Search Results (1,637)

Background: Chest radiography (CXR) is the most frequently performed radiological exam worldwide, but reporting backlogs, caused by a shortage of radiologists, remain a critical challenge in emergency care. Artificial intelligence (AI) triage systems can help alleviate this challenge by differentiating normal from abnormal studies and prioritizing urgent cases for review. This study aimed to externally validate TRIA, a commercial AI-powered CXR triage algorithm (NeuralMed, São Paulo, Brazil). Methods: TRIA employs a two-stage deep learning approach, comprising an image segmentation module that isolates the thoracic region, followed by a classification model trained to recognize common cardiopulmonary pathologies. We trained the system on 275,399 CXRs from multiple public and private datasets. We performed external validation retrospectively on 1045 CXRs (568 normal and 477 abnormal) from a teaching university hospital that was not used for training. We established ground truth using a large language model (LLM) to extract findings from original radiologist reports. An independent radiologist review of a 300-report subset confirmed the reliability of this method, achieving an accuracy of 0.98 (95% CI 0.978–0.988). We compared four ensemble decision strategies for abnormality detection. Performance metrics included sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUROC) with 95% CI. Results: The general abnormality classifier achieved strong performance (AUROC 0.911). Individual pathology models for cardiomegaly, pneumothorax, and effusion showed excellent results (AUROC of 0.968, 0.955, and 0.935, respectively). The weighted ensemble demonstrated the best balance, with an accuracy of 0.854 (95% CI, 0.831–0.874), a sensitivity of 0.845 (0.810–0.875), a specificity of 0.861 (0.830–0.887), and an AUROC of 0.927 (0.911–0.940). Sensitivity-prioritized methods achieving sensitivity >0.92 produced lower specificity (<0.69). False negatives were mainly subtle or equivocal cases, although many were still flagged as abnormal by the general classifier. Conclusions: TRIA achieved robust and balanced accuracy in distinguishing normal from abnormal CXRs. Integrating this system into clinical workflows has the potential to reduce reporting delays, prioritize urgent cases, and improve patient safety. These findings support its clinical utility and warrant prospective multicenter validation. Full article

Although the practice of linking Environmental, Social, and Governance (ESG) metrics to executive compensation (ESG compensation) has become increasingly common worldwide, consistent evidence of its economic consequences for corporate value remains limited. Drawing on agency theory and a sustainable governance perspective, this study examines how responsibility-oriented incentive mechanisms translate into corporate financial performance. Using textual data from a large sample of Chinese listed companies and employing the BERT deep learning model for empirical analysis, the results show that ESG compensation significantly improves subsequent financial performance. Further analysis reveals that this effect is primarily driven by incentives related to the environmental and social dimensions of compensation structures. In addition, ESG compensation enhances firms’ ESG rating performance and reduces rating divergence, thereby lowering stakeholders’ transaction costs. The moderating analysis indicates that managerial ability and financial slack both strengthen the positive effect of ESG compensation on financial performance. Overall, this study uncovers the internal mechanism through which ESG compensation promotes corporate value creation and clarifies its practical implications for sustainable corporate governance. Full article

Background/Objectives: Elbow fractures are the most common injuries in children and are frequently evaluated with plain radiographs in the acute setting. As dedicated pediatric radiology services are not widely available, diagnosis of fractures could be delayed. Since 2023, ChatGPT-4 has offered image analysis capabilities, which has untapped potential for radiographic analysis. This study represents the first evaluation of ChatGPT-4o, a multimodal large language model, in interpreting pediatric elbow radiographs for fracture detection, thereby demonstrating its potential as a generalist AI tool distinct from domain-specific pediatric models. Methods: A curated set of 200 pediatric elbow radiographs (100 normal, 100 abnormal with at least one fracture site, 105 right elbow, and 95 left elbow radiographs) acquired between October 2023 and March 2024 at a tertiary pediatric hospital were analyzed in this case–control study. Each anonymized radiograph was evaluated by ChatGPT-4o via a standardized prompt. ChatGPT-4o’s prediction outputs (fracture vs. no fracture) were subsequently compared against verified radiology reports (ground-truth). Diagnostic performance metrics such as sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), and F1 score were calculated. Results: ChatGPT-4o achieved an overall accuracy of 85% in detecting elbow fractures on pediatric radiographs, with a sensitivity of 87% and specificity of 82%. PPVs and NPVs were 83% and 86%, respectively. The F1 score was 0.85. ChatGPT-4o correctly identified the fracture site in 68 (78%) of the 87 studies in which it had detected fractures accurately. Cohen’s kappa coefficient was 0.69, indicating substantial agreement with actual diagnoses. Conclusions: This study highlights the utility and potential applications of ChatGPT-4o as a valuable point-of-care tool in aiding the detection of pediatric elbow fractures in emergency settings, particularly where specialist access is limited. Full article

