Search Results (59)

Introduction: Optical coherence tomography (OCT) is the gold standard for retinal nerve fibre layer (RNFL) assessment; its high cost and limited accessibility hinder widespread use. This study aims to develop deep learning models that generate RNFL thickness maps from fundus images, providing a cost-effective alternative to OCT. Methods: A dataset of 5000 fundus-OCT image pairs from 5000 unique glaucoma patients was used to train and compare the following four U-Net-based deep learning models: ResU-Net, R2U-Net, Nested U-Net, and Dense U-Net. All models were trained for up to 1000 epochs with early stopping (patience = 50 epochs). Performance was evaluated using Mean Squared Error (MSE), Mean Absolute Error (MAE), Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), and Fréchet Inception Distance (FID). Results: ResU-Net demonstrated the best performance, achieving MSE = 0.00061, MAE = 0.01877, SSIM = 0.9163, PSNR = 32.19 dB, and FID = 30.08. These results represent a 108% improvement in SSIM and a 67% improvement in PSNR compared to previously published benchmark for this task. Conclusions: This study demonstrates that deep learning models, particularly ResU-Net, can generate high-fidelity RNFL thickness maps from fundus photographs, substantially outperforming prior published benchmarks. This approach represents a potential contribution toward accessible glaucoma assessment, contingent upon prospective clinical validation and regulatory evaluation. Full article

Detecting trace metals in soil across geologically diverse terrains remains challenging due to complex mineral–metal interactions and the limited spatial coverage of traditional geochemical tests. This study develops a scalable VIS–NIR–SWIR spectroscopy and machine learning (ML) framework to predict and map soil concentrations of Cr, As, Cu, and Cd in the Aligudarz District, located within the geotectonically complex Sanandaj–Sirjan Zone of western Iran. Laboratory reflectance spectra (~350–2500 nm) obtained from 110 soil samples were pre-processed using derivative filtering, scatter-correction techniques, and genetic algorithm (GA)-based wavelength optimisation to enhance diagnostic absorption features linked to Fe-oxides, clay minerals, and carbonates. Multiple ML-based approaches, including artificial neural networks (ANNs), support vector regression (SVR), and partial least squares regression (PLSR), as well as stepwise multiple linear regression (SMLR), were compared using nested, spatial, and external validation. Nonlinear models, particularly ANNs, exhibited the highest predictive accuracy, with strong generalisation confirmed via an independent test set. GA-selected wavelengths and derivative-enhanced spectra revealed mineralogical controls on metal retention, confirming that spectral predictions reflect underlying geological processes. Ordinary kriging of spectral-ML residuals generated spatially consistent metal-distribution maps that aligned well with local and regional geological features. The integrated framework demonstrates high predictive accuracy and operational scalability, providing a reproducible, field-ready method for rapid geochemical assessment. The findings highlight the potential of VIS–NIR–SWIR spectroscopy, combined with advanced modelling and geostatistics, to support environmental monitoring, mineral exploration, and risk assessment in geologically complex terrains. Full article

A new generalized family of distributions, termed the Cosine–Topp–Leone–Exponentiated Half Logistic–G (Cos–TL–EHL–G) family, is proposed. The primary motivation for introducing this family is to enhance the modelling flexibility of the existing Cosine–Topp–Leone–G class by incorporating a exponentiated half logistic (EHL-G)-based transformation. Two important special cases, namely the Cos–TL–EHL–Weibull (Cos–TL–EHL–W) and Cos–TL–EHL–Log–Logistic (Cos–TL–EHL–LLoG) distributions, are presented. Several mathematical and statistical properties of the proposed family are derived, including series expansions, moments, order statistics, and uncertainty measures. Parameter estimation is carried out using maximum likelihood, least squares, Anderson–Darling, and Cramér–von Mises methods. A Monte Carlo simulation study indicates that the maximum likelihood estimator outperforms the competing estimation techniques. The practical usefulness and robustness of the proposed family are illustrated through applications to two real datasets, where the Cos–TL–EHL–W distribution demonstrates superior performance compared to both nested and non-nested competing models. Full article

