Search Results (1,484)

The presence of Internet of Things (IoT) devices in modern working and living environments is growing rapidly. The data collected in such environments enable us to model users’ behaviour and consequently identify and authenticate them. However, these data may contain information about the user’s current activity, emotional state, or other aspects that are not relevant for authentication. In this work, we employ adversarial deep learning techniques to remove privacy-revealing information from the data while keeping the authentication performance levels almost intact. Furthermore, we develop and apply various techniques to offload the computationally weak edge devices that are part of the machine learning pipeline at training and inference time. Our experiments, conducted on two multimodal IoT datasets, show that P²ESA can be efficiently deployed and trained, and with user identification rates of between 75.85% and 93.31% (c.f. 6.67% baseline), can represent a promising support solution for authentication, while simultaneously fully obfuscating sensitive information. Full article

During red wine production, managing the pomace cap is key for a successful fermentation, allowing the extraction of phenolics and other metabolites and providing the necessary oxygen for yeast activity. In recent years, automatic cap management systems based on the injection of gases have gained popularity, despite the limited scientific information regarding the outcomes of their use. This trial aimed to evaluate the composition of wine during industrial red wine fermentations using an automatic sequential air injection system (i.e., AirMixing MI^TM). Fourteen lots of Cabernet Sauvignon grapes were fermented using four air injection regimes, where the intensity and daily frequency of air injections were set to either low or high. As expected, the treatment combining high-intensity and high-frequency air injection produced the largest dissolved oxygen peaks reaching up to 1.9 mg L⁻¹ per cycle, compared to 0.1 mg L⁻¹ in the low-intensity and low-frequency treatment. Yet, in all cases, little to no accumulation of oxygen overtime was observed. Regarding phenolics, the highest intensity and frequency of air injections led to the fastest increase in total phenolics, anthocyanins, short polymeric pigments, and tannin concentration, although compositional differences among treatments equilibrate by the end of fermentation. The main differences in phenolic compounds observed during fermentation were mediated by temperature variation among wine tanks. Based on these findings, it is advisable to keep the characterizing kinetics of phenolic extraction and expand the study to the aroma evolution of wines fermented with this technology. Full article

Background and Objective: Temperature management is key for reliable operation of permanent magnet synchronous motors (PMSMs). The lumped-parameter thermal network (LPTN) is fast and interpretable but struggles with nonlinear behavior under high power density. We propose OLTEM, a physics-informed deep model that combines LPTN with a thermal neural network (TNN) to improve prediction accuracy while keeping physical meaning. Methods: OLTEM embeds LPTN into a recurrent state-space formulation and learns three parameter sets: thermal conductance, inverse thermal capacitance, and power loss. Two additions are introduced: (i) a state-conditioned squeeze-and-excitation (SC-SE) attention that adapts feature weights using the current temperature state, and (ii) an enhanced power-loss sub-network that uses a deep MLP with SC-SE and non-negativity constraints. The model is trained and evaluated on the public Electric Motor Temperature dataset (Paderborn University/Kaggle). Performance is measured by mean squared error (MSE) and maximum absolute error across permanent-magnet, stator-yoke, stator-tooth, and stator-winding temperatures. Results: OLTEM tracks fast thermal transients and yields lower MSE than both the baseline TNN and a CNN–RNN model for all four components. On a held-out generalization set, MSE remains below 4.0 °C² and the maximum absolute error is about 4.3–8.2 °C. Ablation shows that removing either SC-SE or the enhanced power-loss module degrades accuracy, confirming their complementary roles. Conclusions: By combining physics with learned attention and loss modeling, OLTEM improves PMSM temperature prediction while preserving interpretability. This approach can support motor thermal design and control; future work will study transfer to other machines and further reduce short-term errors during abrupt operating changes. Full article