Pragmatics concerns how people use language and other expressive means, such as nonverbal and paralinguistic cues, to convey intended meaning in the context. Difficulties in pragmatics are common across distinct clinical conditions, motivating validated assessments such as the Assessment Battery for Communication (ABaCo); whether Large Language Models (LLMs) can serve as reliable coders remains uncertain. In this exploratory study, we used Generative Pre-trained Transformer (GPT)-4o as a rater on 2025 item × dimension units drawn from the responses given by 10 healthy older adults (M = 69.8) to selected ABaCo items. Expert human coders served as the reference standard to compare GPT-4o scores. Agreement metrics included exact agreement, Cohen’s κ, and a discrepancy audit by pragmatic act. Agreement was 89.1% with κ = 0.491. Errors were non-random across acts (χ²(12) = 69.4, p < 0.001). After Benjamini–Hochberg False Discovery Rate correction across 26 cells, only two categories remained significant: false positives concentrated in Command and false negatives in Deceit. Missing prosodic and gestural cues likely exacerbate command-specific failures. In conclusion, in text-only settings, GPT-4o can serve as a supervised second coder for healthy-aging assessments of pragmatic competence, under human oversight. Safe clinical deployment requires population-specific validation and multimodal inputs that recover nonverbal cues. Full article

This research proposes an algorithmic machine learning framework aimed at the early evaluation of business ideas. The framework integrates fifteen critical variables organized into five dimensions—innovation, sustainability, the entrepreneurial team, scalability, and initial finances—identified from a systematic review of the literature. Unlike traditional approaches that focus on financial metrics or one-dimensional indicators, this model provides a comprehensive, multidimensional view of entrepreneurial viability in uncertain contexts. Methodologically, the study presents a structured pipeline that incorporates Random Forest, Gradient Boosting, and XGBoost ensemble algorithms, as well as SMOTE data balancing techniques. These techniques address common problems, such as class imbalance and generalization limitations. Theoretically, innovation and sustainability constructs are operationalized alongside entrepreneurial and financial factors, contributing to more consistent, integrative evaluation models. In practical terms, this proposal provides incubators, accelerators, and public policy designers with a replicable and adaptable tool for the early stages of entrepreneurship. While empirical validation is planned for the future, this work lays the methodological groundwork to bridge gaps in the literature and advance more robust predictive models for entrepreneurial evaluation. Full article

