Search Results (399)

This study investigates the relationship between CEO characteristics and firm performance across emerging and developed economies using both panel regression and machine learning techniques. Drawing on Upper Echelons Theory, we examine whether CEO age, tenure, gender, founder status, and appointment origin influence Return on Assets (ROA), Return on Equity (ROE), and market-to-book ratio. We apply the fixed and random effects models for inference and deploy random forest and XGBoost models to determine the feature importance of each CEO trait. Our findings show that CEO tenure consistently predicts improved ROE and ROA, while CEO age and founder status negatively affect firm performance. Female CEOs, though not consistently significant in the baseline models, positively influence market valuation in emerging markets according to interaction models. Firm-level characteristics such as size and leverage dominate CEO traits in explaining performance outcomes, especially in machine learning rankings. By integrating machine learning feature importance, this study contributes an original approach to CEO evaluation, enabling firms and policymakers to prioritise leadership traits that matter most. The findings have practical implications for succession planning, diversity policy, and performance-based executive appointments. Full article

Water pollution from pesticides is a major concern for regulatory agencies worldwide due to expensive detecting mechanisms, delays in the processing of results, and the complexity of the chemical analysis. However, the deployment of monitoring systems utilising the internet of things (IoT) and machine-to-machine communication technologies (M2M) holds promise in overcoming this major global challenge. In this current research, an IoT-based wireless sensor network (WSN) is successfully deployed in rural Kenya at the Kiu watershed, providing in situ pesticide detections and a real-time data visualisation of shallow wells. Kiu is an off-grid community located in an area of intensive agriculture, where residents face a high exposure to pesticides due to farming activities and a reliance on shallow wells for domestic water. The evaluation of path loss models utilising channel characteristics obtained from this study indicate a marked departure from the continuous signal decay with distance. Transmitted packets from deployed sensor nodes indicate minimal mutations of payloads, underscoring systems reliability and data transmission integrity. Additionally, the proposed design significantly reduces the time taken to deliver pesticide measurement results to relevant stakeholders. For the entire monitoring period, pesticide residues were not detected in the selected wells, an outcome validated with lab procedures. These results are attributed to prevailing dry weather conditions which limited the leaching of pesticides to lower layers reaching the water table. Full article

Geothermal systems play a crucial role in understanding Earth’s heat dynamics. The Yunkai Uplift in southern China exemplifies a geothermally rich region characterized by ancient lithologies and high heat flow. This study investigates the geochemical characteristics of geothermal waters in the Yunkai Uplift. Both geothermal and non-thermal water samples were collected along the Xinyi–Lianjiang (XL) Fault Zone and the Cenxi–Luchuan (CL) Fault Zone flanking the core of the Yunkai Mountains. Analytical techniques were applied to examine major ions, trace elements, and dissolved CO₂ and H₂, as well as isotopic characteristics of O, H, Sr, C, and He in water samples, allowing for an investigation of geothermal reservoir temperatures, circulation depths, and mixing processes. The findings indicate that most geothermal waters are influenced by water–rock interactions primarily dominated by granites. The region’s diverse lithologies, change from ancient Caledonian granites and medium–high-grade metamorphic rocks in the central hinterland (XL Fault Zone) to low-grade metamorphic rocks and sedimentary rocks in the western margin (CL Fault Zone). The chemical compositions of geothermal waters are influenced through mixing contacts between diverse rocks of varying ages, leading to distinct geochemical characteristics. Notably, δ¹³C_CO2 values reveal that while some samples exhibit significant contributions from metamorphic CO₂ sources, others are characterized by organic CO₂ origins. Regional heat flow results from the upwelling of mantle magma, supplemented by radioactive heat generated from crustal granites. Isotopic evidence from δ²H and δ¹⁸O indicates that the geothermal waters originate from atmospheric sources, recharged by precipitation in the northern Yunkai Mountains. After infiltrating to specific depths, meteoric waters are heated to temperatures ranging from about 76.4 °C to 178.5 °C before ascending through the XL and CL Fault Zones under buoyancy forces. During their upward migration, geothermal waters undergo significant mixing with cold groundwater (54–92%) in shallow strata. As part of the western boundary of the Yunkai Uplift, the CL Fault Zone may extend deeper into the crust or even interact with the upper mantle but exhibits weaker hydrothermal activities than the XL Fault Zone. The XL Fault Zone, however, is enriched with highly heat-generating granites, is subjected more to both the thermal and mechanical influences of upwelling mantle magma, resulting in a higher heat flow and tension effect, and is more conducive to the formation of geothermal waters. Our findings underscore the role of geotectonic processes, lithological variation, and fault zone activity in shaping the genesis and evolution of geothermal waters in the Yunkai Uplift. Full article