The identification of microRNA (miRNA) biomarkers is crucial in advancing disease research and improving diagnostic precision. Network-based analysis methods are powerful for identifying disease-related biomarkers. However, it is a challenge to generate a robust molecular network that can accurately reflect miRNA interactions and define reliable miRNA biomarkers. To tackle this issue, we propose a disease-related miRNA biomarker identification method based on the knowledge-enhanced bio-network (BIM-Ken) by combining the miRNA expression data and prior knowledge. BIM-Ken constructs the miRNA cooperation network by examining the miRNA interactions based on the miRNA expression data, which contains characteristics about the specific disease, and the information of the network nodes (miRNAs) is enriched by miRNA knowledge (i.e., miRNA-disease associations) from databases. Further, BIM-Ken optimizes the miRNA cooperation network using the well-designed GAE (graph auto-encoder). We improve the loss function by introducing the functional consistency and the difference prompt, so as to facilitate the optimized network to keep the intrinsically important characteristics of the miRNA data about the specific disease and the prior knowledge. The experimental results on the public datasets showed the superiority of BIM-Ken in classification. Subsequently, BIM-Ken was applied to analyze renal cell carcinoma data, and the defined key modules demonstrated involvement in the cancer-related pathways with good discrimination ability. Full article

The likely increase in the frequency and severity of droughts with climate warming will pose an enormous challenge for the conservation of forest biodiversity. Documenting the response of species to recent droughts can inform future conservation actions. Mammals that breed and mature slowly may be especially vulnerable to drought-induced disruption to breeding. The yellow-bellied glider (Petaurus australis, Shaw) is a threatened low-density, arboreal marsupial of eastern Australia. Following a severe drought in 2019, one population had declined by 48% by 2021. The present study investigated whether this population had recovered 3–4 years (2022 and 2023) after that drought. Audio surveys of this highly vocal species were conducted at 42 sites, sampling > 1000 h per year, and producing recordings of 2038–2856 call sequences. The probability of occupancy varied little across the two survey years (0.92–0.97). Local abundance in 2023 had returned to pre-drought levels (45% of occupied sites had ≥3 individuals compared to 6% in 2021). These findings show a recovery from a drought-induced decline required at least 3 years, in keeping with the slow life history traits of this species. This study highlights the importance of considering a species’ life history strategy when evaluating its sensitivity to drought. Full article

Introduction: Radiomics is the extraction of non-invasive and reproducible quantitative imaging features, which may yield mineable information for clinical practice implementation. Quantification of lung function through radiomics could play a role in the management of patients with pulmonary lesions. The aim of this study is to test the capability of radiomic features to predict pulmonary function parameters, focusing on the diffusing capacity of lungs to carbon monoxide (DL_CO). Methods: Retrospective data were retrieved from electronical medical records of patients treated with Stereotactic Body Radiation Therapy (SBRT) at a single institution. Inclusion criteria were as follows: (1) SBRT treatment performed for primary early-stage non-small cell lung cancer (ES-NSCLC) or oligometastatic lung nodules, (2) availability of simulation four-dimensional computed tomography (4DCT) scan, (3) baseline spirometry data availability, (4) availability of baseline clinical data, and (5) written informed consent for the anonymized use of data. The gross tumor volume (GTV) was segmented on 4DCT reconstructed phases representing the moment of maximum inhalation and maximum exhalation (Phase 0 and Phase 50, respectively), and radiomic features were extracted from the lung parenchyma subtracting the lesion/s. An iterative algorithm was clustered based on correlation, while keeping only those most associated with baseline and post-treatment DL_CO. Three models were built to predict DL_CO abnormality: the clinical model—containing clinical information; the radiomic model—containing the radiomic score; the clinical-radiomic model—containing clinical information and the radiomic score. For the models just described, the following were constructed: Model 1 based on the features in Phase 0; Model 2 based on the features in Phase 50; Model 3 based on the difference between the two phases. The AUC was used to compare their performances. Results: A total of 98 patients met the inclusion criteria. The Charlson Comorbidity Index (CCI) scored as the clinical variable most associated with baseline DL_CO (p = 0.014), while the most associated features were mainly texture features and similar among the two phases. Clinical-radiomic models were the best at predicting both baseline and post-treatment abnormal DL_CO. In particular, the performances for the three clinical-radiomic models at predicting baseline abnormal DL_CO were AUC₁ = 0.72, AUC₂ = 0.72, and AUC₃ = 0.75, for Model 1, Model 2, and Model 3, respectively. Regarding the prediction of post-treatment abnormal DL_CO, the performances of the three clinical-radiomic models were AUC₁ = 0.91, AUC₂ = 0.91, and AUC₃ = 0.95, for Model 1, Model 2, and Model 3, respectively. Conclusions: This study demonstrates that radiomic features extracted from healthy lung parenchyma on a 4DCT scan are associated with baseline pulmonary function parameters, showing that radiomics can add a layer of information in surrogate models for lung function assessment. Preliminary results suggest the potential applicability of these models for predicting post-SBRT lung function, warranting validation in larger, prospective cohorts. Full article

