Search Results (175)

Background: The definitions and approaches used to diagnose gestational diabetes (GD) are varied. The two-step approach recommended by the American College of Obstetricians and Gynecologists (ACOG) combines the sensitivity of a glucose challenge test (GCT) with the specificity of a 3-hour oral glucose tolerance test (OGTT). We investigated if minor modification of the two-step procedure can provide improved detection of GD by identifying a risk group of pregnant women with high risk of GD. Methods: We conducted a prospective cohort study of pregnant women enrolled early during pregnancy and followed till delivery. All participants underwent the ACOG-recommended two-step procedure for GD diagnosis. Based on GCT and OGTT results, the participants were divided into four risk groups (RGs): GCT-negative (RG0), GCT-positive but OGTT normal (RG1), single abnormal value on OGTT or raised HbA1c (RG2) and diagnosed GD (RG3). Baseline evaluation included dietary history (24 hour recall) and physical activity. A series of multivariable logistic regression analyses were conducted to estimate the odds of maternal and fetal outcomes. Results: A total of 1041 pregnant women were included in the study, of whom 16 (1.6%) were diagnosed as GD. Our two-step approach identified 48 (4.6%) women as GD, while RG2, RG1 and RG0 comprised 75 (7.2%), 218 (20.9%) and 700 (67.2%), respectively. Compared to RG0, RG2 showed a higher likelihood of antepartum complications [odds ratio and 95% confidence interval 2.38 (1.16–4.15)], any adverse outcome without [2.04 (1.17–3.55)] or with cesarean section [2.09 (1.21–3.61)] and primary cesarean section [1.68 (1.01–2.81)] after adjustment for potential confounders. RG2 was also significantly associated with pregnancy-induced hypertension, meconium-stained amniotic fluid and premature rupture of membranes. Conclusions: In the study participants, we identified a subgroup (RG2) at high risk of GD with perinatal outcomes showing profile consistent with that of GD. Full article

Background/Objectives: Childhood obesity rates in Jordan have reached alarming levels, with 28% of school-age children classified as overweight or obese. School-based gardening interventions show promise for promoting healthy eating behaviors, yet limited research exists in Middle Eastern contexts. This study evaluated the effectiveness of a five-month school-based vegetable gardening and nutrition education intervention on anthropometric measures, dietary intake, and knowledge, attitudes, and practices (KAP) regarding vegetable consumption among Jordanian primary school children. Methods: A quasi-experimental controlled trial was conducted with 216 students (ages 10–12 years) from two demographically matched schools in Amman, Jordan. The intervention group (n = 121) participated in weekly one-hour gardening sessions combined with nutrition education and vegetable tasting activities over five months, while the control group (n = 95) continued the standard curriculum. Outcomes measured at baseline and post-intervention included anthropometric assessments, dietary intake via 24 h recalls, and vegetable-related KAP using a validated questionnaire. Data were analyzed using paired t-tests and repeated measures ANCOVA. Results: The intervention group demonstrated significant improvements in body composition, including reductions in BMI (−1.57 kg/m²), weight (−1.88 kg), and BMI z-score (−0.37), while controls showed minimal increases. Vegetable intake showed significant time × group interaction (p-value = 0.003), with a non-significant increase in the intervention group (2.7 to 2.9 times/day) and a non-significant decrease in the controls (2.5 to 2.4 times/day). Dietary quality improved, including increased fiber intake (+2.36 g/day) and reduced saturated fat consumption (−9.24 g/day). Nutrition knowledge scores increased substantially in the intervention group (+22.31 points) compared to controls (+1.75 points; p-value ≤ 0.001). However, attitudes and practices toward vegetable consumption showed no significant changes. Conclusions: This intervention effectively improved body composition, dietary quality, and nutrition knowledge among Jordanian primary school children. These findings provide evidence for implementing culturally adapted school gardening programs as childhood obesity prevention interventions in Middle Eastern settings, though future programs should incorporate family engagement strategies to enhance behavioral sustainability. Full article

Agricultural drainage canals provide critical habitats for fish species that are highly sensitive to agricultural practices. However, conventional monitoring methods such as capture surveys are invasive and labor-intensive, which means they can disturb fish populations and hinder long-term ecological assessment. Therefore, there is a strong need for effective and non-invasive monitoring techniques. In this study, we developed a practical method using the YOLOv8n deep learning model to automatically detect and quantify fish occurrence in underwater images from a canal in Ibaraki Prefecture, Japan. The model showed high performance in validation (F1-score = 91.6%, Precision = 95.1%, Recall = 88.4%) but exhibited reduced performance under real field conditions (F1-score = 61.6%) due to turbidity, variable lighting, and sediment resuspension. By correcting for detection errors, we estimated that approximately 7300 individuals of Pseudorasbora parva and 80 individuals of Cyprinus carpio passed through the observation site during a seven-hour monitoring period. These findings demonstrate the feasibility of deep learning-based monitoring to capture temporal patterns of fish occurrence in agricultural drainage canals. This approach provides a promising tool for sustainable aquatic ecosystem management in agricultural landscapes and emphasizes the need for further improvements in recall under turbid and low-visibility conditions. Full article