Background: Performance status is often cited as an independent predictor of survival after metastatic spine tumor surgery (MSTS), but its standalone predictive value for short-term outcomes remains unclear. We aimed to evaluate how well Karnofsky (KPS) and Eastern Cooperative Oncology Group performance status (ECOG-PS) predict 90-day survival, a common surgical candidacy threshold, in patients managed with MSTS. Methods: We conducted a retrospective study of 175 adult patients who underwent MSTS at a single institution (2012–2025). All patients had documented preoperative KPS and ECOG-PS scores. Univariable logistic regression was used to assess associations with 90-day survival. Predictive performance was assessed by discrimination (AUC), diagnostic accuracy, calibration (Brier score), and clinical utility (decision curve analysis). Results: The crude 90-day survival rate was 73%. Both KPS (OR 1.02 [95% CI 1.01 to 1.05]; p = 0.001) and ECOG-PS (OR 0.51 [95% CI 0.36 to 0.73]; p < 0.001) were statistically associated with survival. However, discrimination was modest (AUC 0.65 for KPS, 0.68 for ECOG-PS), with the most balanced diagnostic accuracy achieved at KPS ≥ 70 (sensitivity 0.66, specificity 0.62) and ECOG-PS ≤ 2 (sensitivity 0.76, specificity 0.5). Calibration was fair (Brier scores 0.185 and 0.182, respectively). Decision curve analysis showed minimal net benefit across most threshold probabilities, with ECOG-PS performing slightly better at intermediate thresholds (30–60%), the zone of greatest clinical uncertainty. Conclusions: Despite being widely cited as an independent predictor of postoperative survival in patients with metastatic spine disease, performance status assessed via the KPS and ECOG-PS demonstrated only modest overall discriminatory ability, diagnostic accuracy, calibration, and clinical utility when used alone to predict 90-day survival after MSTS. While both scores retained meaningful value at the extremes (i.e., patients with very poor or very good performance status had more predictable outcomes), caution is warranted in intermediate cases, where performance status alone may be insufficient to guide treatment decisions. These findings highlight the critical difference between statistical association and the real-world clinical utility of a single metric to predict outcome in this patient population. Full article

Background: Accurately predicting protein druggability is crucial for successful drug development, as it significantly reduces the time and resources required to identify viable drug targets. However, existing methods often face trade-offs between accuracy, efficiency, and interpretability. This study aims to introduce a lightweight framework designed to address these challenges effectively. Methods: We present a lightweight framework that embeds proteins into four biologically informed, non-Euclidean metric spaces, derived from analyses of amino acid sequences, predicted secondary structures, and curated post-translational modification (PTM) annotations. These representations capture key features such as hydrophobicity profiles, PTM densities, spatial patterns, and secondary structure composition, providing interpretable proxies for structure-related determinants of druggability. This approach enhances our understanding of protein functionality while improving druggability predictability in a biologically relevant context. Results: Evaluated on an Aspirin-binding protein dataset using leave-one-out cross-validation (LOOCV), our distance-based ensemble achieves 92.25% accuracy (AUC = 0.9358) in the whole-protein setting. This performance significantly outperforms common sequence-only baselines in the literature while remaining computationally efficient. Conclusions: On a refined single-chain subset, our framework demonstrates performance comparable to established feature engineering pipelines, highlighting its potential effectiveness in practical applications. Together, these results strongly suggest that biologically grounded, non-Euclidean embeddings provide an effective and interpretable alternative to resource-intensive 3D pipelines for target assessment in drug discovery. This approach not only enhances our ability to assess protein druggability but also streamlines the overall process of target identification and validation. Full article

Early prediction of common bean (Phaseolus vulgaris L.) yield is essential for improving productivity in tropical agricultural systems. In this study, we integrated canopy structural metrics obtained with the Tracing Radiation and Architecture of Canopies (TRAC) system, unmanned aerial vehicle (UAV)-based multispectral measurements (normalized difference vegetation index—NDVI, projected canopy area), and phenological variables collected from stages R6 to R8 under non-limiting nitrogen conditions. Exploratory analyses (correlation, variance inflation factors—VIF), dimensionality reduction (principal component analysis—PCA), and regularized regression (Elastic Net/LASSO), combined with bootstrap stability selection, were applied to identify a parsimonious subset of robust predictors. The final model, composed of six variables, explained approximately 72% of the variability in plant-level grain yield, with acceptable errors (RMSE ≈ 10.67 g; MAE ≈ 7.91 g). The results demonstrate that combining early vigor, radiation interception, and canopy architecture provides complementary information beyond simple spectral indices. This non-destructive framework delivers an efficient model for early yield estimation and supports site-specific management decisions in common bean with high spatial resolution. By enhancing input-use efficiency and reducing waste, this approach contributes to sustainable development and aligns with the global Sustainable Development Goals (SDGs) for climate-resilient agriculture. Full article