Bring them home, keep them home is research based in New South Wales (NSW) Australia, that aims to understand successful and sustainable reunification for Aboriginal families who have children in out-of-home care (OOHC). This research is led by Aboriginal researchers, and partners with Aboriginal organisations. It is informed by the experiences of 20 Aboriginal parents and family members, and more than 200 practitioners and professionals working in child protection and reunification. This paper traces the evolution of Bring them home, keep them home which is now at the forefront of influence for NSW child protection reforms. Using specific examples, it highlights the role of research advocacy and resistance in challenging and disrupting systems in ways that amplify the voices of Aboriginal families and communities and embeds these voices as the foundation for radical innovation for child reunification approaches. The paper shares lessons being learned and insights for Aboriginal-led research with communities in the pursuit of restorative justice, system change, and self-determination. Providing a framework for liberating Aboriginal families from child abduction systems, this paper seeks to offer a truth-telling and practical contribution to the international efforts of Indigenous resistance to child abduction systems. Full article

Multi-focus image fusion (MFIF) is an image-processing method that aims to generate fully focused images by integrating source images from different focal planes. However, the defocus spread effect (DSE) often leads to blurred or jagged focus/defocus boundaries in fused images, which affects the quality of the image. To address this issue, this paper proposes a novel model that embeds the Kolmogorov–Arnold network with convolutional layers in parallel within the U-Net architecture (KCUNet). This model keeps the spatial dimensions of the feature map constant to maintain high-resolution details while progressively increasing the number of channels to capture multi-level features at the encoding stage. In addition, KCUNet incorporates a content-guided attention mechanism to enhance edge information processing, which is crucial for DSE reduction and edge preservation. The model’s performance is optimized through a hybrid loss function that evaluates in several aspects, including edge alignment, mask prediction, and image quality. Finally, comparative evaluations against 15 state-of-the-art methods demonstrate KCUNet’s superior performance in both qualitative and quantitative analyses. Full article

Advanced technologies, such as robotic process automation, blockchain, and machine learning, increase audit efficiency. Nonetheless, some Supreme Audit Institutions (SAIs) have not undergone digital transformation. This research aimed to develop a comprehensive framework for supreme audit institutions to adopt and integrate emerging technologies into their auditing processes using a hybrid theoretical approach based on the TAM (Technology Acceptance Model) and TOE (Technology–Organisation–Environment) models. The framework was informed by insights from nineteen highly experienced experts in the field from eight countries. Through a two-round Delphi questionnaire, the experts provided valuable input on the key factors, challenges, and strategies for successful technology adoption by public sector audit organisations. The findings of this research reveal that technology adoption in SAIs starts with solid management support led by the chief technology officer. They must evaluate the IT infrastructure and readiness for advanced technologies, considering the budget and funding. Integrating solutions like the SAI of Ghana’s Audit Management Information System can significantly enhance audit efficiency. Continuous staff training is essential to build a positive attitude toward new technologies, covering areas like data algorithm auditing and big data analysis. Assessing the complexity and compatibility of new technologies ensures ease of use and cost-effectiveness. Continuous support from technology providers and monitoring advancements will keep SAIs aligned with technological developments, enhancing their auditing capabilities. Full article