Objective: To evaluate the feasibility and effectiveness of wrist-worn wearable devices combined with machine learning (ML) approaches for detecting a diverse array of seizure types beyond generalized tonic–clonic (GTC), including focal, generalized, and subclinical seizures. Materials and Methods: Twenty-eight patients undergoing inpatient video-EEG monitoring at Mayo Clinic were concurrently monitored using Empatica E4 wrist-worn devices. These devices captured accelerometry, blood volume pulse, electrodermal activity, skin temperature, and heart rate. Seizures were annotated by neurologists. The data were preprocessed to experiment with various segment lengths (10 s and 60 s) and multiple feature sets. Three ML strategies, XGBoost, deep learning models (LSTM, CNN, Transformer), and ROCKET, were evaluated using leave-one-patient-out cross-validation. Performance was assessed using area under the receiver operating characteristic curve (AUROC), seizure-wise recall (SW-Recall), and false alarms per hour (FA/h). Results: Detection performance varied by seizure type and model. GTC seizures were detected most reliably (AUROC = 0.86, SW-Recall = 0.81, FA/h = 3.03). Hyperkinetic and tonic seizures showed high SW-Recall but also high FA/h. Subclinical and aware-dyscognitive seizures exhibited the lowest SW-Recall and highest FA/h. MultiROCKET and XGBoost performed best overall, though no single model was optimal for all seizure types. Longer segments (60 s) generally reduced FA/h. Feature set effectiveness varied, with multi-biosignal sets improving performance across seizure types. Conclusions: Wrist-worn wearables combined with ML can extend seizure detection beyond GTC seizures, though performance remains limited for non-motor types. Optimizing model selection, feature sets, and segment lengths, and minimizing false alarms, are key to clinical utility and real-world adoption. Full article

Background/Objectives: Iodine is an important nutrient for the human body, used in the production of thyroid hormones. During pregnancy, a deficiency can cause miscarriage and hypothyroidism, while an excess can cause thyroid dysfunction. Therefore, the objective of this study was to evaluate the factors associated with the iodine nutritional status of pregnant Brazilian women. Methods: This was a cross-sectional, multicenter study conducted with pregnant women over 18 years of age, users of the Unified Health System (SUS). A semi-structured questionnaire was used to obtain sociodemographic information. Iodine status was assessed by urinary iodine concentration (UIC). The iodine content of salt and homemade and industrial seasonings was determined by the titrimetric method. Dietary intake was estimated through a 24-hour dietary recall. The chi-square test and hierarchical multinomial logistic regression were used for statistical analysis. The significance level was set at p ≤ 0.05. Results: Among Brazilian pregnant women, the median UIC was 186.7 µg/L (P25: 118.05 µg/L-P75: 280.93 µg/L). Regarding iodine nutritional status, the prevalence of deficiency was 36.7% (n = 694), above the requirement was 28.7% (n = 543), and excess iodine intake was 3.6% (n = 68). We observed that non-white pregnant women were more likely (OR = 1.83; 95% CI: 1.27–2.64) to have iodine deficiency, and those who did not work were less likely (OR = 0.71; 95% CI: 0.52–0.98). Pregnant women in the last trimester of pregnancy were less likely to have iodine intake above the requirements (OR = 0.52; 95% CI: 0.31–0.88). Conclusions: A substantial proportion of pregnant women had iodine deficiency or intake above the required level. Iodine deficiency is more chance among non-white pregnant women and less chance among those not employed during pregnancy. On the other hand, pregnant women who were in their third trimester of pregnancy were less likely to have iodine intake above the required level. Full article

