Search Results (2,314)

Using neural networks has become the standard technique for medical diagnostics, especially in cancer detection and classification. This work evaluates the performance of Vision Transformer architectures, including Swin Transformer and MaxViT, for several datasets of magnetic resonance imaging (MRI) and computed tomography (CT) scans. We used three training sets of images with brain, lung, and kidney tumors. Each dataset included different classification labels, from brain gliomas and meningiomas to benign and malignant lung conditions and kidney anomalies such as cysts and cancers. This work aims to analyze the behavior of the neural networks in each dataset and the benefits of combining different image modalities and tumor classes. We designed several experiments by fine-tuning the models on combined and individual datasets. The results revealed that the Swin Transformer achieved the highest accuracy, with an average of 99.0% on single datasets and reaching 99.43% on the combined dataset. This research highlights the adaptability of Transformer-based models to various human organs and image modalities. The main contribution lies in evaluating multiple ViT architectures across multi-organ tumor datasets, demonstrating their generalization to multi-organ classification. Integrating these models across diverse datasets could mark a significant advance in precision medicine, paving the way for more efficient healthcare solutions. Full article

The Mirim–Patos–Mangueira coastal lagoon system provides a wide range of ecosystem services. However, its vast territorial extent and the political boundaries that divide it hinder integrated assessments, especially during extreme hydrological events. This study is divided into two parts. First, we assessed the spatial and temporal patterns of water quality in the lagoon system using Sentinel-3/OLCI satellite imagery. Atmospheric correction was performed using ACOLITE, followed by spectral grouping and classification into optical water types (OWTs) using the Sentinel Applications Platform (SNAP). To explore the behavior of water quality parameters across OWTs, Chlorophyll-a and turbidity were estimated using semi-empirical algorithms specifically designed for complex inland and coastal waters. Results showed a gradual increase in mean turbidity from OWT 2 to OWT 6 and a rise in chlorophyll-a from OWT 2 to OWT 4, with a decline at OWT 6. These OWTs correspond, in general terms, to distinct water masses: OWT 2 to clearer waters, OWT 3 and 4 to intermediate/mixed conditions, and OWT 6 to turbid environments. In the second part, we analyzed the response of the Patos Lagoon to flooding in Rio Grande do Sul during an extreme weather event in May 2024. Satellite-derived turbidity estimates were compared with in situ measurements, revealing a systematic underestimation, with a negative bias of 2.6%, a mean relative error of 78%, and a correlation coefficient of 0.85. The findings highlight the utility of OWT classification for tracking changes in water quality and support the use of remote sensing tools to improve environmental monitoring in data-scarce regions, particularly under extreme hydrometeorological conditions. Full article

Investigating brain area functions requires advanced technologies, but meaningful insights depend on correlating neural signals with behavior. Traditional mice behavior annotation methods, including manual and semi-automated approaches, are limited by subjectivity and time constraints. To overcome these limitations, our study employs the YOLO neural network for precise mice tracking and composite RGB frames for behavioral scoring. Our model, trained on over 10,000 frames, accurately classifies sitting, running, and grooming behaviors. Additionally, we provide statistical metrics and data visualization tools. We further combined AI-powered behavior labeling to examine hippocampal neuronal activity using fluorescence microscopy. To analyze neuronal circuit dynamics, we utilized a manifold analysis approach, revealing distinct functional patterns corresponding to transgenic 5xFAD Alzheimer’s model mice. This open-source software enhances the accuracy and efficiency of behavioral and neural data interpretation, advancing neuroscience research. Full article

Breast cancer (BRCA) remains a significant cause of mortality among women, particularly in developing and underdeveloped regions, where early detection is crucial for effective treatment. This research introduces an innovative hybrid model that combines Improved Grey Wolf Optimizer (IGWO) with the Seagull Optimization Algorithm (SOA), forming the IGWO–SOA technique to enhance BRCA detection accuracy. The hybrid model draws inspiration from the adaptive and strategic behaviors of seagulls, especially their ability to dynamically change attack angles in order to effectively tackle complex global optimization challenges. A deep neural network (DNN) is fine-tuned using this hybrid optimization method to address the challenges of hyperparameter selection and overfitting, which are common in DL approaches for BRCA classification. The proposed IGWO–SOA model demonstrates optimal performance in identifying key attributes that contribute to accurate cancer detection using the CBIS-DDSM dataset. Its effectiveness is validated using performance metrics such as loss, F1-score, precision, accuracy, and recall. Notably, the model achieved an impressive accuracy of 99.4%, outperforming existing methods in the domain. By optimizing both the learning parameters and model structure, this research establishes an advanced deep learning framework built upon the IGWO–SOA approach, presenting a robust and reliable method for early BRCA detection with significant potential to improve diagnostic precision. Full article

