Sampling Distribution Research Articles

Weight gain, new-onset overweight or obesity, and their influencing factors during the social distancing era of the COVID-19 pandemic

The prevalence of obesity is gradually increasing in South Korea, and the coronavirus disease 2019 (COVID-19) pandemic has accelerated the weight gain. In response to COVID-19, South Korea implemented targeted distancing measures based on the number of COVID-19 cases, rather than a full lockdown. This study focused on two subgroups that influence the rise in obesity prevalence during the early stages of the COVID-19 pandemic and explored the factors contributing to rapid weight gain in this group during this period. This cross-sectional study involved 3,462 participants recruited through the MicroEmbrain Inc. Online survey panel. Participation emails were sent by the research company to the online survey panel, and individuals who voluntarily agreed to participate responded to the survey. The participants consisted of residents aged 19 and above from Seoul, with sample distribution taking considering sex, age, and region. Height, and weight before and during the COVID-19 pandemic were self-reported. The analysis revealed that compared with that of pre-COVID-19 statistics, 11 % of all participants entered the overweight or obese categories post COVID-19, and among those who were overweight or obese before COVID-19, 42.5 % reported weight gain. Both groups had relatively younger individuals, higher stress levels, spent more time alone, and had increased usage of online platforms including online food purchasing. Post-COVID-19 wt gain among individuals with pre-existing overweight or obesity status was associated with a higher frequency of solitary drinking (odds ratio [OR] = 1.407, 95 % confidence interval [CI] = 1.024–1.932), increased online leisure time (OR = 1.336, 95 % CI = 1.063–1.678), and increased use of meal delivery services (OR = 1.307, 95 % CI = 1.026–1.664). The prevalence of obesity has continued to rise after COVID-19 pandemic, highlighting the need for interventions that mitigate the negative impact of the pandemic on health behavior.

Enhancing identification performance of cognitive impairment high-risk based on a semi-supervised learning method

BackgroundCognitive assessment plays a pivotal role in the early detection of cognitive impairment, particularly in the prevention and management of cognitive diseases such as Alzheimer’s and Lewy body dementia. Large-scale screening relies heavily on cognitive assessment scales as primary tools, with some low sensitivity and others expensive. Despite significant progress in machine learning for cognitive function assessment, its application in this particular screening domain remains underexplored, often requiring labor-intensive expert annotations. AimsThis paper introduces a semi-supervised learning algorithm based on pseudo-label with putback (SS-PP), aiming to enhance model efficiency in predicting the high risk of cognitive impairment (HR-CI) by utilizing the distribution of unlabeled samples. DataThe study involved 189 labeled samples and 215,078 unlabeled samples from real world. A semi-supervised classification algorithm was designed and evaluated by comparison with supervised methods composed by 14 traditional machine-learning methods and other advanced semi-supervised algorithms. ResultsThe optimal SS-PP model, based on GBDT, achieved an AUC of 0.947. Comparative analysis with supervised learning models and semi-supervised methods demonstrated an average AUC improvement of 8% and state-of-art performance, repectively. ConclusionThis study pioneers the exploration of utilizing limited labeled data for HR-CI predictions and evaluates the benefits of incorporating physical examination data, holding significant implications for the development of cost-effective strategies in relevant healthcare domains.

Attention mechanism based on deep learning for defect detection of wind turbine blade via multi-scale features

Abstract An enhanced wind turbine blade surface defect detection algorithm, CGIW-YOLOv8, has been introduced to tackle the problems of uneven distribution of defect samples, confusion between defects and background, and variations in target scales that arise during drone maintenance of wind turbine blades. This algorithm is given based on the YOLOv8 model. Initially, a data augmentation method based on geometric changes and Poisson mixing was used to enrich the dataset and address the problem of uneven sample distribution. Subsequently, the incorporation of the Coordinate Attention (CA) mechanism into the Backbone network improved the feature extraction capability in complex backgrounds. In the Neck, the Reparameterized Generalized Feature Pyramid Network (Rep-GFPN) was introduced as a path fusion strategy and multiple cross-scale connections are fused, which effectively enhances the multi-scale expression ability of the network. Finally, the original CIOU loss function was replaced with Inner-WIoU, which was created by applying the Inner-IoU loss function to the Wise-IoU loss function. It improved detection accuracy while simultaneously speeding up the model's rate of convergence. Experimental results show that the mAP of the method for defect detection reaches 92%, which is 5.5% higher than the baseline network. The detection speed is 120.5 FPS, which meets the needs of real-time detection.

Radioactivity determination in depth distribution of stone samples of Kalaburagi District, Karnataka, India.