Precision livestock farming (PLF), by integrating multimodal sensor data, provides opportunities to enhance welfare monitoring and management in small ruminants. This study evaluated whether environmental, physiological, and behavioral measurements—including the temperature–humidity index (THI), carbon dioxide (CO₂) and ammonia (NH₃) concentrations measured at the barn level, body condition score (BCS), rectal and ocular temperatures, GPS-derived locomotion metrics, accelerometry data, and fixed animal traits—can serve as key predictors of welfare and productivity in dairy sheep. Data were collected from 90 ewes: all animals underwent the same repeated welfare assessments, while 30 of them were additionally equipped with GPS–accelerometer sensor collars; environmental conditions were continuously recorded for the entire flock, generating 773 complete multimodal records. All predictive models were developed using data from all 90 ewes; collar-derived behavioral variables were included only for individuals equipped with GPS–accelerometer collars. Nine regression methods (linear regression (LR), partial least square regression (PLSR), elastic net (EN), mixed-effects models, random forest (RF), extreme gradient boosting (XGBoost), support vector regression (SVR), neural networks (multilayer perceptron, MLP), and an ensemble of RF–XGBoost–EN were evaluated using a combination of nested cross-validation (CV) and leave-one-animal-out CV (LOAOCV) to ensure robustness and generalization at the individual animal level. Nonlinear models—particularly RF, XGBoost, SVR, and the ensemble—consistently delivered superior performance across traits. For behavioral (e.g., daily distance movement) and thermal indicators (e.g., medial canthus temperature), the highest predictive capacity (R² ≈ 0.60–0.70) was achieved, while moderate predictive capacity (R² ≈ 0.40–0.50 and ≈0.35–0.45), respectively, was observed for respiratory rate and milk yield, reflecting their multifactorial nature. Feature importance analyses underscored the relevance of THI, CO₂, NH₃, concentrations, and BCS across results. Overall, these findings demonstrate that multimodal sensor fusion can effectively support the prediction of welfare and productivity indicators in intensively reared dairy sheep and emphasize the need for larger and more diverse datasets to further enhance model generalizability and model transferability. Full article

Traditional electrical resistivity tomography (ERT) technology confronts bottlenecks such as the volume effect in the detection of termite nests in levees, while the ERT based on deep learning has insufficient interpretation accuracy due to small sample data. This study proposes an intelligent ERT diagnosis framework that integrates generative adversarial networks (GANs) with semantic segmentation models. The GAN-enhanced networks (GFU-Net and GFL-Net) are developed, incorporating a Squeeze-and-Excitation (SE) attention mechanism to suppress false anomalies. Additionally, a comprehensive loss function combining binary cross-entropy (BCE) and the Focal loss function is used to address the issue of sample imbalance. Using forward modeling based on the finite difference method (FDM), a termite nest hidden danger ERT dataset, which includes seven types of high-resistance anomaly configurations, is generated. Numerical simulations demonstrate that GFL-Net achieves a mean intersection-over-union (mIoU) of 97.68% and a spatial positioning error of less than 0.04 m. In field validation on a red clay embankment in Jiangxi Province, this method significantly improves the positioning accuracy of hidden termite nests compared to traditional least squares (LS) inversion. Excavation verification results show that the maximum error in the horizontal center and top burial depth of the termite nest identified by GFL-Net is less than 7% and 16%, respectively. The research findings provide reliable technical support for the accurate identification of termite nest hidden dangers in embankments. Full article

Underwater passive acoustic monitoring (PAM) serves as a core approach pervasively applied to the long-term, non-invasive detection of biological acoustic signals. Dolphin whistles serve as a fundamental aspect of vocal communication, exhibiting intricate frequency-modulated structures. Robust detection of these whistles is essential for dolphin species diversity conservation, yet performance is frequently compromised by underwater background noise, leading to significant degradation in detection reliability. To address this issue, this paper presents an unsupervised enhancement method based on Dually-Regularized Non-Negative Matrix Factorization (DR-NMF). Beyond a standard data fidelity term, the proposed framework integrates two specialized regularizers, including Overlapping Group Shrinkage and Group Lasso. The former promotes time–frequency continuity of whistle ridges, while the latter adaptively eliminates redundant bases, achieving an improved trade-off between structural integrity and noise suppression. The optimization procedure employed a combination of majorization–minimization, iteratively reweighted least squares, and proximal gradient techniques, all of which were implemented within an alternating minimization scheme featuring nested inner–outer iterations. This architecture ensures stable convergence and computational practicality. Extensive experimental evaluations under diverse low signal-to-noise ratio (SNR) conditions reveal that the proposed method achieves a substantial improvement in recall without compromising precision, resulting in consistent enhancements in frame-level $F_1$-scores. When applied to real-world dolphin whistle recordings, our method outperforms existing baseline approaches, demonstrating remarkable robustness in detecting whistle signals when amidst challenging marine environmental noise. Full article