This study provides a comprehensive assessment and synthesis of the literature on short-selling. It performs a lexicometric analysis, providing a quantitative review of 1093 peer-reviewed journal articles to identify and illustrate the main themes in short-selling research. Almost half the published literature on short-selling is thematically clustered around portfolio management techniques. Other key themes involve short-selling as it relates to risk management, strategic management, and market irregularities. Descending hierarchical classification examines the overall structure of the textual corpus of the short-selling literature and the relationships between its key terms. Similarity analysis reveals that the short-selling literature is highly concentrated, with most conceptual groups closely aligned and fitting into overlapping or conceptually similar areas. Some notable groups highlight prior short-selling studies of market dynamics, behavioral factors, technological advancements, and regulatory frameworks, which can serve as a foundation for market regulators to make more informed decisions that enhance overall market stability. Additionally, this study proposes a conceptual framework in which short-selling can be either a driver or an outcome by integrating the literature on its antecedents, consequences, explanatory variables, and boundary conditions. Finally, it suggests directions for future research. Full article

This study proposes an application framework based on Large Language Models (LLMs) to analyze multimodal heterogeneous data in the power sector and introduces the CLB-BER model for classifying user electricity consumption behavior. We first employ the Euclidean–Cosine Dynamic Windowing (ECDW) method to optimize the adjustment phase of the CLUBS clustering algorithm, improving the classification accuracy of electricity consumption patterns and establishing a mapping between unlabeled behavioral features and user types. To overcome the limitations of traditional clustering algorithms in recognizing emerging consumption patterns, we fine-tune a pre-trained DistilBERT model and integrate it with a Softmax layer to enhance classification performance. The experimental results on real-world power grid data demonstrate that the CLB-BER model significantly outperforms conventional algorithms in terms of classification efficiency and accuracy, achieving 94.21% accuracy and an F1 score of 94.34%, compared to 92.13% accuracy for Transformer and lower accuracy for baselines like KNN (81.45%) and SVM (86.73%); additionally, the Improved-C clustering achieves a silhouette index of 0.63, surpassing CLUBS (0.62) and K-means (0.55), underscoring its potential for power grid analysis and user behavior understanding. Our framework inherently preserves temporal symmetry in consumption patterns through dynamic sequence alignment, enhancing its robustness for real-world applications. Full article

We live in an era where digitalization and omnichannel strategies significantly transform retail landscapes, and accurate spatial analytics from Geographic Information Systems (GIS) can deliver substantial competitive benefits. Nonetheless, despite evident practical advantages for specific targeting strategies and operational efficiency, the degree of GIS integration into academic marketing literature remains ambiguous. Clarifying this uncertainty is beneficial for advancing theoretical understanding and ensuring retail strategies fully leverage robust, data-driven spatial intelligence. To examine the intellectual development of the field, co-occurrence analysis, topic mapping, and citation structure visualization were performed on 4952 peer-reviewed articles using the Bibliometrix R package (version 4.3.3) within R software (version 4.4.1). The results demonstrate that although GIS-based methods have been effectively incorporated into fields like site selection and spatial segmentation, traditional marketing research has not yet entirely adopted them. One of the study’s key findings is the distinction between “author keywords” and “keywords plus,” where researchers concentrate on novel topics like omnichannel retail, artificial intelligence, and logistics. However, “Keywords plus” still refers to more traditional terms such as pricing, customer satisfaction, and consumer behavior. This discrepancy presents a misalignment between current research trends and indexed classification practices. Although the mainstream retail research lacks terminology connected to geomarketing, a theme evolution analysis reveals a growing focus on technology-driven and sustainability-related concepts associated with the Retail 4.0 and 5.0 paradigms. These findings underscore a conceptual and structural deficiency in the literature and indicate the necessity for enhanced integration of GIS and spatial decision support systems (SDSS) in retail marketing. Full article

