116320 research outputs found
Sort by
From Technology Idea to Value Proposition
No full textStrategic use of managerial tools might simplify technology entrepreneurs’ struggle in developing value propositions for new technologies. This chapter introduces the Technology Application Selection framework and Business Model approaches that can help managers and entrepreneurs to turn their technology-based ideas into value propositions. By doing so, they speed up their process of technology commercialization
Can an ‘informed’ general population sample be comparable to a patient sample? A case study of preferences for chemotherapy induced peripheral neuropathy
Get PDFBackground: In health care preference studies, a general population sample may be the only viable option. However, they lack the understanding of treatment/care of patient samples. This study investigated the impact of providing extra information on general population comprehension of discrete choice experiment (DCE) choice sets. Preferences were compared between an informed and ‘naïve’ general population sample and a patient sample. This was investigated in the context of eliciting preferences for features of a chemotherapy induced peripheral neuropathy (CIPN) assessment tool. Methods: A general population sample was randomised to two arms. Arm 1 (N = 167) received written information and some pictures about CIPN and Arm 2 (N = 168) received extra information in the form of a short video and moving images. These responses were compared to a patient sample (N = 117) that received the same information as Arm 1. All respondents completed 8 choice sets each. Results: Arms 1 and 2 of the general population sample had no preference differences, although respondents in Arm 2 had an easier time identifying differences between assessment options than those in Arm 1. The patient and general population sample had overlapping preferences for some attribute parameters, while differences were more in terms of strength of preference rather than differences in preferences. Conclusions: Extra information can improve general population understanding of DCE choice sets. However, it was not found to bring general population preferences closer to the patient sample. This has implications when considering willingness to pay by patients versus general population
Decarbonising Electricity
No full textThe current shift to renewable energy is dominated by globalised energy companies building large-scale wind and solar plants. This book discusses the consequences and possibilities of this shift in India, Germany, and Australia, focusing on regions which have now largely decarbonised electricity generation. The authors show how centralised models of energy provision are maintained, and chart their impacts in terms of energy geography, social stratification, and socio-ecological appropriation. The chapters emphasise the prominent role played by state regulation, financial incentives, and public infrastructure for corporate renewables, arguing that public provision should be re-purposed for distributed renewables, social equity in affected regions, and for wider social benefit. This interdisciplinary book provides fertile building ground for research in - and application of - future energy transitions. It will appeal to students, researchers, and policy makers from anthropology, sociology, politics and political economy, geography, and environmental and sustainability studies
Artificial Intelligence Use in Primary Care: Attitudes, Concerns, and Readiness among Health Professionals in Metropolitan Australia
No full textBackground: Artificial intelligence (AI) holds promise for improving efficiency and decision support in primary care. However, little is known about how primary care professionals in Australia, particularly within metropolitan regions, currently perceive and use AI tools. Objective: To evaluate the use of and attitudes toward AI among general practice clinicians and staff in Australia, including familiarity, confidence, perceived benefits and concerns, policy awareness, and readiness for adoption. Methods: A cross-sectional survey was conducted among general practitioners, non-GP specialists, allied health practitioners, nurses, and administrative staff in Australian primary care settings. The questionnaire assessed participants’ experience, familiarity and confidence using AI (rated on 5-point scales), concerns, policy awareness, understanding of AI bias, willingness to adopt AI, and preferred areas of application. Descriptive statistics and exploratory subgroup comparisons by role and experience were performed; these were descriptive observations given the small sample size. Results: A total of 39 primary care professionals were recruited. Overall familiarity with AI was low (mean ratings ~2/5), and self-rated confidence in using AI tools was modest. The most cited concerns were data privacy, AI errors or “hallucinations,” and integration challenges. Awareness of official AI policies was limited. Around two-thirds acknowledged AI’s potential biases. Notably, 64% expressed willingness to incorporate AI into practice; none explicitly refused. Participants saw AI as most useful for drafting documentation, handling administrative tasks, and monitoring follow-up. Conclusion: Primary care professionals in a metropolitan region of Sydney, Australia, show cautious optimism toward AI. While familiarity is limited, many are open to using AI for streamlining tasks. Addressing data security, reliability, and system integration concerns will be key, along with increasing education and supportive policies
Profiling of Burkholderia pseudomallei variants derived from Queensland's clinical isolates.