Nest defense is a key component of avian reproductive success, yet its intensity and expression often depend on ecological and social contexts. We investigated the nest defense behaviors of Saunders’s Gulls (Saundersilarus saundersi) breeding in Incheon Bay of South Korea in 2022 in relation to nest density and perceived threats. Using decoy presentations of three heterospecifics, Oriental Magpie (Pica serica; diurnal avian nest predator), common raccoon dog (Nyctereutes procyonoides; nocturnal mammalian nest predator), and Little Tern (Sternula albifrons; neutral co-nester), we quantified latency to respond, bombing attack rate, and the number of mobbing individuals at high- and low-density nesting sites within a breeding colony. Mixed models revealed that latency to respond and attack rates varied strongly with stimulus type, with diurnal predator magpies eliciting the fastest and most intense responses, followed by nocturnal predator raccoon dogs and co-nester terns. Nest density influenced the number of mobbing individuals, which was significantly greater at high-density sites. Principal Component Analysis reduced the three behavioral metrics into a composite score, which correlated negatively with latency and positively with bombing attack rate and mobbing intensity. This score varied with both nest density and stimulus type. Our findings demonstrate that Saunders’s Gulls adjust their nest defense strategies according to both the social context and predator type, highlighting the importance of density-dependent collective nest defense in colonial breeders. Full article

Event–event relation extraction (ERE) is an important and challenging task in natural language processing. At present, state-of-the-art ERE methods mainly adopt supervised learning, especially deep learning, which needs a large number of high-quality labeled event corpora. However, these methods will face the challenge of few-shot learning for extracting Chinese multiple event–event relations. Complex deep learning models often cannot converge effectively on small Chinese event corpora. And the manual event relation labeling is a very time-consuming and uncertain work. This paper proposes a joint extraction model for multiple Chinese event–event relations based on weighted double consistency constraint learning, named the Chinese event–event relations miner (CERMiner), to extract multiple types of Chinese emergency event–event relations jointly. After encoding event pairs from their contexts, a group of weighted double consistency constraint, including common sense constraints and domain constraints, are designed and integrated into model learning to accelerate model convergence on few-shot corpora. To evaluate the effectiveness of the CERMiner model, we conduct experiments on the CEC dataset, which contains three relation types—CE, EC, and AC—with 697, 200, and 242 instances, respectively. We report Precision, Recall, and F1-score as evaluation metrics. Our method achieves 84.8%, 72.7%, and 78.2% in Precision, Recall, and $F_1$-score, respectively, outperforming the SGT baseline by 1.7% in $F_1$-score. These results demonstrate that the proposed model can better realize joint extraction of multiple Chinese emergency event–event relations in low-resource environments compared to existing state-of-the-art methods. Full article

From both theoretical and applied perspectives, the trapezoidal fuzzy numbers are widely relevant fuzzy sets. In this paper, we show that the four kinds of common metrics—the supremum metric, the $L_p$-type $d_p$ metrics, the sendograph metric, and the endograph metric—are equivalent on the trapezoidal fuzzy numbers. In fact, we obtain a stronger result: the convergence induced by these four kinds of metrics on the trapezoidal fuzzy numbers is equivalent to the convergence of the corresponding representation quadruples of the trapezoidal fuzzy numbers in ${\mathbb{R}}^4$. The latter convergence is very easy to verify. Our results give a fundamental understanding of these four kinds of common metrics on the trapezoidal fuzzy numbers and provide a quick judgment condition for the convergence induced by them. Full article

Research shows that individuals with heart failure are 60% more likely to develop dementia because of their shared metabolic risk factors. Developing a classification model to differentiate between these two conditions effectively is crucial for improving diagnostic accuracy, guiding clinical decision-making, and supporting timely interventions in older adults. This study proposes a novel method for dementia classification, distinguishing it from its common comorbidity, heart failure, using blood testing and personal data. A dataset comprising 11,124 imbalanced electronic health records of older adults from hospitals in Chiang Rai, Thailand, was utilized. Conditional tabular generative adversarial networks (CTGANs) were employed to generate synthetic data while preserving key statistical relationships, diversity, and distributions of the original dataset. Two groups of ensemble models were analyzed: the boosting group—extreme gradient boosting, light gradient boosting machine—and the bagging group—random forest and extra trees. Performance metrics, including accuracy, precision, recall, F1-score, and area under the receiver-operating characteristic curve were evaluated. Compared with the synthetic minority oversampling technique, CTGAN-based synthetic data generation significantly enhanced the performance of ensemble learning models in classifying dementia and heart failure. Full article