The activity concentrations of 238U, 232Th and 40K were found using high-resolution gamma-ray spectrometry on depth profile samples collected from the Kalaburagi districts. This study aims to ascertain the radioactivity Changes concerning the depth profile. With values of 51.64±0.50 Bqkg-1 for 238U, 58.77±0.23 for 232Th and 313.92±3.57 for 40K, respectively, the depth profile samples of Jayanagar in the Kalaburagi region exhibit significant activity concentrations among the measured values. Moreover, estimates of the K/Th and K/U ratios have been made, based on surface-level collected samples, which may have values that differ from the samples under study.

Characterisation of biocondensate microfluidic flow using array-detector FCS

BackgroundBiomolecular condensation via liquid-liquid phase separation (LLPS) is crucial for orchestrating cellular activities temporospatially. Although the rheological heterogeneity of biocondensates and the structural dynamics of their constituents carry critical functional information, methods to quantitatively study biocondensates are lacking. Single-molecule fluorescence research can offer insights into biocondensation mechanisms. Unfortunately, as dense condensates tend to sink inside their dilute aqueous surroundings, studying their properties via methods relying on Brownian diffusion may fail. MethodsWe take a first step towards single-molecule research on condensates of Tau protein under flow in a microfluidic channel of an in-house developed microfluidic chip. Fluorescence correlation spectroscopy (FCS), a well-known technique to collect molecular characteristics within a sample, was employed with a newly commercialised technology, where FCS is performed on an array detector (AD-FCS), providing detailed diffusion and flow information. ResultsThe AD-FCS technology allowed characterising our microfluidic chip, revealing 3D flow profiles. Subsequently, AD-FCS allowed mapping the flow of Tau condensates while measuring their burst durations through the stationary laser. Lastly, AD-FCS allowed obtaining flow velocity and burst duration data, the latter of which was used to estimate the condensate size distribution within LLPS samples. ConclusionStudying biocondensates under flow through AD-FCS is promising for single-molecule experiments. In addition, AD-FCS shows its ability to estimate the size distribution in condensate samples in a convenient manner, prompting a new way of investigating biocondensate phase diagrams. General significanceWe show that AD-FCS is a valuable tool for advancing research on understanding and characterising LLPS properties of biocondensates.

A novel domain adaptive method for gearbox fault diagnosis using maximum multiple-classifier discrepancy network

Abstract Domain adaptive gearbox fault diagnosis methods have made impressive achievements for the past several years. However, most of the traditional domain adaptive methods have significant limitations under fluctuating operating conditions. The acquired acceleration signals will result in different signal vibration spectra and peak vibration amplitudes due to the different working conditions between the source and target domains. There is an obvious discrepancy between the distribution of fault samples in the source domain and the target domain, which in turn makes it difficult to classify the target domain samples with fuzzy fault category boundaries. Therefore, how to measure the discrepancy between two distributions has been an important research direction in machine learning. A good metric helps to discover better features and build better models. In this paper, a novel domain adaptive method for gearbox fault diagnosis using maximum multiple-classifier discrepancy network (MMCDN) is proposed. The sparse stack autoencoder is used by the MMCDN as a feature extractor for fault feature extraction, and a kind of composite distance is adopted for domain discrepancy measurement of source and target domain features for domain alignment. Then the extracted features are input into a three-classifier of the model for adversarial training. The trained model classifiers have high performance in fault classification. The combination of domain adaptation and multi-classifier discrepancy output can effectively solve the impact of working condition changes and the misclassification problem for fuzzy samples with class boundaries. Experimental validation with two planetary gearbox datasets shows that the MMCDN has more favorable diagnostic accuracy and performance than other methods.

AoUPRS: A Cost-Effective and Versatile PRS Calculator for the All of Us Program.

The All of Us (AoU) Research Program provides a comprehensive genomic dataset to accelerate health research and medical breakthroughs. Despite its potential, researchers face significant challenges, including high costs and inefficiencies associated with data extraction and analysis. AoUPRS addresses these challenges by offering a versatile and cost-effective tool for calculating polygenic risk scores (PRS), enabling both experienced and novice researchers to leverage the AoU dataset for significant genomic discoveries. AoUPRS is implemented in Python and utilizes the Hail framework for genomic data analysis. It offers two distinct approaches for PRS calculation: the Hail MatrixTable (MT) and the Hail Variant Dataset (VDS). The MT approach provides a dense representation of genotype data, while the VDS approach offers a sparse representation, significantly reducing computational costs. In performance evaluations, the VDS approach demonstrated a cost reduction of up to 99.51% for smaller scores and 85% for larger scores compared to the MT approach. Both approaches yielded similar predictive power, as shown by logistic regression analyses of PRS for coronary artery disease, atrial fibrillation, and type 2 diabetes. The empirical cumulative distribution functions (ECDFs) for PRS values further confirmed the consistency between the two methods. AoUPRS is a versatile and cost-effective tool that addresses the high costs and inefficiencies associated with PRS calculations using the AoU dataset. By offering both dense and sparse data processing approaches, AoUPRS allows researchers to choose the approach best suited to their needs, facilitating genomic discoveries. The tool's open-source availability on GitHub, coupled with detailed documentation and tutorials, ensures accessibility and ease of use for the scientific community.