Get PDFBurkholderia pseudomallei (Bp), an environmental bacterium and opportunistic pathogen endemic to tropical regions, is highly adaptive and thrives in diverse environments, from soil to human hosts. Bacterial adaptation is critical for survival, virulence modulation, and persistence during infection and can manifest as colony morphotype variation (CMV). Although Bp adaptation has been studied, CMV remains poorly understood. Here, we characterized five clinical Bp isolates exhibiting heterogeneous populations with rough and smooth colony morphologies. We used phenotypic assays, whole-genome sequencing, and proteomics to investigate the molecular pathways reflecting CMV, by comparing smooth and rough morphotypes. Although phenotypic differences in protease activity, hemolysis, mucoidy, iron uptake, and antibiotic sensitivity-including to antimicrobial agents commonly used to treat infections-were rare, these traits alone could not distinguish morphotypes or groups of isolates. Genomic comparisons revealed either no differences or limited isolate-specific mutations, which do not explain the overall difference in phenotypes. In contrast, proteomic analysis uncovered consistent shifts in protein abundance related to virulence, including quorum sensing, DNA methylation, and secretion systems. Rough variants showed higher abundance of EPS-associated proteins, the BpsI3/R3 quorum-sensing system, and the global regulator ScmR, whereas smooth variants displayed higher abundances of proteins belonging to type III/VI secretion and siderophore biosynthesis pathways. These findings suggest that CMV is driven by phase variation and regulatory mechanisms rather than punctual genomic modifications. Our study underscores the limitations of phenotype or genome-based classification alone in the context of CMV and highlights the value of integrated multi-omics approaches to uncover CMV-associated biomarkers, with potential applications in diagnostics and the development of targeted therapies against persistent and drug-resistant Bp infections.IMPORTANCEBurkholderia pseudomallei (Bp), the causative agent of melioidosis, is endemic to Australia, Asia, Africa, and the Americas. It predominantly affects Indigenous populations and individuals suffering from diabetes, chronic lung or kidney disease, or alcoholism. Bp is known for its exceptional genomic and phenotypic plasticity, enabling rapid adaptation to diverse environments. This adaptability is reflected by colony morphotype variation (CMV), including reversible phase variation between smooth and rough colonies. In this study, we report rough and smooth colonies from clinical samples and emphasize the importance of characterizing CMV through multi-omics approaches rather than relying solely on genomics and phenotypic traits. By integrating genomic, phenotypic, and proteomic data, we identified that a limited number of mutations, including one in a putative regulatory element, likely drive major molecular changes between morphotypes. These affect the expression of quorum-sensing systems, the transcriptional regulator ScmR, DNA methyltransferase, and virulence-associated genes
Novel Deep Learning Model for Glaucoma Detection Using Fusion of Fundus and Optical Coherence Tomography Images.
Get PDFGlaucoma is a leading cause of irreversible blindness worldwide, yet early detection can prevent vision loss. This paper proposes a novel deep learning approach that combines two ophthalmic imaging modalities, fundus photographs and optical coherence tomography scans, as paired images from the same eye of each patient for automated glaucoma detection. We develop separate convolutional neural network models for fundus and optical coherence tomography images and a fusion model that integrates features from both modalities for each eye. The models are trained and evaluated on a private clinical dataset (Bangladesh Eye Hospital and Institute Ltd.) consisting of 216 healthy eye images (108 fundus, 108 optical coherence tomography) from 108 patients and 200 glaucomatous eye images (100 fundus, 100 optical coherence tomography) from 100 patients. Our methodology includes image preprocessing pipelines for each modality, custom convolutional neural network/ResNet-based architectures for single-modality analysis, and a two-branch fusion network combining fundus and optical coherence tomography feature representations. We report the performance (accuracy, sensitivity, specificity, and area under curve) of the fundus-only, optical coherence tomography-only, and fusion models. In addition to a fixed test set evaluation, we perform five-fold cross-validation, confirming the robustness and consistency of the fusion model across multiple data partitions. On our fixed test set, the fundus-only model achieves 86% accuracy (AUC 0.89) and the optical coherence tomography-only model, 84% accuracy (AUC 0.87). Our fused model reaches 92% accuracy (AUC 0.95), an absolute improvement of 6 percentage points and 8 percentage points over the fundus and OCT baselines, respectively. McNemar's test on pooled five-fold validation predictions (b = 3, c = 18) yields χ2=10.7 (p = 0.001), and on optical coherence tomography-only vs. fused (b_o = 5, c_o = 20) χo2=9.0 (p = 0.003), confirming that the fusion gains are significant. Five-fold cross-validation further confirms these improvements (mean AUC 0.952±0.011. We also compare our results with the existing literature and discuss the clinical significance, limitations, and future work. To the best of our knowledge, this is the first time a novel deep learning model has been used on a fusion of paired fundus and optical coherence tomography images of the same patient for the detection of glaucoma
Hydrogen/methane explosion loads and their effects on high-performance concrete: A comprehensive review
Get PDFAs the global energy sector transitions toward sustainability, hydrogen and natural gas (methane) are emerging as pivotal fuels. However, the explosive nature of these fuels poses substantial risks, highlighting the need for precise explosion-loading predictions and robust blast-resistant infrastructure. Although high-performance concrete (HPC) and ultra-high-performance concrete (UHPC) show promise for such infrastructure, their performance under gaseous explosions remains insufficiently understood. This review consolidated current methods for predicting hydrogen/methane explosion loads and for assessing structural response of HPC/UHPC members. Experimental tests (under unconfined, semi-confined, confined, vented, and congested conditions), empirical models (TNT equivalence, multi-energy), and numerical simulations (ranging from one-step to detailed reaction CFD) were examined. Recent advancements in data-driven prediction, such as machine learning and graph neural networks, show potential for improving prediction speed. Particularly, the SALE method, a computationally efficient approach based on user-defined detonation parameters, demonstrated its ability to model a wide range of gas detonations and structural damage scenarios in hydrocodes like LS-DYNA. Key gaps include the lack of dimensionless predictive models and universal data-driven frameworks for diverse blast scenarios. Future research should focus on improving deflagration-load predictions, expanding experimental and numerical databases, and integrating advanced machine learning techniques with numerical simulations to ensure the resilience and safety of HPC/UHPC systems
Data sources:
Access Repository Dashboard
Do you manage OPUS? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!