Background and Objectives: Investigating the adherence to antihypertensive medication and identifying patients with low adherence allows targeted interventions to improve therapeutic outcomes. Artificial intelligence (AI) offers advanced tools for analyzing medication adherence data. This study aimed to develop and validate several predictive models for non-adherence, using patient-reported data collected via a structured questionnaire. Materials and Methods: A cross-sectional, multi-center study was conducted on 3095 hypertensive patients from community pharmacies. A structured questionnaire was administered, collecting data on sociodemographic factors, medical history, self-monitoring behaviors, and informational exposure, alongside medication adherence measured using the Romanian-translated and validated ARMS (Adherence to Refills and Medications Scale). Five machine learning models were developed to predict non-adherence, defined by ARMS quartile-based thresholds. The models included Logistic Regression, Random Forest, and boosting algorithms (CatBoost, LightGBM, and XGBoost). Models were evaluated based on their ability to stratify patients according to adherence risk. Results: A total of 79.13% of respondents had an ARMS Score ≥ 15, indicating a high prevalence of suboptimal adherence. Better adherence was statistically associated (adjusted for age and sex) with more frequent blood pressure self-monitoring, a reduced salt intake, fewer daily supplements, more frequent reading of medication leaflets, and the receipt of specific information from pharmacists. Among the ML models, CatBoost achieved the highest ROC AUC Scores across the non-adherence classifications, although none exceeded 0.75. Conclusions: Several machine learning models were developed and validated to estimate levels of medication non-adherence. While the performance was moderate, the results demonstrate the potential of AI in identifying and stratifying patients by adherence profiles. Notably, to our knowledge, this study represents the first application of permutation and SHapley Additive exPlanations feature importance in combination with probability-based adherence stratification, offering a novel framework for predictive adherence modelling. Full article

Accurate monitoring of animal behaviors enables improved management in precision livestock farming (PLF), supporting critical applications including health assessment, estrus detection, parturition monitoring, and feed intake estimation. Although both contact and non-contact sensing modalities are utilized, wearable devices with embedded sensors (e.g., accelerometers, pressure sensors) offer unique advantages through continuous data streams that enhance behavioral traceability. Focusing specifically on contact sensing techniques, this review examines sensor characteristics and data acquisition challenges, methodologies for processing behavioral data and implementing identification algorithms, industrial applications enabled by recognition outcomes, and prevailing challenges with emerging research opportunities. Current behavior classification relies predominantly on traditional machine learning or deep learning approaches with high-frequency data acquisition. The fundamental limitation restricting advancement in this field is the difficulty in maintaining high-fidelity recognition performance at reduced acquisition rates, particularly for integrated multi-behavior identification. Considering that the computational demands and limited adaptability to complex field environments remain significant constraints, Tiny Machine Learning (Tiny ML) could present opportunities to guide future research toward practical, scalable behavioral monitoring solutions. In addition, algorithm development for functional applications post behavior recognition may represent a critical future research direction. Full article

Approximately 18% of U.S. children experience cognitive and behavioral challenges, with both genetic and environmental contributors. We examined if household insecticides, particularly those used in and around the home and on pets, are associated with neurodevelopmental changes. Data were from children aged 24–60 months in the CHARGE study with the following classifications: autism spectrum disorder (ASD, n = 810), developmental delay (DD, n = 192), and typical development (TD, n = 531). Exposure to indoor, outdoor, and pet insecticides was reported for the period from three months pre-conception to the second birthday. Cognitive and adaptive functioning were assessed using the Mullen Scales of Early Learning and Vineland Adaptive Behavior Scales. Linear regression was used to evaluate associations by diagnostic group, adjusting for confounders. Flea/tick soaps, shampoos, and powders used during year two were significantly associated with lower cognitive and adaptive scores in children with ASD after FDR correction. Flea/tick skin treatments in early pregnancy were associated with reduced scores in the DD group, though not significant after correction, especially when used with high frequency. No associations were observed in TD children. These findings underscore the need to examine early-life exposure to non-agricultural insecticides as modifiable risk factors for neurodevelopment. Full article

Smart home technology enhances the quality of life, particularly with respect to in-home care and health monitoring. While video-based methods provide accurate behavior analysis, privacy concerns drive interest in non-visual alternatives. This study proposes a Bluetooth Low Energy (BLE)-enabled indoor positioning and behavior recognition system, integrating machine learning techniques to support sustainable and privacy-preserving health monitoring. Key optimizations include: (1) a vertically mounted Data Collection Unit (DCU) for improved height positioning, (2) synchronized data collection to reduce discrepancies, (3) Kalman filtering to smooth RSSI signals, and (4) AI-based RSSI analysis for enhanced behavior recognition. Experiments in a real home environment used a smart wristband to assess BLE signal variations across different activities (standing, sitting, lying down). The results show that the proposed system reliably tracks user locations and identifies behavior patterns. This research supports elderly care, remote health monitoring, and non-invasive behavior analysis, providing a privacy-preserving solution for smart healthcare applications. Full article