Background/Objectives: Excessive weight gain is a growing concern among people living with HIV (PWH) receiving integrase strand transfer inhibitor (INSTI)-based regimens as first-line antiretroviral therapy (ART), as it may contribute to multimorbidity. The mechanisms driving weight gain in INSTI users are not fully understood but are thought to be multifactorial. This study examines the plasma metabolome associated with weight gain in PWH on INSTI-based regimens. Methods: We conducted a nested case–control study within the randomized clinical trial MICTLAN (NCT06629480). Sixty-six participants were randomized to receive INSTI-based regimens, either bictegravir/tenofovir alafenamide/emtricitabine (BIC/TAF/FTC) or dolutegravir/abacavir/lamivudine (DTG/ABC/3TC), and followed for 18 months. Weight gain >10% relative to baseline was considered a primary endpoint and used as a criterium to categorize cases (n = 28) and controls (n = 38). Anthropometric and clinical measurements, plasma insulin, and metabolomic profiles were assessed at baseline and 18 months post-ART. Plasma untargeted metabolomics was performed using liquid chromatography–mass spectrometry (LC-MS/MS) to identify metabolomic changes linked to weight gain. Bioinformatic tools, including Partial Least Squares Discriminant Analysis (PLS-DA), volcano plots, and KEGG pathway enrichment analysis, were used to analyze plasma metabolomes and identify significant differential metabolites. Results: Weight gain at 18 months in PWH on INSTI-based ART was associated with insulin resistance, as measured by HOMA-IR (OR 3.23; 95% CI 1.14–9.10; p = 0.023), and visceral adipose tissue thickness > 4 cm (OR 4.50; 95% CI 1.60–13.03; 9.10; p = 0.004), and hypertriglyceridemia (OR 3.9; 95% CI 1.38–10.94; p = 0.008). Baseline HIV RNA viral load >50,000 copies/mL (OR 8.05; 95% CI 2.65–24.43; p = 0.0002) was identified as a baseline predictor of weight gain (aOR 6.58 (1.83–23.58); p = 0.004). In addition, accumulation of circulating medium-chain acylcarnitines, indicative of mitochondrial dysfunction, and insulin resistance were linked to weight gain in PWH on INSTI-based regimens after 18 months of therapy. Conclusions: This metabolomic study identified metabolites reflecting mitochondrial dysfunction, dysregulated lipid metabolism, and altered amino acid metabolism as key mechanisms underlying insulin resistance and weight gain in PWH on INSTI-based ART. Full article