Field hockey involves repetitive high-intensity movements and physical contact, posing a high risk of injury. However, studies developing injury prediction models without relying on expensive tools such as GPS remain limited. This study aimed to develop an explainable AI model that predicts injury occurrence using only simple questionnaire-based data and visually identifies key predictors. Survey data were collected from 239 adult players registered with the Korea Field Hockey Association in 2024, including university and professional team athletes. Ten variables were used: sex, team affiliation, playing experience, player level, warm-up duration, weekly training hours and days, and physical indicators (age, height, weight). Injury was defined as an event within the past year that resulted in being unable to train for more than 24 h. Logistic Regression, Random Forest, and XGBoost models were compared. The final model—Logistic Regression—underwent SHAP-based visualization for interpretability. The Logistic Regression model showed the best performance in recall (0.6810 ± 0.0983), F1-score (0.6260 ± 0.0499), and AUC (0.6515 ± 0.0393). SHAP analysis identified Group, Training Time, Weight, and Player Level as key predictors, and visualized their contributions to individual predictions. This study demonstrates that a lightweight, interpretable injury prediction model using only simple survey data can achieve practical performance. This approach offers valuable insights for real-world applications and the development of injury prevention strategies. Full article

Background: Individuals with food allergies typically need to avoid specific allergens, leading to distinct dietary choices. Their food product intake may therefore vary from that of the general population, potentially leading to differences in their intake of nutrients and other food compounds. Methods: We compared food compound and nutrient group intakes between the general Dutch adult population (n = 415) and food allergic Dutch adult patients with either milk and/or egg allergies (n = 16), peanut and/or tree nut allergies (n = 35) or a combination of milk/egg and peanut/tree nut allergies (n = 22). We translated 24-hour dietary recall data into food compound intake values. We used a mixed effects ANOVA model and considered compound intakes statistically significantly different at FDR-corrected p < 0.05. Additionally, compounds with uncorrected p < 0.01 were explored for potential relevance. Results: A total of 489 compounds or nutrient groups were included in the statistical analysis. Milk/egg and mixed allergic patients had significantly lower intakes of beta-lactose, butyric acid, caproic acid, caprylic acid, capric acid, lauric acid, myristic acid, myristoleic acid, conjugated linoleic acid, and remainder saturated fatty acids (p < 0.05, FDR corrected), with mean intake factors of 1.6–3.2 and 1.3–2.9 lower, respectively, than the general population. In addition, 36 other compounds showed intake differences with a p < 0.01 without FDR correction. There were no statistically significant differences between the peanut/tree nut allergy group and the general population. Conclusions: Our study shows significantly lower intakes of 10 mainly dairy-derived compounds by the milk/egg and mixed-allergic patients, presenting the potential for long-term health consequences and the need for supplementation a relevant consideration, warranting further research. Full article

Objective: This study aimed to evaluate the reliability and validity of a Food Frequency Questionnaire (FFQ) designed for use in epidemiological studies among populations in Fujian, China. Methods: From September to December 2023, adults aged 18 years and above residing in Fujian Province, southeastern China, were recruited via online survey promotion. Participants completed the FFQ twice with a one-month interval and also completed a 3-day 24 h dietary recall (3d-24HDR), covering two weekdays and one weekend day, during the same period. The reliability of the FFQ was assessed by comparing the average intake of food groups and nutrients between the two FFQs, using Spearman correlation coefficients, intraclass correlation coefficients (ICCs), and weighted Kappa coefficients based on tertile classification. Validity was evaluated by comparing the average intake values from the FFQs and the 3d-24HDR using similar methods, including Spearman correlation, weighted Kappa statistics, and Bland–Altman analysis. Results: A total of 152 participants completed two FFQs (for reliability assessment), and 142 participants completed the 3d-24HDR (for validity assessment). Spearman correlation coefficients for food group intake between the two FFQs ranged from 0.60 to 0.80, with ICCs ranging from 0.53 to 0.91. For energy and nutrient intake, Spearman coefficients ranged from 0.66 to 0.96, and ICCs ranged from 0.57 to 0.97. After tertile classification, less than 15% of participants were misclassified into distant categories. The weighted Kappa coefficients for food groups and nutrients ranged from 0.37 to 0.71 and 0.43 to 0.88, respectively. In comparison with the 3d-24HDR, Spearman correlations for food groups and nutrients ranged from 0.41 to 0.72 and 0.40 to 0.70, respectively. The proportion of participants classified into the same or adjacent tertile was 78.8–95.1%. Weighted Kappa coefficients and Bland–Altman plots indicated acceptable agreement between the FFQ and 3d-24HDR for most nutrients. Conclusions: The FFQ used in this study demonstrated good reliability and moderate-to-good validity. It is suitable for use in dietary assessment in gastric cancer epidemiological studies in Fujian, China. Full article

