Search Results (1,093)

The relentless drive to meet global demand for poultry products has pushed for rapid intensification in chicken farming, dramatically boosting efficiency and yield. Yet, these gains have exposed a host of complex welfare challenges that have prompted scientific scrutiny and ethical reflection. In this review, I critically evaluate recent innovations aimed at mitigating such concerns by drawing on advances in behavioral science and digital monitoring and insights into biological adaptations. Specifically, I focus on four interconnected themes: First, I spotlight the complexity of avian sensory perception—encompassing vision, auditory capabilities, olfaction, and tactile faculties—to underscore how lighting design, housing configurations, and enrichment strategies can better align with birds’ unique sensory worlds. Second, I explore novel tools for gauging emotional states and cognition, ranging from cognitive bias tests to developing protocols for identifying pain or distress based on facial cues. Third, I examine the transformative potential of computer vision, bioacoustics, and sensor-based technologies for the continuous, automated tracking of behavior and physiological indicators in commercial flocks. Fourth, I assess how data-driven management platforms, underpinned by precision livestock farming, can deploy real-time insights to optimize welfare on a broad scale. Recognizing that climate change and evolving production environments intensify these challenges, I also investigate how breeds resilient to extreme conditions might open new avenues for welfare-centered genetic and management approaches. While the adoption of cutting-edge techniques has shown promise, significant hurdles persist regarding validation, standardization, and commercial acceptance. I conclude that truly sustainable progress hinges on an interdisciplinary convergence of ethology, neuroscience, engineering, data analytics, and evolutionary biology—an integrative path that not only refines welfare assessment but also reimagines poultry production in ethically and scientifically robust ways. Full article

Background: Trends in global ageing underscore the rising burden of age-related cognitive decline and concomitant cardiometabolic diseases, including type 2 diabetes mellitus (T2DM). Carnosine, a naturally occurring dipeptide with anti-inflammatory, antioxidant and anti-glycating properties, has shown promise in animal models and limited human studies for improving cognitive function, insulin resistance and T2DM, but its therapeutic effects on cognition remain unclear. The aim of this study is to assess the effects of carnosine on cognitive function in individuals with prediabetes or well-controlled T2DM. Methods: This is a secondary analysis of a double-blind randomised controlled trial (RCT), whereby 49 adults with prediabetes or early-stage well-controlled T2DM were randomised to receive 2 g of carnosine or identical placebo daily for 14 weeks. At baseline and follow-up, cognitive function was assessed as a secondary outcome using the Digit-Symbol Substitution Test, Stroop test, Trail Making Tests A & B, and the Cambridge Automated Neuropsychological Test Battery (CANTAB). Results: In total, 42 adults (23 males and 19 females) completed the trial. There were no differences in participant anthropometry or cognitive functioning between carnosine and placebo groups at baseline (all p > 0.1). After the 14-week supplementation period, there were no differences between carnosine and placebo groups in change and follow-up values for any cognitive measures including Stroop, Digit Symbol Substitution Sest, Trail Making A/B or CANTAB (all p > 0.05). Adjustments for baseline cognitive scores, diabetic status, level of education, age or interaction effects with participants’ sex did not change the results. Conclusions: Carnosine supplementation did not improve cognitive measures in individuals with prediabetes or T2DM in this study. While larger trials may provide further insights, alternative factors—such as the relatively young and healthy profile of our cohort—may have contributed to the lack of observed effect. Future research should examine individuals with existing cognitive impairment or those at higher risk of cognitive decline to better define the therapeutic potential of carnosine in this context. Full article

Background: Radiomic features have been extensively used with machine learning and other Artificial Intelligence methods in medical imaging problems. Coronavirus Disease 2019 (COVID-19), which has been spreading worldwide since 2020, has motivated scientists to develop automatic COVID-19 recognition systems, to enhance the clinical routine in overcrowded hospitals. Purpose: To develop an automated system of recognizing COVID-19 and Community-Acquired Pneumonia (CAP) using radiomic features extracted from whole lung chest Computed Tomography (CT) images. Radiomic feature extraction from whole lung CTs simplifies the image segmentation for the malignancy region of interest (ROI). Methods: In this work, we used radiomic features extracted from CT images representing whole lungs to train various machine learning models that are capable of identifying COVID-19 images, CAP images and healthy cases. The CT images were derived from an open access data set, called COVID-CT-MD, containing 76 Normal cases, 169 COVID-19 cases and 60 CAP cases. Results: Four two-class models and one three-class model were developed: Normal–COVID, COVID–CAP, Normal–CAP, Normal–Disease and Normal–COVID–CAP. Different algorithms and data augmentation were used to train each model 20 times on a different data set split, and, finally, the model with the best average performance was selected for each case. The performance metrics of Accuracy, Sensitivity and Specificity were used to assess the performance of the different systems. Since COVID-19 and CAP share similar characteristics, it is challenging to develop a model that can distinguish these diseases. Result: The results were promising for the models finally selected for each case. The accuracy for the independent test set was 83.11% in the Normal–COVID case, 88.77% in the COVID–CAP case, 93.97% in the Normal–CAP case and 94.13% in the Normal–Disease case, when referring to two-class cases, while, in the three-class case, the accuracy was 78.55%. Conclusion: The results obtained suggest that radiomic features extracted from whole lung CT images can be successfully used to distinguish COVID-19 from other pneumonias and normal lung cases. Full article