Generative AI tools are becoming increasingly common in education. They make many tasks easier, but they also raise questions about how students interact with information and whether their ability to think critically might be affected. Although these tools are now part of many learning processes, we still do not fully understand how they influence cognitive behavior or digital maturity. This study proposes a model to help identify different user profiles based on how they engage with AI in educational contexts. The approach combines fuzzy clustering, the Analytic Hierarchy Process (AHP), and explainable AI techniques (SHAP and LIME). It focuses on five dimensions: how AI is used, how users verify information, the cognitive effort involved, decision-making strategies, and reflective behavior. The model was tested on data from 1273 users, revealing three main types of profiles, from users who are highly dependent on automation to more autonomous and critical users. The classification was validated with XGBoost, achieving over 99% accuracy. The explainability analysis helped us understand what factors most influenced each profile. Overall, this framework offers practical insight for educators and institutions looking to promote more responsible and thoughtful use of AI in learning. Full article

Expansive soils, characterized by the presence of surface and subsurface cracks, over-consolidation, and swell-shrink properties, present significant challenges to slope stability in geotechnical engineering. Despite extensive research, preventing geohazards associated with expansive soils remains unresolved. This study investigates shallow sliding failures in slopes of highly expansive soils induced by rainfall, using model tests to explore deformation and mechanical behavior under cyclic wetting and drying conditions, focusing on the interaction between soil properties and environmental factors. Model tests were conducted in a wedge-shaped box filled with Nanyang expansive clay from Henan, China, which is classified as high-plasticity clay (CH) according to the Unified Soil Classification System (USCS). The soil was compacted in four layers to maintain a 1:2 slope ratio (i.e., 1 vertical to 2 horizontal), which reflects typical expansive soil slope configurations observed in the field. Monitoring devices, including moisture sensors, pressure transducers, and displacement sensors, recorded changes in soil moisture, stress, and deformation. A static treatment phase allowed natural crack development to simulate real-world conditions. Key findings revealed that shear failure propagated along pre-existing cracks and weak structural discontinuities, supporting the progressive failure theory in shallow sliding. Cracks significantly influenced water infiltration, creating localized stress concentrations and deformation. Atmospheric conditions and wet-dry cycles were crucial, as increased moisture content reduced soil suction and weakened the slope’s strength. These results enhance understanding of expansive soil slope failure mechanisms and provide a theoretical foundation for developing improved stabilization techniques. Full article

This study aimed to longitudinally analyze changes in body mass index (BMI) among Korean middle and high school students before and after the COVID-19 pandemic. Data were obtained from the national-level Physical Activity Promotion System (PAPS), collected between 2018 and 2024. A total of 171,705 adolescents aged 13 to 18 were included in the analysis (86,542 males and 85,163 females), with a mean age of 15.2 years (SD = 1.68). Time-series analysis and two-way analysis of variance (ANOVA) were conducted to examine differences in BMI by year, sex, region (capital vs. non-capital), and urban–rural classification. The results indicated a significant increase in BMI during the pandemic period (2020–2022), peaking in 2022, followed by a gradual decline thereafter. Notably, male students and those living in rural or non-capital areas consistently exhibited higher BMI levels, suggesting structural disparities in access to physical activity opportunities and health resources. This study employed the Socio-Ecological Model and the Health Equity Framework as theoretical lenses to interpret BMI changes not merely as individual behavioral outcomes but as consequences shaped by environmental and policy-level determinants. The findings underscore the need for equity-based interventions in physical education and health policy to mitigate adolescent health inequalities during future public health crises. Full article

Emerging digital technologies are transforming how consumers participate in financial markets, yet their benefits depend critically on the speed, reliability, and transparency of the underlying platforms. Online stock trading platforms must maintain high efficiency underload to ensure a good user experience. This paper presents performance analysis under various load conditions based on the containerized stock exchange system. A comprehensive data logging pipeline was implemented, capturing metrics such as API response times, database query times, and resource utilization. We analyze the collected data to identify performance patterns, using both statistical analysis and machine learning techniques. Preliminary analysis reveals correlations between application processing time and database load, as well as the impact of user behavior on system performance. Association rule mining is applied to uncover relationships among performance metrics, and multiple classification algorithms are evaluated for their ability to predict user activity class patterns from system metrics. The insights from this work can guide optimizations in similar distributed web applications to improve scalability and reliability under a heavy load. By framing performance not merely as a technical property but as a determinant of financial decision-making and well-being, the study contributes actionable insights for designers of consumer-facing fintech services seeking to meet sustainable development goals through trustworthy, resilient digital infrastructure. Full article