With coastal tourism depending on clean beaches and litter surveys remaining manual, sparse, and costly, this study coupled centimeter-resolution UAV imagery with a Grid R-CNN detector to automate debris mapping on five beaches of Wonsan Island, Korea. Thirty-one Phantom 4 flights (0.83 cm GSD) produced 31,841 orthoimages, while 11 debris classes from the AI Hub dataset trained the model. The network reached 74.9% mAP and 78%/84.7% precision–recall while processing 2.87 images s⁻¹ on a single RTX 3060 Ti, enabling a 6 km shoreline to be surveyed in under one hour. Georeferenced detections aggregated to 25 m grids showed that 57% of high-density cells lay within 100 m of the beach entrances or landward edges, and 86% within 200 m. These micro-patterns, which are difficult to detect in meter-scale imagery, suggest that entrance-focused cleanup strategies could reduce annual maintenance costs by approximately one-fifth. This highlights the potential of centimeter-scale GeoAI in supporting sustainable beach management. Full article

Introduction: Obesity is associated with cardiometabolic diseases, and chrononutrition has become a novel weight loss strategy. However, few have characterized chrononutrition patterns among people attempting weight loss. This study characterizes chrononutrition patterns in a nationally representative sample of U.S. adults who attempted weight loss in the past year through dietary modifications by weight change and adiposity. Methods: This cross-sectional analysis utilizes NHANES 2017–2020 data. Chrononutrition patterns were assessed using 24 h dietary recalls. Participants self-reported weight loss attempts in the past year and if they tried using diet modification. Weight change (loss, maintenance, and gain) was defined based on differences in current weight and weight one year prior. We used latent profile analysis and descriptive statistics. Results: The sample included 2107 participants who attempted weight loss in the past year through diet modification (median age 47; 58% women and 62% white). Individuals who gained weight (vs. loss) had longer hours between waketime and the first eating (1.78 vs. 1.62 h, p = 0.024), consumed a lower proportion of calories later in the day (43% vs. 52%, p < 0.001), and ate less frequently (5.20 vs. 5.43 episodes, p = 0.008). Participants with obesity had the shortest eating window (11.77 vs. 12.22 h, p = 0.02) despite a longer delay between waketime and the first eating (1.80 vs. 1.29 h, p < 0.001) and lower eating frequency (5.16 vs. 5.97, p < 0.001). Conclusions: Variations in eating timing, eating episodes, and caloric distribution suggest that chrononutrition may play a role in personalized weight management strategies. Full article

During crisis events, digital information volume can increase by over 500% within hours, with social media platforms alone generating millions of crisis-related posts. This volume creates critical challenges for emergency responders who require timely access to the concise subset of accurate information they are interested in. Existing approaches strongly rely on the power of large language models. However, the use of large language models limits the scalability of the retrieval procedure and may introduce hallucinations. This paper introduces a novel multi-stage text retrieval framework to enhance information retrieval during crises. Our framework employs a novel three-stage extractive pipeline where (1) a topic modeling component filters candidates based on thematic relevance, (2) an initial high-recall lexical retriever identifies a broad candidate set, and (3) a dense retriever reranks the remaining documents. This architecture balances computational efficiency with retrieval effectiveness, prioritizing high recall in early stages while refining precision in later stages. The framework avoids the introduction of hallucinations, achieving a 15% improvement in BERT-Score compared to existing solutions without requiring any costly abstractive model. Moreover, our sequential approach accelerates the search process by 5% compared to the use of a single-stage based on a dense retrieval approach, with minimal effect on the performance in terms of BERT-Score. Full article

Effective incident detection is essential for emergency response and transportation management. Traditional methods relying on stationary technologies are often costly and provide limited coverage, prompting the exploration of crowdsourced data such as Waze. While Waze offers extensive coverage, its data can be unverified and unreliable. This study aims to identify factors affecting the reliability of Waze alerts and develop a predictive model to distinguish true incidents from false alerts using real-time Waze data, thereby improving emergency response times. Real crash data from the New Jersey Department of Transportation (NJDOT) and crowdsourced data from Waze were matched using the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to differentiate true and false alerts. A binary logit model was constructed to reveal significant predictors such as time categories around peak hours, road type, report ratings, and crash type. Findings indicate that the likelihood of accurate Waze alerts increases during peak hours, on streets, and with higher report ratings and major crashes. Additionally, multiple machine learning-based predictive models were developed and evaluated to forecast in real time whether Waze alerts correspond to actual incidents. Among those models, the Random Forest model achieved the highest overall accuracy (82.5%) and F1-score (82.8%), and an Area Under the Receiver Operating Characteristic Curve (AUC-ROC) of 0.90, demonstrating its robustness and reliability for real-time incident detection. Gradient Boosting, with an AUC-ROC of 0.90 and Area Under the Precision–Recall Curve (AUC-PR) of 0.90, also performed strongly, particularly excelling at predicting true alerts. The analysis further emphasized the importance of key predictors such as time of day, report ratings, and road type. These findings provide actionable insights for enhancing the accuracy of incident detection and improving the reliability of crowdsourced traffic alerts, supporting more effective traffic management and emergency response systems. Full article