In today’s digital era, social media platforms like Instagram have become deeply embedded in daily life, generating billions of content items each day. This vast stream of publicly accessible data presents a unique opportunity for researchers to gain insights into human behaviour and personality. However, leveraging such unstructured and highly variable data for psychological analysis introduces significant challenges, including data sparsity, noise, and ethical considerations around privacy. This study addresses these challenges by exploring the potential of machine learning to infer personality traits from Instagram content. Motivated by the growing demand for scalable, non-intrusive methods of psychological assessment, we developed a personality prediction system combining convolutional neural networks (CNNs) and random forest (RF) algorithms. Our model is grounded in the Big Five Personality framework, which includes Extraversion, Agreeableness, Conscientiousness, Neuroticism, and Openness. Using data collected with informed consent from 941 participants, we extracted visual features from their Instagram images using two pretrained CNNs, which were then used to train five RF models, each targeting a specific trait. The proposed system achieved an average mean absolute error of 0.1867 across all traits. Compared to the PAN-2015 benchmark, our method demonstrated competitive performance. These results highlight that using social media data for personality prediction offers potential applications in personalized content delivery, mental health monitoring, and human–computer interactions. Full article

Epileptic seizure detection and classification from EEG recordings faces significant challenges due to extreme class imbalance. Analysis of the Temple University Hospital Seizure (TUSZ) dataset reveals imbalance ratios of 150:1 between common and rare seizure types, with high temporal heterogeneity (seizure durations of 1–1638 s). We propose a cascaded deep learning architecture with two specialized CNNs: a binary detector followed by a multi-class classifier. This approach decomposes the classification problem, reducing the maximum imbalance from 150:1 to manageable levels (9:1 binary, 5:1 type). The architecture implements a high-confidence filtering mechanism (threshold = 0.9), creating a 99.5% pure dataset for type classification, dynamic class-weighted optimization proportional to inverse class frequencies, and information flow refinement through progressive stages. Loss dynamics analysis reveals that our weighting scheme strategically redistributes optimization attention, reducing variance by 90.7% for majority classes while increasing variance for minority classes, ensuring all seizure types receive proportional learning signals regardless of representation. The binary classifier achieves 99.64% specificity and 98.23% sensitivity (ROC-AUC = 0.995). The type classifier demonstrates >99% accuracy across seven seizure categories with perfect (100%) classification for three seizure types despite minimal representation. Cross-dataset validation on the University of Bonn dataset confirms robust generalization (96.0% accuracy) for binary seizure detection. This framework effectively addresses multi-level imbalance in neurophysiological signal classification with hierarchical class structures. Full article