Psychoacoustic metrics offer a valuable complement to traditional noise evaluation methods for gear transmissions, as they account for the human perception of sound quality rather than relying solely on physical measurements. While parameters such as overall sound pressure level (SPL) and spectral content quantify noise intensity and frequency distribution, they often fail to reflect subjective annoyance caused by tonal or high-frequency components common in gear systems. This review provides a structured overview of how psychoacoustic metrics—including loudness, sharpness, roughness, fluctuation strength, and tonality—are applied in the analysis of gear transmission noise. Relevant studies were identified through a comprehensive search across multiple scientific databases, with 54 meeting the inclusion criteria. The findings highlight both the benefits and limitations of these metrics, and present examples of their industrial application in automotive and mechanical engineering contexts. The review also identifies gaps in current research, particularly in integrating psychoacoustic evaluation with predictive modelling and machine learning, and suggests directions for future work. Full article

Depression is a common mental disorder during gestation, posing potential risks to fetal development and leading to behavioral and psychiatric alterations in offspring. Pharmacological intervention, particularly with selective serotonin reuptake inhibitors (SSRIs), is often necessary. This study investigated the effects of fluoxetine (F) on behavioral and memory changes in rodent offspring following maternal gestational and lactation treatment, as well as potential alterations in hippocampal cellularity compared to control (C) progeny. Methodologies included the Morris water maze, elevated plus maze, activity monitoring, parental behavior assessments, and isotropic fractionation for the quantification of hippocampal cells and neurons. Results indicated that maternal fluoxetine exposure significantly affected the body mass, brain weight, and hippocampal metrics of the offspring, aligning with the ‘selfish brain’ hypothesis. Notably, dams treated with fluoxetine showed reduced parental care, leading to offspring with increased activity levels but no changes in anxiety-like behaviors. However, while there was a decline in learning and memory retention, as assessed by the Morris water maze, working and reference memory did not differ significantly from those of controls. This study establishes an association between fluoxetine treatment, increased hippocampal neuron density, and behavioral changes related to memory and hyperactivity, with implications for understanding behavioral disorders and informing future therapeutic interventions. Full article

Hamstring strain injuries (HSI) are common in soccer and remain challenging to predict, as traditional risk factors often fail to capture the multifactorial nature of injury susceptibility. This prospective cohort study aimed to develop and internally validate a machine learning-assisted logistic regression model for predicting hamstring injuries in amateur soccer players using preseason clinical and strength-related variables. A total of 120 male players were followed for one competitive season (30 weeks). Baseline predictors included age, body mass index, previous injury, and bilateral isometric hip and knee strength measured via handheld dynamometry. Twenty initial predictors were reduced to ten through symmetrical uncertainty feature ranking before training a logistic regression model with elastic-net regularization (training set: n = 83; test set: n = 37) using nested four-fold cross-validation. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration metrics, and confusion matrices. During follow-up, 21 players sustained at least one HSI (32 events; 28% reinjuries), yielding an events-per-variable ratio of 2.1, below ideal thresholds and suggesting possible overfitting. On the independent test set, the model achieved an accuracy of 64.9%, AUC of 0.68 (95% CI 0.52–0.84), calibration slope of 0.85, and intercept of −0.12, with a sensitivity of 60% and specificity of 65.6%. Dominant-leg hip abduction strength was the only statistically significant predictor (OR = 0.82, 95% CI 0.70–0.96), while permutation importance analyses identified previous hamstring injury as the most stable contributor to model performance. Neither age nor hamstring isometric strength demonstrated predictive value. Although model discrimination was moderate and calibration indicated mild overfitting, findings reinforce the prognostic relevance of prior injury and suggest that reduced hip abduction strength may serve as an emerging candidate marker. This study, classified as a TRIPOD Category 2 model (development without external validation), provides preliminary, hypothesis-generating evidence supporting the use of multivariate strength and history-based predictors in future, larger-scale injury prediction research. Full article