Evaluating bone tunnels is crucial for assessing functional recovery after anterior cruciate ligament reconstruction. Conventional methods are imprecise, time-consuming, and labor-intensive. This study introduces a novel deep learning-based system for accurate bone tunnel segmentation and assessment. The system has two primary stages. Firstly, the ResNet50-Unet network is employed to capture the bone tunnel area in each slice. Subsequently, in the bone texture analysis, the open-source software 3D Slicer is leveraged to execute three-dimensional reconstruction based on the segmented outcomes from the previous stage. The ResNet50-Unet network was trained and validated using a newly developed dataset named tunnel bone segmentation (TB-Seg). The outcomes reveal commendable performance metrics, with mean intersection over union (mIoU), mean average precision (mAP), precision, and recall on the validation set reaching 76%, 85%, 88%, and 85%, respectively. To assess the robustness of our innovative bone texture system, we conducted tests on a cohort of 24 patients, successfully extracting bone volume/total volume, trabecular thickness, trabecular separation, trabecular number, and volumetric information. The system excels with substantial significance in facilitating the subsequent analysis of the intricate interplay between bone tunnel characteristics and the postoperative recovery trajectory after anterior cruciate ligament reconstruction. Furthermore, in our five randomly selected cases, clinicians utilizing our system completed the entire analytical workflow in a mere 357–429 s, representing a substantial improvement compared to the conventional duration exceeding one hour. Full article

Background/Objectives: Food insecurity during pregnancy is associated with higher risks of negative physical outcomes for both mother and child. This study aims to understand experiences of food insecurity among low-income Singaporean pregnant women and its impact on nutritional status. Methods: In this cross-sectional, mixed-methods study, 49 food-insecure pregnant women were recruited from KK Women’s and Children’s Hospital between November 2021 and November 2023, among which 11 in-depth interviews were conducted. Questionnaires, anthropometric measurements, 24-Hour dietary recalls, metabolic and nutritional blood tests were conducted for all subjects. Descriptive quantitative analysis was performed and integrated with qualitative thematic analysis to explain findings. Results: On average, women were overweight pre-pregnancy (body mass index 26.1 ± 6.9 kg/m²) and had low haemoglobin and 25-hydroxyvitamin D levels. Calorie intake and intake from major food groups did not meet recommendations during pregnancy, except for “Grains”. From interviews, effects of financial constraints, how participants managed their food supply and pregnancy-related symptoms, supported findings from 24-Hour dietary recalls. Conclusions: Food insecurity led to suboptimal nutritional status and diets in Singaporean pregnant women despite appearing well-nourished. Further exploration of perspectives of food-insecure mothers, healthcare providers and welfare organisations is needed to devise long-term solutions to improve food security and alleviate malnutrition. Full article

Construction projects’ unsatisfactory performance has been linked to factors influencing individuals’ well-being and mental alertness on projects. Drowsiness is a significant indicator of sleep deprivation and fatigue, so being able to identify the cognitive and physical preparedness of workers on site to engage in construction tasks is important. As a consequence of the strenuous nature of the work involved in construction, long work hours, and environmental conditions, drowsiness is commonplace and has received less attention despite being a leading cause of accidents occurring on-site. Detecting drowsiness is essential for determining the safety and well-being of site workers. This study presents a vision-based approach using an improved version of the You Only Look Once (YOLOv8) algorithm for real-time drowsiness exposure among construction workers. The proposed method leverages computer vision techniques to analyze facial and eye features, enabling the early detection of signs of drowsiness, effectively preventing accidents, and enhancing on-site safety. The model showed significant precision and efficiency in detecting drowsiness from the given dataset, accomplishing a drowsiness class with a mean average precision (mAP) of 92%. However, it also exhibited difficulties handling imbalanced classes, particularly the underrepresented ‘Awake with PPE’ class, which was detected with high precision but comparatively lower recall and mAP. This highlighted the necessity of balanced datasets for optimal deep learning performance. The YOLOv8 model’s average mAP of 78% in drowsiness detection compared favorably with other studies employing different methodologies. The system improves productivity and reduces costs by preventing accidents and enhancing worker safety. However, limitations, such as sensitivity to lighting conditions and occlusions, must be addressed in future iterations. Full article