications, Social Computing & Networking (ISPA/BDCloud/SocialCom/SustainCom), 2023)

    Collaborative learning is a widely used privacy-preserving distributed training framework where users participate in global training using gradients instead of disclosing their private data. However, gradient inversion attacks have challenged the privacy of this approach by reconstructing private inputs from gradients. While prior works have proposed various defenses against gradient inversion attacks, their privacy assessments have mainly focused on untrained models, lacking consideration for the trained model, which should be the primary focus in collaborative learning. In this context, we first conduct a comprehensive privacy evaluation across all stages of collaborative learning. We uncover the limitations of existing defenses in providing sufficient privacy protection for trained models. To address this challenge, we introduce GradPrivacy, a novel framework tailored to safeguard the privacy of trained models without compromising their performance. GradPrivacy comprises two key components: the amplitude perturbation module, which perturbs gradient parameters associated with critical features to thwart attackers from reconstructing essential input feature information, and the deviation correction module, which effectively maintains model performance by correcting deviations in model update directions from previous rounds. Extensive evaluations demonstrate that GradPrivacy successfully achieves effective privacy preservation, surpassing state-of-the-art methods in terms of the privacy-accuracy trade-off.

  • Conference output
    People’s Perception and Expectation of Moral Settings in Autonomous Vehicles: An Australian Case
    Rafiee, Amir; Breakey, Hugh; Wu, Yong; Sattar, Abdul (International Conference on Computer Ethics, International Conference on Computer Ethics: Philosophical Enquiry 2023, 2023)

    While Autonomous Vehicles (AVs) can handle the majority of driving situations with relative ease, it is indeed challenging to design a system whose safety performance will fit every situation. Technology errors, misaligned sensors, malicious actors and bad weather can all contribute to imminent collisions. If we assume that the wide-spread use and adoption of AVs is a necessary condition of the many societal benefits that these vehicles have promised to offer, then it is quite clear that any reasonable ethics policy should also consider the various user expectations with which they interact, and the larger societies in which they are implemented. In this paper we aim to evaluate Australian’s perception and expectation on personal AVs relating to various ethical settings. We do this using a survey questionnaire, where the participants are shown 6 dilemma situations involving an AV, and are asked to decide which outcome is the most acceptable to them. We have designed the survey questions with consideration for previous research and have excluded any selection criteria which we believed were biased or redundant in nature. We enhanced our questionnaire by informing participants about the legal implications of each crash scenario. We also provided participants with a randomised choice which we named an Objective Decision System (ODS). If selected, the AV would consider all possible outcomes for a given crash scenario and choose one at random. The randomised decision is non-weighted, which means that all possible outcomes are treated equally. We will use the survey analysis, to list and prioritise Australian’s preferences on personal AVs when dealing with an ethical dilemma, that can help manufacturers in programming and governments in developing AV policies. Finally, we make some recommendations for further researchers as we believe such questionnaires can help arouse people’s curiosity in the various ways that an AV could be programmed to deal with a dilemma situation and would encourage AV adoption.

  • Conference output
    Single Sequence Based Feature Engineering for Convolutional Neural Networks Towards RNA Contact Map Prediction
    Rashid, MA; Paliwal, KK (2023 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE), 2023)

    Features are crucial for deep learning models as they help understand complex data, learn patterns, and make accurate predictions. Feature engineering can sometimes be computationally expensive, but it can speed up the training and inference phases of the deep learning models. By providing a more concise and informative representation, it reduces the number of parameters and computations required in the down-stream operations. Well-chosen features can enhance a model's ability to represent data in a structured and meaningful way. They help learn hierarchical dependencies in data, reduce the dimensionality of the input data, and generalize from the training data to unseen data to make the model robust. In this work, we present a self-supervised learning model for feature generation from RNA sequences towards applying in deep learning models for RNA contact map prediction. We test the efficacy of our extracted features by comparing the prediction performance with the prediction performance obtained by the features extracted using the state-of-the-art language foundation model, RNA-FM. We found our approach promising.