Deciphering the Mortar Composition of Ancient Chinese Pagodas: A Comparative Analysis from the Tang to Ming Dynasties in Shaanxi Province

The mortar materials from the internal and external walls of ancient Chinese pagodas in Shaanxi (Luoshan Temple Pagoda (847–858 AD), Daxiang Pagoda (960–1279 AD), and Qianjin Pagoda (1609 AD)) were investigated. Qualitative and semi-quantitative techniques including scanning electronic microscopy (SEM), energy-dispersive spectroscopy (EDS), X-ray diffraction (XRD), attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR) and thermogravimetric analysis with differential scanning calorimetry (TGA-DSC) were employed. The physical and chemical characteristics including microstructural morphology, chemical composition, element distribution, and organic additives of the mortar samples were studied. SEM indicated that the mortar from the external wall exhibits a tightly bonded matrix, with the irregularly shaped particles, and the carbonation of the calcium carbonate in the mortars closely resembles that of aragonite crystals. The results of EDS, XRD, FTIR and TGA-DSC revealed that mortar from external wall of these ancient pagodas were air-hardened lime, and complete carbonization with no traces of calcium hydroxide, while mortars from internal wall primarily comprises local clay and lime. In addition, ATR-FTIR and iodine-starch discoloration tests confirmed that the presence of starch in the external wall of Daxiang Pagoda, whereas no organic additives were detected in the Qianjin and Luoshan Temple Pagoda.

XGBoost-B-GHM: An Ensemble Model with Feature Selection and GHM Loss Function Optimization for Credit Scoring

Credit evaluation has always been an important part of the financial field. The existing credit evaluation methods have difficulty in solving the problems of redundant data features and imbalanced samples. In response to the above issues, an ensemble model combining an advanced feature selection algorithm and an optimized loss function is proposed, which can be applied in the field of credit evaluation and improve the risk management ability of financial institutions. Firstly, the Boruta algorithm is embedded for feature selection, which can effectively reduce the data dimension and noise and improve the model’s capacity for generalization by automatically identifying and screening out features that are highly correlated with target variables. Then, the GHM loss function is incorporated into the XGBoost model to tackle the issue of skewed sample distribution, which is common in classification, and further improve the classification and prediction performance of the model. The comparative experiments on four large datasets demonstrate that the proposed method is superior to the existing mainstream methods and can effectively extract features and handle the problem of imbalanced samples.

A robust domain distribution alignment discriminative network driven by physical samples for rotor-bearing fault diagnosis

The diagnostic performance of most intelligent models depends on having a sufficient number of labeled samples, which are difficult to obtain in real-world scenarios. Physics-guided sample generation emerges as a promising remedy for addressing the scarcity of labeled samples during model training. However, existing studies have tended to overlook the coupling effect between the marginal and conditional distributions of physical and real samples, leading to domain confusion and a subsequent reduction in model generalization and diagnostic accuracy. Therefore, this study introduces a novel approach: the physically driven sample-based domain distribution alignment discriminative network (DDADN), which integrates dynamic modeling and domain adaptation. Specifically, a lumped parameter model is established to generate complete defect samples, subsequently optimized using the Pearson correlation coefficient (PCC). Next, an enhanced joint distribution adaptation (EJDA) mechanism is employed to harmonize distributions across domains, effectively aligning both overall and class-specific distributions within the domains and thus achieving adaptive domain confusion. Meanwhile, to ensure robust feature classification and extraction, an adaptive softmax (A-softmax) loss function is introduced. Finally, in unlabeled real operating conditions, the proposed multitask scheme adeptly transfers knowledge between physical and real signals, yielding impressive classification accuracies of 97.78% and 98.34% for the experimental and Case Western Reserve University (CWRU) datasets, respectively. These results underscore the superiority of the proposed scheme in terms of learning transferable features and maintaining high diagnostic accuracy and generalization capability.