Objectives: The erythrocyte sedimentation rate (ESR) is a diagnostic test that is employed worldwide to assess a patient’s inflammatory status. Like all laboratory tests, it requires reference intervals (RIs) to support clinical decision making and facilitate accurate diagnosis. In this study, we aimed to generate RIs for the automatic analyzers VES-MATIC 5 (VM5) and CUBE 30 touch (C30T) compared to the gold standard method. Methods: A total of 989 (presumably healthy) participants from Arezzo Hospital in Italy were enrolled. The ESR RIs were established according to CLSI for all three methods. Results: The analysis pointed out significant differences between women and men and age-related increases in ESRs obtained via all three analytical methods. The average and median values resulting from VM5 and C30T were, respectively, 1 mm/h smaller and higher than the gold standard. The RIs were calculated based on three clusters: the first pertained to patients aged ≥ 18 but ≤ 49 years; the second pertained to patients aged ≥ 50 but ≤ 69 years; the last comprised patients aged ≥ 70 years. Due to the clear overlap between these ranges and the statistical analysis, we identified only one range for females ≥ 18 years (Westergren: 1–22 mm/h; VM5: 1–22 mm/h; C30T: 1–25 mm/h). For the male participants, two separate RIs were proposed: one for those aged ≥ 18 but < 69 years (Westergren: 1–14 mm/h; VM5: 1–14 mm/h; C30T: 1–18 mm/h) and one for those aged 70 years or above (Westergren: 1–22 mm/h; VM5: 1–23 mm/h; C30T: 1–29 mm/h). Conclusions: The proposed RI for automated analyzers C30T and VM5 agreed with the reference method and can be adopted to measure ESRs within EDTA blood samples. Full article

Transportation agencies often rely on manual surveys to monitor seat belt compliance; however, these methods are limited by surveyor fatigue, reduced visibility due to tinted windows or low lighting, and restricted geographic coverage, making manual surveys prone to errors and unrepresentative of the broader driving population. This paper presents an automated seat belt detection system leveraging the YOLO11 neural network on video footage captured by a tethered uncrewed aerial vehicle (UAV). The objectives are to (1) develop a robust system for detecting seat belt use at stop-controlled intersections, (2) evaluate factors affecting detection accuracy, and (3) demonstrate the potential of UAV-based compliance monitoring. The model was tested in real-world scenarios at a single-lane and a complex multi-lane stop-controlled intersection in Iowa. Three studies examined key factors influencing detection accuracy: (i) seat belt–shirt color contrast, (ii) sunlight direction, and (iii) vehicle type. System performance was compared against manual video review and large language model (LLM)-assisted analysis, with assessments focused on accuracy, resource requirements, and computational efficiency. The model achieved a mean average precision (mAP) of 0.902, maintained high accuracy across the three studies, and outperformed manual methods in reliability and efficiency while offering a scalable, cost-effective alternative to LLM-based solutions. Full article

This study presents an automated computer vision system for assessing the cleanliness of plastic mirror caps used in the automotive industry after a washing process. These components are highly visible and require optimal surface conditions prior to painting, making the detection of residual contaminants critical for quality assurance. The system acquires high-resolution monochrome images under various lighting configurations, including natural light and infrared (IR) at 850 nm and 940 nm, with different angles of incidence. Four blob detection algorithms—adaptive thresholding, Laplacian of Gaussian (LoG), Difference of Gaussians (DoG), and Determinant of Hessian (DoH)—were implemented and evaluated based on their ability to detect surface impurities. Performance was assessed by comparing the total detected blob area before and after the cleaning process, providing a proxy for both sensitivity and false positive rate. Among the tested methods, adaptive thresholding under 30° natural light produced the best results, with a statistically significant z-score of +2.05 in the pre-wash phase and reduced false detections in post-wash conditions. The LoG and DoG methods were more prone to spurious detections, while DoH demonstrated intermediate performance but struggled with reflective surfaces. The proposed approach offers a cost-effective and scalable solution for real-time quality control in industrial environments, with the potential to improve process reliability and reduce waste due to surface defects. Full article