The affected family members (AFM) of relatives with substance use problems (RSU) play an important role in supporting their relatives to enter substance use treatment. This study investigated the help-seeking behaviours for their relatives by AFM in Brazil, including the characteristics of those who sought help and the risk factors for delaying it. A secondary analysis from a national cross-sectional study of 3030 AFM was performed. Participants were recruited from a range of services focused on AFM across each of the five Brazilian regions (North, Northeast, Central-West, Southeast, South). While 92.7% sought help, 66.0% delayed for an average of 37.2 (SD 70.71) months. Help seeking was associated with higher socioeconomic status and being from the Southeastern region. Barriers included the relative refusing help (31.5%) and the belief that help was not needed (20.6%). Longer delays were associated with female AFM, residents in the Central-West region, non-parents, older RSU, alcohol use, and withdrawal coping strategies. The findings show disparities in help-seeking behaviour across socioeconomic groups, regions, and substance types, highlighting the need for better healthcare workforce distribution and targeted interventions to educate AFMs on the importance of engagement with healthcare services. Full article

Cisheteronormativities inform and distort what LGBTQ+ young people’s bodies can be and do, and what choices about the body are possible, profoundly impacting mental health. This article presents findings from a UK study examining ‘what works’ in early intervention mental health support for LGBTQ+ youth to examine how these impacts can be addressed. Data were collected across 12 mental health support services via the following: interviews with LGBTQ+ youth aged 12–25, service staff/volunteers, and parents/carers (n = 93); document review; and non-participant observation. In analysis, ‘Body’ was identified as a key principle underpinning effective early intervention mental health support. This article presents three key areas: the ability to name and define the body; the body’s ability to ‘do’; and the ability to make informed decisions about one’s body, life, and future. This article highlights the urgent importance of upholding bodily autonomy for LGBTQ+ youth if efforts to address mental health inequalities are to have any chance at success. Full article

Smart cities are widely regarded as a promising solution to urbanization challenges; however, environmental aspects such as outdoor thermal comfort and urban heat island are often less addressed than social and economic dimensions of sustainability. To address this gap, we developed and evaluated an affordable, scalable, and cost-effective weather station platform, consisting of a centralized server and portable edge devices to facilitate urban heat island and outdoor thermal comfort studies. This edge device is designed in accordance with the ISO 7726 (1998) standards and further enhanced with a positioning system. The device can regularly log parameters such as air temperature, relative humidity, globe temperature, wind speed, and geographical coordinates. Strategic selection of components allowed for a low-cost device that can perform data manipulation, pre-processing, store the data, and exchange data with a centralized server via the internet. The centralized server facilitates scalability, processing, storage, and live monitoring of data acquisition processes. The edge devices’ electrical and shielding design was evaluated against a commercial weather station, showing Mean Absolute Error and Root Mean Square Error values of 0.1 and 0.33, respectively, for air temperature. Further, empirical test campaigns were conducted under two scenarios: “stop-and-go” and “on-the-move”. These tests provided an insight into transition and response times required for urban heat island and thermal comfort studies, and evaluated the platform’s overall performance, validating it for nuanced human-scale thermal comfort, urban heat island, and bio-meteorological studies. Full article

This study introduces a novel forecast combination method for monthly Japanese tourism demand, analyzed at both aggregated and disaggregated levels, including tourist, business, and other travel purposes. The sample period spans from January 1996 to December 2018. Initially, the time series data were decomposed into high and low frequencies using the Ensemble Empirical Mode Decomposition (EEMD) technique. Following this, Autoregressive Integrated Moving Average (ARIMA), Neural Network (NN), and Support Vector Machine (SVM) forecasting models were applied to each decomposed component individually. The forecasts from these models were then combined to produce the final predictions. Our findings indicate that the two-stage forecast combination method significantly enhances forecasting accuracy in most cases. Consequently, the combined forecasts utilizing EEMD outperform those generated by individual models. Full article