Analysis of compression response and coordinated deformation mechanism of bi-crystal Al-Cu-Mg-Ag alloy using CPFEM

In this paper, the compression behavior of Al-Cu-Mg-Ag bi-crystal samples at room temperature under different orientation combinations was studied. The stress-strain distribution of bi-crystal samples during the compression process was studied based on the crystal plastic finite element method (CPFEM). The deformation mismatch and slip transfer caused by the orientation combination of bi-crystal are discussed. The results show that the Al-Cu-Mg-Ag alloy bi-crystal samples with different orientations exhibit an anisotropic hardening response under compression. The combination of soft and hard orientation can coordinate the dislocation drive of the yield point, delay the elastoplastic transition of the yield point, and greatly improve the strength of the alloy. The mechanism of high strength of heterogeneous structure is proposed in this paper, which provides an important reference for the design of high strength heat resistant aluminum alloy for automobiles.

Distribution and abundance of ichthyoplankton in natural al artificial reefs in Edremit Bay, Northern Aegean Sea, Türkiye

Edremit Bay is one of the important areas for small–scale fishing in the North Aegean Sea. For this reason, more detailed ichthyoplankton studies should be continued to evaluate the fish stock status of the region, calculate the adult fish biomass, and determine the spawning period, place, and time of fish species. In addition, from 2011 to 2013, artificial reefs were constructed in Edremit Bay. Regular investigation of fish eggs and larvae in the region influenced by the Edremit Bay artificial reef would provide better knowledge of the structure and function of the local ecosystem and provide a basis for monitoring the marine ecosystem in the area. The current research examined the species abundance and distribution of surface plankton samples collected in 2015 and 2016. The evaluation also included the influence of abiotic environmental factors such as temperature, pH, chlorophyll a (chl–a), salinity, depth, and dissolved oxygen. Throughout the research, 3345 eggs and 176 larvae were analyzed. 57 species of eggs and larvae were identified. Eggs were not found in the natural reef (S3) in the summer and winter of 2015, while the highest egg abundance (212.83·100 m-3) was recorded in the natural reef (S1) in the summer of 2016. Larvae were not found in the natural reefs (S2 and S3) in the summer, autumn and winter of 2015 and in the artificial reefs (S4 and S5) in the summer and autumn in 2015 while the highest larvae abundance (8.12·100 m-3) on the artificial reef (S1) in summer 2016. Although the natural reefs had the highest number of species (30 species), the artificial reefs showed the lowest species diversity (17 species).

Influence of Long Milling Time on the Electrical Resistivity of Nanocrystalline Ni2MnSn Heusler Alloy Obtained by Mechanosynthesis.

A Ni2MnSn Heusler alloy was obtained as a single B2 phase after 12 h of mechanical milling. The influence of prolonged milling on the phase stability was analysed for milling times up to 50 h, related to mean crystallite size, lattice strain, and electrical resistivity. The nature of the powders in the milled range was found to be nanocrystalline, with a mean crystallite size of about 33 ± 2 nm. An evaluation of the internal stresses induced by milling was performed, a linear behaviour was found, and a coefficient of the internal stress increase with milling time was proposed. Particle size distributions of milled samples were analysed, and the morphology of the powders was visualised by scanning electron microscopy. The elemental distribution of milled samples was quantified by energy-dispersive X-ray spectroscopy. Electrical resistivity measurements were performed on compacted samples, and their behaviour with milling time was analysed.

Modified Greenwood statistic and its application for statistical testing

In this paper, we explore the modified Greenwood statistic, which, in contrast to the classical Greenwood statistic, is properly defined for random samples from any distribution. The classical Greenwood statistic, extensively examined in the existing literature, has found diverse and interesting applications across various domains. Furthermore, numerous modifications to the classical statistic have been proposed. The modified Greenwood statistic, as proposed and discussed in this paper, shares several key properties with its classical counterpart. Emphasizing its stochastic monotonicity within three broad classes of distributions – namely, generalized Pareto, α−stable, and Student’s t distributions – we advocate for the utilization of the modified Greenwood statistic in testing scenarios. Our exploration encompasses three distinct directions. In the first direction, we employ the modified Greenwood statistic for Gaussian distribution testing. Our empirical results compellingly illustrate that the proposed approach consistently outperforms alternative goodness-of-fit tests documented in the literature, particularly exhibiting superior efficacy for small sample sizes. The second considered problem involves testing whether the distribution of a random sample has infinite variance. The last proposition suggests using the modified Greenwood statistic for testing of a given distribution. The presented simulation study strongly supports the efficiency of the proposed approach in the considered problems. Theoretical results and power simulation studies are further validated by real data analysis.