Low carbon dioxide (CO₂) emissions make nuclear energy crucial in decarbonizing the economy. In this context, nuclear safety, and especially the operation of nuclear power plants, remains a critical issue. This article presents a new fractal cluster method of control that improves the quality of assessing fuel element cladding integrity, which is critical for nuclear and environmental safety. The proposed non-destructive testing method allows for detecting defects on the inner and outer cladding surfaces without removing the elements from the nuclear reactor, which ensures prompt response and prevention of radiation leakage. Studies have shown that the fractal dimension of the cladding surface, which varies from 2.1 to 2.5, indicates significant heterogeneity caused by mechanical damage or corrosion, which can affect its integrity. The density analysis of defect clusters allows quantifying their concentration per unit area, which is an important indicator for assessing the risks associated with the operation of nuclear facilities. The data obtained are used to assess the impact of defects on the vessel’s integrity and, in turn, on nuclear safety. The monitoring results are transmitted in real time to the operator’s automated workstation, allowing for timely decision making to prevent radioactive releases and improve environmental safety. The proposed method is a promising tool for ensuring reliable quality control of the fuel element cladding condition and improving nuclear and environmental safety. While the study is based on VVER-1000 reactor data, the flexibility of the proposed methodology suggests its potential applicability to other reactor types, opening avenues for broader implementation in diverse nuclear systems. Full article

Objective: To determine if prenatal socioeconomic status (SES) is associated with childhood working memory (WM), we constructed a more precise, integrative measure of WM using variables from multiple tasks that may provide a more representative measure of WM. Study Design: We used data from a prospective birth cohort study in Mexico City, Mexico, with N = 515 children aged 6–9 years. Prenatal SES was measured using the Mexican Association of Marketing Research and Public Opinion Agencies (AMAI) index. We created a latent variable for nonverbal working memory using multiple tasks (Cambridge Neuropsychological Test Automated Battery spatial working memory, operant chamber Delayed Match to Sample and Incremental Repeated Acquisition). Structural equation models were used to assess associations between prenatal SES and nonverbal working memory, adjusting for child demographics (e.g., age and sex), prenatal exposures (e.g., exposures to lead, arsenic, and secondhand smoke), and family (current SES, maternal IQ) variables. Results: Children had a mean age of 6.6 years [SD 0.6], and 50.5% were boys. Using confirmatory factor analysis, we constructed a latent variable of nonverbal working memory, which was measurement invariant across child sex. Prenatal SES was associated with childhood nonverbal working memory (standardized factor loading = 0.17; p = 0.004). These associations were modified by child sex. Higher prenatal SES was significantly associated with higher childhood WM in females (standardized factor loading = 0.26; p = 0.002), but not in males. Conclusions: Prenatal socioeconomic status is a predictor of childhood working memory, but it may be a stronger predictor for girls compared with for boys. Full article

Antepartum Cardiotocography (CTG) is a biomedical sensing technology widely used for fetal health monitoring. While the visual interpretation of CTG traces is highly subjective, with the inter-observer agreement as low as 29% and a false positive rate of approximately 60%, the Dawes–Redman system provides an automated approach to fetal well-being assessments. However, it is primarily designed to rule out adverse outcomes rather than detect them, resulting in a high specificity (90.7%) but low sensitivity (18.2%) in identifying fetal distress. This paper introduces PatchCTG, an AI-enabled biomedical time series transformer for CTG analysis. It employs patch-based tokenisation, instance normalisation, and channel-independent processing to capture essential local and global temporal dependencies within CTG signals. PatchCTG was evaluated on the Oxford Maternity (OXMAT) dataset, which comprises over 20,000 high-quality CTG traces from diverse clinical outcomes, after applying the inclusion and exclusion criteria. With extensive hyperparameter optimisation, PatchCTG achieved an AUC of 0.77, with a specificity of 88% and sensitivity of 57% at Youden’s index threshold, demonstrating its adaptability to various clinical needs. Its robust performance across varying temporal thresholds highlights its potential for both real-time and retrospective analysis in sensor-driven fetal monitoring. Testing across varying temporal thresholds showcased it robust predictive performance, particularly with finetuning on data closer to delivery, achieving a sensitivity of 52% and specificity of 88% for near-delivery cases. These findings suggest the potential of PatchCTG to enhance clinical decision-making in antepartum care by providing a sensor-based, AI-driven, objective tool for reliable fetal health assessment. Full article