Neonicotinoids (NEOs) have emerged as viable alternatives to conventional organophosphate pesticides and are widely used in agriculture, horticulture, and household applications. However, the increasing frequency and concentration of NEOs detected in water, sediments, soil, and other environmental media have raised significant concerns about their threats to ecosystems and public health globally. This review paper compiles and integrates key findings from previous studies to analyze the overall occurrence and distribution trends of NEOs in sediments, soil, and other environmental media in China from 2019 to 2024, which has updated and analyzed new data and advanced the knowledge that the previous literature disclosed. The main findings of this work were that over the past decades, NEOs have been consistently detected in sediments, soils, and other environmental media at concentrations ranging from 1 to 10 ng g⁻¹ dw. Acetamiprid (ACE), imidacloprid (IMI), clothianidin (CLO), and thiamethoxam (THM) are the most frequently detected NEOs in sediments and soil. It was found from this work that the threshold concentration of NEOs in soil is very limited, and there are no official acceptable toxic levels of NEOs in soil/water/sediments. Only few countries have conducted the work, at the initial phase, on regulating NEOs and have established their regulatory threshold levels. The associated ecological risks and levels of human exposure in soil have been evaluated, revealing that imidacloprid and thiamethoxam present higher risks for long-term environmental contamination due to their relatively higher concentrations. In contrast, acetamiprid, clothianidin, dinotefuran, and thiacloprid exhibited lower environmental persistence, potentially posing lower ecological risks. These trends imply the need for more focused monitoring and regulatory efforts for compounds like imidacloprid, which exhibit higher concentrations in environmental media. Despite these findings, the contamination of NEOs in sediments and soils is still considered to receive insufficient attention, particularly in northern and western China. Furthermore, the presence of NEOs in other environmental media, including indoor dust, wheat grains, vegetables, and teas, warrants further investigation and concern. Full article

This paper introduces a novel framework for addressing domain adaptation challenges in large language models (LLMs), emphasising privacy-preserving synthetic data generation and efficient fine-tuning. The proposed framework employs a multi-stage approach that includes document ingestion, relevance assessment, and automated dataset creation. This process reduces the need for extensive technical expertise while safeguarding data privacy. We evaluate the framework’s performance on domain-specific tasks in fields such as biobanking and public health, demonstrating that models fine-tuned using our method achieve results comparable to larger proprietary models. Crucially, these models maintain their general instruction-following capabilities, even when adapted to specialised domains, as shown through experiments with 7B and 8B parameter LLMs. Key components of the framework include continuous pre-training, supervised fine-tuning (SFT), and reinforcement learning methods such as direct preference optimisation (DPO), which together provide a flexible and configurable solution for deploying LLMs. The framework supports both local models and API-based solutions, making it scalable and accessible. By enabling privacy-preserving, domain-specific adaptation without requiring extensive expertise, this framework represents a significant step forward in the deployment of LLMs for specialised applications. The framework significantly lowers the barrier to domain adaptation for small- and medium-sized enterprises (SMEs), enabling them to utilise the power of LLMs without requiring extensive resources or technical expertise. Full article

Background: This study explores barriers and facilitators to psychological safety during medical procedures among individuals diagnosed with chronic illnesses in childhood. Psychological safety in healthcare, detected via neuroception and the autonomic nervous system’s responses to perceived safety or threat, is essential for the well-being and mental health of chronically ill patients, especially those with early diagnoses. Methods: Using Polyvagal Theory as a framework, semi-structured interviews were conducted with six participants (aged 20–64) who experienced chronic disease from a young age. The Neuroception of Psychological Safety Scale (NPSS) guided thematic exploration to understand participants’ experiences. Thematic analysis identified key themes that reflect contributors and detractors to psychological safety during medical care. Results: Four primary themes were developed: (1) knowledge empowerment through information and facilitated inquiry, (2) holistic acknowledgment of psychological and social impacts, (3) the role of parental involvement in healthcare interactions, and (4) the need for an individualised, patient-centred approach. Participants expressed a need for psychological support integrated with their medical treatment and the importance of autonomy and clear communication. Conclusions: Psychological safety is central to medical experiences for chronically ill individuals and requires a patient-centred, psychologically informed approach. Emphasising tailored support, family involvement, and comprehensive mental health consideration can foster more effective care and enhance patients’ long-term well-being. Full article