  • Conference output
    Water and energy integration towards post-mine transition
    Clerc, C; Sharma, V; Pagliero, L; Abbasi, B; Parshley, J; Fourie, A; Tibbett, M (16th International Conference on Mine Closure, 2023)

    Mining is resource-intensive and often located in remote and climate-fragile areas with limited resources, requiring complementary developments in water treatment and power generation services to sustain its activities (Owen & Kemp, 2018). Communities that surround mine sites usually rely on the mining company for their water and energy supply. Since many jurisdictions require mining companies to close all service infrastructure as part of closure planning, mine closure raises concerns over the future sustainability of this mining-dependent infrastructure. Mine closure, therefore, can disrupt critical access to water and energy for local communities and impact their post-mining transition plans and development opportunities (Christmann, 2017; Vivoda et al., 2019). Interest in exploring the socio-economic impacts of mine closure has been increasing worldwide. Yet, research that considers how mine closure intersects with local needs to align sustainable planning is sparse (Bainton & Holcombe, 2018). Understanding the nature of water and energy challenges induced by mine closure in remote and climate-vulnerable regions will help guide communities to successfully transition post-mining (Carvalho, 2017). As such, an integrated approach to water and energy access and security is needed. Integrated planning of water and energy access allows the optimal use of resources, enabling a sustainable supply post-mining. Silo-ed interventions that promise a clean and affordable supply of energy without considering its interface with, and role in, sustainable water management, have been ineffective in achieving sustained benefits over the long term (United Nations et al., 2014). An integrated approach not only helps balance resource quality and quantity for productive livelihoods but, in doing so, optimises the cost-efficiency of solutions while minimising associated social and environmental impacts (De Oliveira et al., 2022). This paper presents an exploratory study of what an integrated multi-dimensional analysis for water and energy access post-mine may look like. The novelty of our approach is its grounding in inter-disciplinary literature encompassing diverse disciplines: mining and mine closure, development challenges in remote contexts, climate change adaptation, and integrated energy and water regional-scale planning.

  • Conference output
    Reducing Object Detection Uncertainty from RGB and Thermal Data for UAV Outdoor Surveillance
    Sandino, J; Caccetta, PA; Sanderson, C; Maire, F; Gonzalez, F (2022 IEEE Aerospace Conference (AERO), 2022)

    Recent advances in Unmanned Aerial Vehicles (UAVs) have resulted in their quick adoption for wide a range of civilian applications, including precision agriculture, biosecurity, disaster monitoring and surveillance. UAVs offer low-cost platforms with flexible hardware configurations, as well as an increasing number of autonomous capabilities, including take-off, landing, object tracking and obstacle avoidance. However, little attention has been paid to how UAVs deal with object detection uncertainties caused by false readings from vision-based detectors, data noise, vibrations, and occlusion. In most situations, the relevance and understanding of these detections are delegated to human operators, as many UAVs have limited cognition power to interact autonomously with the environment. This paper presents a framework for autonomous navigation under uncertainty in outdoor scenarios for small UAVs using a probabilistic-based motion planner. The framework is evaluated with real flight tests using a sub 2 kg quadrotor UAV and illustrated in victim finding Search and Rescue (SAR) case study in a forest/bushland. The navigation problem is modelled using a Partially Observable Markov Decision Process (POMDP), and solved in real time onboard the small UAV using Augmented Belief Trees (ABT) and the TAPIR toolkit. Results from experiments using colour and thermal imagery show that the proposed motion planner provides accurate victim localisation coordinates, as the UAV has the flexibility to interact with the environment and obtain clearer visualisations of any potential victims compared to the baseline motion planner. Incorporating this system allows optimised UAV surveillance operations by diminishing false positive readings from vision-based object detectors.

  • Conference output
    Detecting Malicious Blockchain Transactions Using Graph Neural Networks
    Jeyakumar, Samantha Tharani; Eugene Yugarajah, Andrew Charles; Hóu, Zhé; Muthukkumarasamy, Vallipuram (Distributed Ledger Technology 7th International Symposium, SDLT 2023, 2024)

    The adoption of blockchain technology within various critical infrastructures is on the rise. Concurrently, there has been a corresponding increase in its misuse, primarily through the exploitation of its pseudo-anonymous characteristic. Encouraging blockchain adoption and improving security in the decentralised environment require techniques to detect wallets and/or smart contracts owned by malicious entities. Illegal activities such as dark market trades, money laundering, and receiving unlawful payments are performed by connecting various wallets or smart contracts in a meticulous way. A graph can be a potential representation to visualise such interconnections via various patterns, and graph-based data may represent the topological structure of the blockchain network. Recently, Graph Neural Networks (GNN) have been widely used for analysing the structure of complex networks and identifying patterns. This is the first work that considers a generalised graph representation for the Bitcoin and Ethereum networks and analyses their behaviour using a combination of heterogeneous GNN framework’s GraphSAGE and Graph Attention Network (GAT). The classification results reveal that the proposed approach modestly improved Bitcoin network analysis, whereas Ethereum smart contract analysis needs further investigation in terms of incorporating other aspects of smart contracts, such as code-base, byte length, and lifetime features.