Masonry-lined tunnels form a vital part of the world’s operational railway networks. However, in many cases their structural condition is deteriorating, so it is vital to undertake regular condition assessments to ensure their safety. In order to reduce costs and improve the repeatability of these assessments, automated deep learning-based tunnel analysis workflows have been proposed. However, for such methods to be applied in practice to a safety-critical situation, it is necessary to validate their conclusions. This study analysed how uncertainty quantification methods can be used to assess the test time performance of neural networks trained for masonry joint segmentation without the laborious labelling of additional ground truths. It applies test-time augmentation (TTA) and Monte Carlo dropout (MCD) to evaluate both the aleatoric and epistemic uncertainties of a selection of trained models. It then shows how these can be used to generate uncertainty maps to aid an engineer’s interpretation of the neural network output. Full article

This paper presents a comparative study of binarization techniques for automated defect detection in dye penetrant testing (DPT) images. We evaluate established methods, including global, adaptive, and histogram-based thresholding, against three novel machine learning-assisted approaches, Soft Binarization (SoBin), Delta Binarization (DeBin), and Convolutional Autoencoder Binarization (AutoBin), using a real-world dataset from an automated DPT system inspecting stainless steel pipes. Performance is assessed with both pixel-level and region-level metrics, with particular emphasis on the influence of defect saturation. Defect saturation is quantified as the mean saturation value of all pixels belonging to a given defect, and defects are grouped into ten categories spanning from low (60–68) to high (132–140) mean saturation. Our results demonstrate that for lower mean defect saturation values, methods such as AutoBin_Triangle, HSV_global_70, and SoBin achieve superior Intersection over Union (IoU) and high true positive rates. In contrast, methods based primarily on global thresholding of the saturation channel tend to perform competitively on images with higher defect saturation levels, reflecting their sensitivity to stronger color signals. Moreover, depending on the method, nearly perfect region-level true positive rates (${\mathrm{TPR}}_{\mathrm{region}}$) or minimal false positive rates (${\mathrm{FPR}}_{\mathrm{region}}$) can be attained, emphasizing the trade-off that different models offer distinct strengths and weaknesses, which necessitates selecting the optimal method based on the specific quality control requirements and risk tolerances of the industrial process. These findings underscore the critical importance of defect saturation as a cue for both human and computer vision systems and provide valuable insights for developing robust automated quality control and predictive quality algorithms. Full article

Background/Objectives: This study aimed to evaluate the clinical interchangeability of intraocular pressure (IOP) measurements between a non-contact tonometer (NCT), the TRK-3 OMNIA, and Goldmann applanation tonometry (GAT) and to assess the repeatability of TRK-3 measurements. Methods: This prospective, multicenter study included 120 subjects stratified into three IOP groups based on GAT measurements: low IOP (7–16 mmHg), intermediate IOP (>16 to <23 mmHg), and high IOP (≥23 mmHg). The study eye was randomly selected from each subject. IOP was measured using both TRK-3 OMNIA and GAT following a standardized protocol. Agreement between the two methods was evaluated using Bland–Altman analysis, 95% limits of agreement (LoA), and equivalence testing via the two one-sided test (TOST) approach with a predefined ±5 mmHg margin. Linear regression analysis was performed to characterize the relationship between TRK-3 and GAT measurements. The repeatability of TRK-3 measurements was assessed using the intraclass correlation coefficient (ICC), repeatability limit, and coefficient of variation (CV). Results: Across all subjects, the mean difference between TRK-3 OMNIA and GAT IOP measurements was −0.2 mmHg. TRK-3 OMNIA overestimated IOP in the low IOP group (mean difference: 2.1 mmHg, LoA: −1.2 to 5.4 mmHg) and underestimated in the high IOP group (mean difference: −2.4 mmHg, LoA: −5.9 to 1.1 mmHg), while agreement was highest in the intermediate IOP group (−0.2 mmHg, LoA: −2.9 to 2.5 mmHg). Despite the systematic trend, equivalence testing confirmed statistical equivalence across all groups, with 90% confidence intervals (CI) of 1.7 to 2.5 mmHg (low IOP group), −0.6 to 0.2 mmHg (intermediate IOP group), and −2.9 to −2.0 mmHg (high IOP group). Linear regression analysis found a strong correlation (r = 0.92) between TRK-3 and GAT. The TRK-3 OMNIA demonstrated good repeatability, with an ICC of 0.94, a repeatability limit of 3.12 mmHg, and a CV of 5.65%. The repeatability limits were 2.22 mmHg, 2.60 mmHg, and 4.19 mmHg in the low, intermediate, and high IOP groups, respectively. Conclusions: TRK-3 OMNIA and GAT measurements showed strong agreement, statistical equivalence, and a high correlation, supporting their clinical interchangeability. TRK-3 also demonstrated high repeatability. These findings indicate that this automated non-contact tonometer provides reliable and repeatable IOP measurements, supporting its integration into routine clinical workflows. Full article