Addressing privacy concerns is one of the key challenges facing the development of Internet of Things (IoT)-based systems (IoTSs). As IoT devices often collect and process personal and sensitive information, strict privacy policies must be defined and enforced to keep data secure and safe, ensuring security and regulatory compliance. Any data breach could compromise the security of the system, leading to various types of threats and attacks, some of which could even endanger human life. Therefore, it is crucial to design and build a comprehensive and general privacy framework for the development of IoTSs. This framework should not be limited to specific IoTS domains but should be general enough to support and cover most IoTS domains. In this paper, we present a framework that assists developers by (i) enabling them to build IoTSs that comply with privacy standards, such as the General Data Protection Regulation (GDPR), and (ii) providing a simplified and practical approach to identifying and addressing privacy concerns. In addition, the framework enables developers to implement effective countermeasures. Full article

Accurate forecasting of river water flow helps to warn of floods and droughts in advance, provides a basis for the rational allocation of water resources, and at the same time, offers important support for the safe operation of hydropower stations and water conservancy projects. Water flow is characterized by time series, but the existing models focus on the positive series when LSTM is applied, without considering the different contributions of the water flow series to the model at different moments. In order to solve this problem, this study proposes a river water flow prediction model, named AT-BiLSTM, which mainly consists of a bidirectional layer and an attention layer. The bidirectional layer is able to better capture the long-distance dependencies in the sequential data by combining the forward and backward information processing capabilities. In addition, the attention layer focuses on key parts and ignores irrelevant information when processing water flow data series. The effectiveness of the proposed method was validated against an actual dataset from the Shizuishan monitoring station on the Yellow River in China. The results confirmed that compared with the RNN model, the proposed model significantly reduced the MAE, MSE, and RMSE on the dataset by 27.16%, 42.01%, and 23.85%, respectively, providing the best predictive performance among the six compared models. Moreover, this attention mechanism enables the model to show good performance in 72 h (3 days) forecast, keeping the average prediction error below 6%. This implies that the proposed hybrid model could provide a decision base for river flow flood control and resource allocation. Full article

Combining a large collection of patient data and advanced technology, healthcare organizations can excel in medical research and increase the quality of patient care. At the same time, health records present serious privacy and security challenges because they are confidential and can be breached through networks. Even traditional methods with federated learning are used to share data, patient information might still be at risk of interference while updating the model. This paper proposes the Privacy-Preserving Federated Learning with Homomorphic Encryption (PPFLHE) framework, which strongly supports secure cooperation in healthcare and at the same time providing symmetric privacy protection among participating institutions. Everyone in the collaboration used the same EfficientNet-B0 architecture and training conditions and keeping the model symmetrical throughout the network to achieve a balanced learning process and fairness. All the institutions used CKKS encryption symmetrically for their models to keep data concealed and stop any attempts at inference. Our federated learning process uses FedAvg on the server to symmetrically aggregate encrypted model updates and decrease any delays in our server communication. We attained a classification accuracy of 83.19% and 81.27% when using the APTOS 2019 Blindness Detection dataset and MosMedData CT scan dataset, respectively. Such findings confirm that the PPFLHE framework is generalizable among the broad range of medical imaging methods. In this way, patient data are kept secure while encouraging medical research and treatment to move forward, helping healthcare systems cooperate more effectively. Full article