  • Conference output
    Revisiting variety ratings for ratoon stunting disease
    Ngo, CN; Gibbs, L; Garlic, K; Wei, X; Bhuiyan, SA; Allsopp, PG (44th Annual Conference of the Australian Society of Sugar Cane Technologists (ASSCT 2023), 2023)

    Variety resistance ratings are provided by Sugar Research Australia for a number for sugarcane diseases through SPIDNet. These diseases rely on resistance as little can be done in regard to their management. However, for ratoon stunting disease (RSD), one of the most important diseases in the Australian industry, this has not been the case in recent years. For RSD, the term “resistance” is somewhat misleading and can impart a false sense of security if viewed as for other diseases. All known varieties can acquire RSD, which can lead to yield losses and also in turn act as an infection source. Enzyme-linked immunoassays (EB-EIA) were previously used to estimate Leifsonia xyli subsp. xyli (Lxx) bacteria titres and to assign a resistance rating. Quantitative polymerase chain reaction (qPCR) offers a more specific and sensitive method of detecting Lxx. The qPCR method can also provide more accurate estimates of Lxx titre. This study compared the leaf sheath biopsy (LSB) sample type in the plant crop and the xylem-extract sample type in plant and first-ratoon crops with the aim of developing a variety rating method.

  • Conference output
    Shifting Attitudes and Uncertain Futures: The Endangerment of Sinasina Sign Language (Papua New Guinea)
    Rarrick, Samantha (Foundation for Endangered Languages (FEL) XXIII/2019: Causes of Language Endangerment, 2019)

    Sign languages of Papua New Guinea (PNG) are currently underreported and underdocumented. Yet, like many spoken languages of this region, these languages are highly likely to be endangered. Here, I present findings from ongoing work with a small and endangered sign language of the highlands of PNG, Sinasina Sign Language (SSSL). Our attempts to apply existing frameworks for assessing vitality for this language uncovered issues related to (i) language endangerment evaluations as imprecise predictors of language loss; and (ii) changing attitudes towards deafness and sign language use within the Kere community where SSSL is used. This paper addresses these attitudes, possibilities for the vitality of this language if at least one deaf Kere child is born soon, and implications for evaluating language endangerment globally. Recommendations for adjustments to existing frameworks are also recommended to better address diverse language situations by overtly including potential for new users and likelihood of language shift. I conclude by addressing implications for the goals of evaluating endangerment and by arguing that endangerment situations should not be viewed as static; they are constantly changing as language users and circumstances change, often in unpredictable ways.

  • Conference output
    A rapid method of screening sugarcane clones for resistance to red rot
    Bhuiyan, SA; Natarajan, S; Garlick, K; Rapmund, A; Eglinton, J (44th Annual Conference of the Australian Society of Sugar Cane Technologists (ASSCT 2023), 2023)

    Red rot, caused by the fungus Colletotrichum falcatum, is an important sugarcane disease. It is the most damaging disease in some South Asian countries causing significant yield and crop losses. In the Sugar Research Australia (SRA) variety-development program, advanced clones in the final stage of selection are screened for resistance to red rot before being released for commercial production. The conventional method of red rot rating in the field using 12-month-old mature cane is time-consuming, labour-intensive and is impacted by the weather and crop lodging. A rapid method of screening for resistance to red rot was developed (two-eye-setts method) and compared to historical ratings obtained from a conventional method. In addition, two other methods, controlled condition testing (CCT), a method used in India, and a leaf midrib inoculation method were tested. In the two-eye-setts method, six clones with known ratings were inoculated with red rot culture through holes made in the middle of the two-eye-setts and incubated at 30°C and 90% relative humidity for 2 weeks. Inoculated setts were split longitudinally and visually assessed for symptoms using standard disease indices and photographed. The images were analysed using the machine-learning algorithm Classification and Regression Tree to estimate the percentage of symptomatic pixels as image cover. Symptom expression was poor in the setts inoculated using the CCT method. The leaf midrib method showed no differences among the inoculated leaves. The two-eye-setts method produced excellent symptoms in all inoculated setts, and visual indices of disease showed strong correlation (r=0.99) with the historical ratings of clones. Image cover correlated strongly with disease indices (r=0.93) and historical red rot rating (r=0.88). The two-eye-setts method along with image analysis can substantially shorten the time required for screening for red rot s from over a year to about 3weeks. This approach will be implemented to screen for resistance to red rot in the SRA variety-development program.