Web accessibility is essential for e-government in the Philippines to ensure that all citizens, including those with disabilities, can access important information and services. This study evaluates government web accessibility using the Web Content Accessibility Guidelines 2.0 from the World Wide Web Consortium and web presence based on the Government Website Template Design guidelines. A combination of automated testing tools and visual inspections was used for the assessment. Results showed significant discrepancies between web presence and web accessibility. Web presence compliance ranged from 28% to 82.67%, averaging 53.43%. Web accessibility scored higher, with compliance rates ranging from 62.32% to 97.1% and an average of 82.5%. This indicates that while many government agencies have focused on accessibility, there is a need to improve their digital services and visibility. A well-structured and user-friendly website is vital. However, without expanded online services, mobile accessibility, and transactional features, the full potential of digital governance remains untapped. Future studies are directed to aid government agencies with adopting accessible design principles, conducting regular audits, collaborating with disability advocacy groups, and integrating assistive technologies to foster a more inclusive and efficient digital government ecosystem. Full article

This study details the design and testing of a custom test rig for evaluating the performance of both open and shrouded rotors. The rig includes a two-axis load cell that is directly connected to the rotor to measure the rotor thrust separated from the total thrust when testing shrouded rotors and ensure accurate torque measurements, independent of external structural influences. Moreover, a main load cell is used to measure the total thrust for both configurations (open and shrouded rotor), as it is connected to the entire setup. Rotor RPM is monitored by capturing the voltage frequency from the BLDC motor, controlled using a Pololu Maestro Controller through the electronic speed controller. A shunt resistance is used to calculate the current through the electric Brushless Direct Current (BLDC) motor and by measuring the voltage, the electric power is calculated. By combining both mechanical and electrical power measurements, the BLDC motor’s efficiency is calculated. Automated data collection is conducted using National Instruments DAQ systems, with averaged measurements of thrust, torque, RPM, current, and voltage. Two rotors are tested to obtain performance data for both open and shrouded configurations. Additionally, a computational study is carried out to account for the aerodynamic effects of the rig’s structural elements. Uncertainty analysis is employed to assess the reliability of the experimental results by quantifying the numerical errors associated with both random and systematic errors encountered during the rotor’s performance evaluation. Full article

Background/Objectives: To develop an automated deep learning-based bone age prediction model using the Tanner–Whitehouse (TW3) method and evaluate its feasibility by comparing its performance with that of pediatric radiologists. Methods: The hand and wrist radiographs of 560 Korean children and adolescents (280 female, 280 male, mean age 9.43 ± 2.92 years) were evaluated using the TW3-based model and three pediatric radiologists. Images with bony destruction, congenital anomalies, or non-diagnostic quality were excluded. A commercialized AI solution built upon the Rotated Single Shot MultiBox Detector (SSD) and EfficientNet-B0 was used. Bone age measurements from the model and radiologists were compared using the paired t-tests. Linear regression analysis was performed and the coefficient of determination (r²), mean absolute error (MAE), and root mean square error (RMSE) were measured. A Bland–Altman analysis was conducted and the proportion of bone age predictions within 0.6 years of the radiologists’ assessments was calculated. Results: The TW3-based model demonstrated no significant differences between bone age measurements and radiologists, except for participants <6 and >13 years old (overall, p = 0.874; 6–8 years, p = 0.737; 8–9 years, p = 0.093; 9–10 years, p = 0.301; 10–11 years, p = 0.584; 11–13 years, p = 0.976; <6 or >13 years, p < 0.001). There was a strong linear correlation between the model prediction and radiologist assessments (r² = 0.977). The RMSE and MAE values of the model were 0.529 (95% CI, 0.482–0.575) and 0.388 (95% CI, 0.361–0.417) years. Overall, 82.3% of bone age model predictions were within 0.6 years of the radiologists’ interpretation. Conclusions: Automated deep learning-based bone age assessment has the potential to reduce radiologists’ workload and provide standardized measurements for clinical decision making. Full article