Objective: Vitamin D deficiency (VDD) is common in pregnancy and may affect lipid metabolism. The underlying mechanisms are multifactorial, but most evidence so far comes from non-pregnant populations. This study aims to identify metabolites and metabolic patterns associated with VDD in early pregnancy and to evaluate their relationships with maternal lipid profiles. Methods: A nested case–control research was carried out in the Zhoushan Pregnant Women Cohort (ZPWC). Cases were defined as women with VDD (25(OH)D < 20 ng/mL), and controls (≥20 ng/mL) were matched 1:1 using propensity scores based on age, pre-pregnancy BMI, gestational week, and calendar year at blood sampling. The untargeted metabolomics of first-trimester maternal plasma were measured. Metabolic profiles were analyzed using partial least squares-discriminant analysis (PLS-DA). Principal component analysis (PCA) was applied to visualize group separation, and metabolite set enrichment analysis (MSEA) was performed to reveal biologically relevant metabolic patterns. Associations between VDD-related metabolite components in early pregnancy and lipid levels in mid-pregnancy were assessed using linear regression models. Results: 44 cases and 44 controls were selected for the study. There were 60 metabolites identified as being connected to VDD. Among these, 26 metabolites, primarily glycerophospholipids and fatty acyls, exhibited decreased levels in the VDD group. In contrast, 34 metabolites showed increased levels, mainly comprising benzene derivatives, carboxylic acids, and organooxygen compounds. PCA based on these metabolites explained 52.8% of the total variance (R²X = 0.528) across the first six principal components (PC1: 16.4%, PC2: 10.6%, PC3: 9.2%, PC4: 6.3%, PC5: 5.7%, PC6: 4.6%). PC2, dominated by lineolic acids and derivatives, was negatively associated with total cholesterol (TC), triglycerides (TG), and low-density lipoprotein cholesterol (LDL-C) (all p < 0.01). PC3, dominated by glycerophosphocholines, was negatively associated with TC, TG, and high-density lipoprotein cholesterol (HDL-C) (all p < 0.05). MSEA revealed significant enrichment of the pantothenate and CoA biosynthesis pathway after multiple testing correction (FDR < 0.05). Conclusions: This study reveals distinct metabolic alterations linked to VDD and suggests potential mechanisms underlying its association with maternal lipid metabolism in early pregnancy. Full article

Bacterial leaf streak (BLS), caused by Xanthomonas translucens pv. undulosa, has become an intermittent yet economically significant disease of wheat in the Upper Midwest during the last decade. Because chemical and cultural controls remain ineffective, breeders rely on developing resistant varieties, yet visual ratings in inoculated nurseries are labor-intensive, subjective, and time-consuming. To accelerate this process, we combined unmanned-aerial-vehicle hyperspectral imaging (UAV-HSI) with a carefully tuned chemometric workflow that delivers rapid, objective estimates of disease severity. Principal component analysis cleanly separated BLS, leaf rust, and Fusarium head blight, with the first component explaining 97.76% of the spectral variance, demonstrating in-field pathogen discrimination. Pre-processing of the hyperspectral cubes, followed by robust Partial Least Squares (RPLS) regression, improved model reliability by managing outliers and heteroscedastic noise. Four variable-selection strategies—Variable Importance in Projection (VIP), Interval PLS (iPLS), Recursive Weighted PLS (rPLS), and Genetic Algorithm (GA)—were evaluated; rPLS provided the best balance between parsimony and accuracy, trimming the predictor set from 244 to 29 bands. Informative wavelengths clustered in the near-infrared and red-edge regions, which are linked to chlorophyll loss and canopy water stress. The best model, RPLS with optimal preprocessing and variable selection based on the rPLS method, showed high predictive accuracy, achieving a cross-validated R² of 0.823 and cross-validated RMSE of 7.452, demonstrating its effectiveness for detecting and quantifying BLS. We also explored the spectral overlap with Sentinel-2 bands, showing how UAV-derived maps can nest within satellite mosaics to link plot-level scouting to landscape-scale surveillance. Together, these results lay a practical foundation for breeders to speed the selection of resistant lines and for agronomists to monitor BLS dynamics across multiple spatial scales. Full article

The Engel algorithm was created to solve chip-firing games and can be used to find the stationary distribution for absorbing Markov chains. Kaushal et al. developed a matlab-based version of the generalized Engel algorithm based on Engel’s probabilistic abacus theory. This paper introduces a modified version of the generalized Engel algorithm, which we call the modified Engel algorithm, or the mEngel algorithm for short. This modified version is designed to address issues related to non-absorbing Markov chains. It achieves this by breaking down the transition matrix into two distinct matrices, where each entry in the transition matrix is calculated from the ratio of the numerator and denominator matrices. In a nested iteration setting, these matrices play a crucial role in converting non-absorbing Markov chains into absorbing ones and then back again, thereby providing an approximation of the solutions of non-absorbing Markov chains until the distribution of a Markov chain converges to a stationary distribution. Our results show that the numerical outcomes of the mEngel algorithm align with those obtained from the power method and the canonical decomposition of absorbing Markov chains. We provide an example, Torrence’s problem, to illustrate the application of absorbing probabilities. Furthermore, our proposed algorithm analyzes the Monopoly transition matrix as a form of non-absorbing probabilities based on the rules of the Monopoly game, a complete information dynamic game, particularly the probability of landing on the Jail square, which is determined by the order of the product of the movement, Jail, Chance, and Community Chest matrices. The Long Jail strategy, the Short Jail strategy, and the strategy for getting out of Jail by rolling consecutive doubles three times have been formulated and tested. In addition, choosing which color group to buy is also an important strategy. By comparing the probability distribution of each strategy and the profit return for each property and color group of properties, and the color group property, we find which one should be used when playing Monopoly. In conclusion, the mEngel algorithm, implemented using R codes, offers an alternative approach to solving the Monopoly game and demonstrates practical value. Full article