Background: After an older person moves into a nursing home, the next of kin often continues to participate in the care provided there. This participation in care may contribute valuable knowledge of the preferences and wishes of the older person, thereby helping nursing staff deliver personalised care. Objectives: The aim of this study was to explore how next of kin experience their participation in the care of older persons residing in nursing homes after educating nursing staff about participation in palliative care. Methods: This follow-up study used a qualitative design based on semi-structured interviews with 37 next of kin. A thematic analysis was applied. Results: Participation of the next of kin involved active communication and collaboration with nursing staff, expressed in three themes: striving to achieve co-created care, navigating involvement through presence, and building commitment through communication and information. The dual role of being an emotionally close next of kin and a participant in the relative’s care was challenging and became increasingly burdensome and often overwhelming when the older person’s health deteriorated. Conclusions: This study reveals the need to develop and implement a policy for the participation of next of kin in the care of older people living in nursing homes. In addition, support groups can increase well-being through dialogue with other next of kin, thereby alleviating emotional strain. Increased implementation of life stories and the use of digital communication would keep the next of kin informed about the older person’s condition, especially when they cannot be present in person. Life story is a valuable tool for person-centred care and strengthens the relationships between the next of kin, the older person, and the nursing staff. Full article

Background: Understanding the determinants of children’s outdoor play is an important element for child development and broader public health outcomes. There is growing evidence that children’s opportunities for play, particularly outdoor risky play, are diminishing. Parents are concerned with keeping their child safe while affording them independence to play. This study explored parents’ attitudes to risky play and practices around children’s independent mobility in Ireland with the aim of informing public health strategies promoting healthy childhood environments. Methods: An online survey comprising validated scales and standardised questions was completed by a nationally represented sample of 376 parents of children up to 16 years. Data was analysed via descriptive statistics, chi-square tests, and regression analysis. Results: A total of 376 participants accessed the survey, of which 349 completed it. A total of 84% of participants were female. A total of 74% agreed that children need regular exposure to actual risk to develop risk management skills, and 71% trusted their children to play safely. Chi-square tests reveal significant associations between outdoor play in the rain and school travel (p < 0.01), and appropriate age to begin activities at home and in educational settings (p < 0.05). A moderate association was found between the method of school travel and children’s permission to play in the rain (Cramer’s V = 0.51). Respondents considered supervision to be a necessity to ensure their children’s safety. Overall, the results indicate that parents were risk-averse in three of the six categories of risky play, namely, play near dangerous elements, play with adult tools, and out-of-sight play. Conclusions: This study presents a descriptive analysis of findings from the Ireland State of Play Survey. Findings indicate that although parents recognise the benefits of risky play, there is some contradiction between parental attitudes and actual practices, with a lack of willingness or confidence in permitting their children to participate in all such activities. Full article

With the increase in keeping exotic companion mammals as pets, concerns about antimicrobial resistance (AMR) and its impact on animal and human health are growing. Guinea pigs, a popular pet in Hong Kong and globally, have limited studies regarding antimicrobial culture and sensitivity results. We reviewed bacteriologic and antimicrobial sensitivity results from clinically ill pet guinea pigs from 2019 to 2023 using data from the City University Veterinary Diagnostic Laboratory. Of the 234 clinical samples from 22 veterinary clinics in Hong Kong, 134 (57.3%) showed positive bacterial growth, of which 23 (17.2%) showed mixed bacterial growth. In total, 156 bacterial isolates were identified. Gram-positive bacteria (n = 104, 66.7%) were most commonly recovered, representing 25 bacterial species, most commonly Streptococcus spp., Staphylococcus spp., and Corynebacterium spp. The majority of positive samples were from the integument (43.6%) and urinary tract (33.8%). A total of 85.9% of all isolates were resistant to at least one antimicrobial agent, with over 40% of isolates exhibiting resistance to three or more antimicrobial agents, and 27.6% were multidrug resistant (resistant to at least one agent in three or more antimicrobial classes). High resistance rates were observed for penicillin (45.6%), gentamicin (43.7%), doxycycline (42.1%), and azithromycin (36.3%). In contrast, isolates were highly susceptible to ceftazidime (84.1%), chloramphenicol (82.6%), ciprofloxacin (72.7%), and marbofloxacin (72.2%). These findings highlight the high frequency of AMR in this population of clinically ill pet guinea pigs in Hong Kong and the need for informed and judicious antimicrobial use. Full article