This paper examines the relationship between board gender diversity and Environmental, Social, and Governance (ESG) disclosure in developed economies. Using a sample of forty-five firms across developed countries between 2012 and 2023, the analysis employs Bloomberg’s ESG disclosure score as a proxy. In terms of methodology, both pooled ordinary least squares (OLS) and fixed effects regression models are employed. However, to mitigate potential endogeneity concerns, the study employs an instrumental variable approach and dynamic panel regression techniques to provide robust causal inference. The findings offer two significant insights. In accordance with critical mass theory, firms with a minimum of three female directors demonstrate a significant positive relationship between board gender diversity and ESG disclosure. This indicates that achieving a critical level of female representation is essential for fostering meaningful improvements in ESG disclosure scores. Second, firms with merely one or two female directors, often considered token representation, exhibit a negative significant impact on ESG disclosure. Additionally, within the UK context, board gender diversity is positively associated with ESG disclosure, suggesting that institutional frameworks and regulatory environment shape this relationship. Full article

Background: Displaced femoral neck fractures (FNFs) can be treated with hemiarthroplasty (HA) or total hip arthroplasty (THA), with THA typically offered to fitter patients. Sarcopenia increases complications and mortality after hip arthroplasty. The psoas muscle–L3 vertebra ratio (PML3) is a sarcopenia marker. This study evaluated PML3’s role in predicting postoperative outcomes and guiding surgical decision-making. Methods: A retrospective study was conducted at a single centre between January 2021 and December 2024. PML3 was measured on computed tomography (CT) at the L3 vertebra level for patients with displaced FNFs, comparing postoperative outcomes between HA and THA cohorts. Results: Eighty-three patients (fifty-seven female, twenty-six male) were analysed. Forty-three underwent THA, and forty underwent HA. Postoperative complications were higher in HA patients (48% vs. 21%, p = 0.019), with lower 30-day survival (90% vs. 98%). Median PML3 in the HA group was 0.70 mm² (IQR: 0.47–1.47), lower than in the THA group (1.34 mm², IQR: 1.00–1.78, p = 0.002). However, PML3 values for patients that suffered complications (irrespective of surgical decision) were essentially the same; HA, 0.57 mm² (IQR: 0.43–1.83); THA 0.56 mm² (IQR: 0.41–1.05, p = 0.847). ROC analysis showed PML3 as an acceptable predictor of postoperative complications, with an AUC of 0.71. Conclusions: Lower PML3 values correlate with higher postoperative complications and mortality following THA or HA for displaced FNFs, confirming its role as a prognostic marker. Some THA complications in low-PML3 patients might have been avoided by selecting less invasive HA, suggesting THA should be reserved for those with greater muscle reserves. Full article

Digital twin (DT) technology has become a key enabler for prognostics and health management (PHM) in complex industrial systems, yet scaling predictive models for multi-component degradation (MCD) scenarios remains challenging, particularly when transferring insights from predictive models of smaller systems developed with limited data to larger systems. To address this, a physics-informed neural network (PINN) framework that integrates a standardized scaling methodology, enabling scalable DT analytics for MCD prognostics, was developed in this paper. Our approach employs a systematic DevOps workflow that features containerized PINN DT analytics deployed on a Kubernetes cluster for dynamic resource optimization, a real-time DT platform (PTC ThingWorx™), and a custom API for bidirectional data exchange that connects the cluster to the DT platform. A key contribution of this paper is the scalable DT model, which facilitates transfer learning of degradation patterns across heterogeneous hydraulic systems. Three (3) hydraulic system configurations were modeled, analyzing multi-component filter degradation under pump speeds of 700–900 RPM. Trained on limited data from a reference system, the scaled PINN model achieved 88.98% accuracy for initial degradation detection at 900 RPM—outperforming an unscaled baseline of 64.13%—with consistent improvements across various speeds and thresholds. This work advances PHM analytics by reducing costs and development time, providing a scalable framework for cross-system DT deployment. Full article