In this study, we examined the role of intellectual humility (IH) as an antecedent of individual attitude toward sustainable tourism viewed from the lens of personality trait theory, virtue ethics theory, and regenerative tourism principles within a stakeholder framework. Data were collected via Qualtrics in an online survey of 233 adults in the United States. A series of confirmatory factor analyses (CFA) were applied to the data to test the measurement model. In addition, a bifactor CFA was found to have acceptable fit and appropriate in controlling for common method variance. A series of covariance-based structural equations models (SEMs) was estimated to test the hypothesized model while controlling for common method variance in addition to individual age and gender. Using the chi-square difference test for nested model comparison, we found that intellectual humility was a significant antecedent of the negative ecological impact of tourism (β = 0.14, p < 0.01) while its relationships with economic and social impacts of travel became non-significant after controlling for common method variance. Pro-social tendency, operationalized as HEXACO Honesty–Humility, was also a significant antecedent of the negative ecological impact (β = 0.17) and positive economic impact (β = −0.34) of tourism, after controlling for common method variance. Despite its limitations due to its cross-sectional design and use of self-report data in the U.S., this study was novel in introducing intellectual humility as an important virtue to be cultivated at the individual level to achieve a holistic approach to sustainable tourism, especially in shaping destination choices. In addition, the study highlights the need to detect common method variance in self-report data via bifactor CFA to avoid erroneous reporting of significant findings, hampering our collective research efforts to address climate change and its impact. Full article

With the continuous deterioration of air quality and the increasingly serious environmental problem of air pollution, accurate air quality prediction is of great significance for environmental governance. Air quality index (AQI) prediction based on deep learning is currently a hot research topic. The neural network model method currently used for prediction has difficulty effectively coping with the high volatility of AQI data and capturing the complex nonlinear relationships and long-term dependencies in the data. To address these issues, this paper proposes multivariate air quality forecasting with a residual nested LSTM neural network based on the discrete stationary wavelet transform (DSWT) model. Firstly, the DSWT data-decomposition technique decomposes each AQI data point into multiple sub-signals. Then, each sub-signal is sent to the NLSTM layer for processing to capture the temporal relationships between different pollutants. The processed results are then combined, using residual connections to mitigate issues of gradient vanishing and explosion during the model training process. The inverse mean squared error method is combined with the simple weighted average method, to serve as the weight-update approach. Back propagation is then applied, to dynamically adjust the weights based on the prediction accuracy of each sample, further enhancing the model’s prediction accuracy. The experiment was conducted on the air quality index dataset of 12 observation stations in and around Beijing. The results show that the proposed model outperforms several existing models and data-processing methods in multi-task AQI prediction. There were significant improvements in mean absolute error (MAE), root mean square error (RMSE), mean absolute percentage error (MAPE), and R square (R²). Full article

High-sensitivity C-reactive protein (hs-CRP) is a marker of systemic inflammation and is associated with developing dyslipidemia. However, the causality between hs-CRP and dyslipidemia remains unresolved. This study aimed to investigate the relationship between hs-CRP concentrations and dyslipidemia and to explore the potential causal link using Mendelian randomization (MR) analysis. A nested case–control study was conducted with 1174 participants, and genotype data were analyzed using the Korean Chip. A genome-wide association study (GWAS) identified rs76400217 as a suitable instrumental variable (IV) due to its significant association with hs-CRP (p < 10⁻⁸). Logistic regression models, adjusted for confounders, were used to evaluate the association between hs-CRP and dyslipidemia. An MR analysis was performed using a two-stage least squares (2SLS) method, with rs76400217 as the IV to assess causality. Logistic regression showed a significant association between hs-CRP concentrations and dyslipidemia (OR 2.08, 95% CI: 1.81–2.39, p < 0.001). This association remained significant after adjusting for factors such as age, sex, alcohol consumption, and BMI. The MR analysis using rs76400217 as the IV confirmed the strong associations with hs-CRP concentrations (p < 0.001) in all models, but the causality between hs-CRP and dyslipidemia was not statistically significant. Thus, no evidence of a causal relationship between hs-CRP and the risk of dyslipidemia was found in the Korean population. The strong association observed between hs-CRP and dyslipidemia may be due to other contributing factors rather than a direct cause. Full article

Background: Alterations in maternal lipid metabolism have been elucidated by several studies in relation to macrosomia. However, the lipidome of the intrauterine compartment associated with macrosomia, particularly in early pregnancy, remains largely unknown. Objectives: (1) To compare the lipidomic profile of early 2nd trimester amniotic fluid (AF) of healthy mothers with normal body mass index who gave birth to large-for-gestational age (LGA) versus appropriate-for-gestational age (AGA) infants; and (2) to examine if insulin and glucose concentrations in AF were associated with the AF lipidomic profile. Methods: In this nested case–control study, bio-banked AF samples were collected from pregnant women undergoing routine amniocentesis at 12–22 weeks of gestation. A subsample of 15 LGA infants (cases) were contrasted with 15 AGA infants (controls). An untargeted lipidomics analysis using liquid chromatography quadrupole time-of-flight mass spectrometry was conducted. Univariate and multivariate statistical analyses (principal component analysis and partial least-squares discriminant analysis) were used to extract differentially abundant (DA) features with high variable importance in projection (VIP) scores. Results: LGA AF was characterized by elevations of 30 phosphatidic acid species. Among other DA features, sphingomyelin (SM 14:0;O2/20:1) had the highest VIP score and was markedly elevated in LGA AF. Neither insulin nor glucose was associated with 2nd trimester AF lipidomic profiles in these healthy, normal-weight mothers. Conclusion: These findings provide evidence of early dysregulated lipid metabolism in healthy, normal-weight mothers with LGA infants. Full article

Cooperative MIMO communication systems play an important role in the development of future sixth-generation (6G) wireless systems incorporating new technologies such as massive MIMO relay systems, dual-polarized antenna arrays, millimeter-wave communications, and, more recently, communications assisted using intelligent reflecting surfaces (IRSs), and unmanned aerial vehicles (UAVs). In a companion paper, we provided an overview of cooperative communication systems from a tensor modeling perspective. The objective of the present paper is to provide a comprehensive tutorial on semi-blind receivers for MIMO one-way two-hop relay systems, allowing the joint estimation of transmitted symbols and individual communication channels with only a few pilot symbols. After a reminder of some tensor prerequisites, we present an overview of tensor models, with a detailed, unified, and original description of two classes of tensor decomposition frequently used in the design of relay systems, namely nested CPD/PARAFAC and nested Tucker decomposition (TD). Some new variants of nested models are introduced. Uniqueness and identifiability conditions, depending on the algorithm used to estimate the parameters of these models, are established. Two families of algorithms are presented: iterative algorithms based on alternating least squares (ALS) and closed-form solutions using Khatri–Rao and Kronecker factorization methods, which consist of SVD-based rank-one matrix or tensor approximations. In a second part of the paper, the overview of cooperative communication systems is completed before presenting several two-hop relay systems using different codings and configurations in terms of relaying protocol (AF/DF) and channel modeling. The aim of this presentation is firstly to show how these choices lead to different nested tensor models for the signals received at destination. Then, by capitalizing on these models and their correspondence with the generic models studied in the first part, we derive semi-blind receivers to jointly estimate the transmitted symbols and the individual communication channels for each relay system considered. In a third part, extensive Monte Carlo simulation results are presented to compare the performance of relay systems and associated semi-blind receivers in terms of the symbol error rate (SER) and channel estimate normalized mean-square error (NMSE). Their computation time is also compared. Finally, some perspectives are drawn for future research work. Full article