IC points weight learning-based GCN and improving feature distribution for industrial fault diagnosis

Industrial fault diagnosis (FD) is crucial to detect the causes of faults in a timely manner and solve the problems to maintain stable production. Adequately considering the spatial structure of the samples and their features is an urgent problem for better FD. This paper focused on the points located in the within-class boundaries and between-class intersection region (IC points) and proposed IC point weight learning-based graph convolution network (GCN) and improving feature distribution (ICWGCN-FD) method. Firstly, ICWGCN-FD assigns appropriate weights to IC points specially allocated by graph learning layer. Secondly, a cosine distance loss term is particularly designed to make the IC points closer to the center of their respective classes and further away from their nearest neighbor points belongs to other classes. Finally, the spatial feature distribution of the overall samples is improved under the influence of the message passing function possessed by GCN, thus presenting a clearer distribution. Two cases from industrial simulation are carried out to validate the practical feasibility and superiority of ICWGCN-FD, compared with other classic and recent graph neural networks. In addition, the 3D visualization results correspond to the principles of the method, and an ablation experiment successfully verified the validity of each component.

Image Feature Extraction Using Symbolic Data of Cumulative Distribution Functions

Symbolic data analysis is an emerging field in statistics with great potential to become a standard inferential technique. This research introduces a new approach to image feature extraction using the empirical cumulative distribution function (ECDF) and distribution function of distribution values (DFDV) as symbolic data. The main objective is to reduce the dimension of huge pixel data by organizing them into more coherent pixel-intensity distributions. We propose a partitioning method with different breakpoints to capture pixel intensity variations effectively. This results in an ECDF representing the proportion of pixel intensities and a DFDV representing the probability distribution at specific points. The novelty of this approach lies in using ECDF and DFDV as symbolic features, thus summarizing the data and providing a more informative representation of the pixel value distribution, facilitating image classification analysis based on intensity distribution. The experimental results underscore the potential of this method in distinguishing image characteristics among existing image classes. Image features extracted using this approach promise image classification analysis with more informative image representations. In addition, theoretical insights into the properties of DFDV distribution functions are gained.

Obtaining Conservative Estimates of Integrated Profitability for a Single-Period Product in an Own-Branding-and-Manufacturing Enterprise with Multiple Owned Channels

The achievable capacity index (ACI) is a simple and efficient approach for estimating the profitability of newsboy-type products, wherein profitability is defined as the probability of achieving the target profit by optimizing the order quantity. At present, the ACI is applicable to single retail stores (i.e., single demand) but not to multiple sales channels (i.e., multiple demand). This paper presents an integrated achievable capacity index (IACI) by which to measure the aggregate profitability of multiple mutually independent channels under normally distributed demand. An unbiased IACI estimator is also developed, to which is applied the Taylor expansion to approximate its sampling distribution, wherein the sizes, means, and variances of demand differ in each channel. Furthermore, overestimates due to sampling error are avoided by deriving the lower confidence bound for the IACI. This paper also provides generic tables to aid managers seeking conservative estimates of profitability. The applicability of the proposed scheme is demonstrated numerically using a real-world example involving an own-branding-and-manufacturing (OBM) enterprise with multiple owned channels.

Determination of sample size for a multinomial model coupled with the phenology model

Predicting the timing of phenological events is important in agriculture, especially high-revenue products. A project sponsored by USDA-ARS had the objective of adapting a previously developed model for estimating proportions of insects in different development stages as a function of temperature (degree) and time (days) for predicting bloom in almond orchards. Data for the model normally form a two-way table of counts, with rows corresponding to sample percentages of different development stages and columns to sampling times. In this study, we report a technique developed to estimate sample sizes of multinomial and product multinomial models using a method of moments and determine the empirical coverage of sample size. This study aims to determine an appropriate sample size for data collection. This involves establishing a sampling distribution for the Pearson statistic, defined as the product of the sample size and the deviance of empirical proportions from population proportions. The intended outcome is to predict the optimal timing for harvesting crops at desired development stages when coupled with the phenology model, for which variability of the maximum likelihood estimates of the phenology model depends on sample size.

A high reticulocyte count is a risk factor for the onset of metabolic dysfunction-associated steatotic liver disease: Cross-sectional and prospective studies of data of 310,091 individuals from the UK Biobank.

Background and Aims: Metabolic dysfunction-associated steatotic liver disease (MASLD) poses a considerable health risk. Nevertheless, its risk factors are not thoroughly comprehended, and the association between the reticulocyte count and MASLD remains uncertain. This study aimed to explore the relationship between reticulocyte count and MASLD. Methods: A total of 310,091 individuals from the UK Biobank were included in this cross-sectional study, and 7,316 individuals were included in this prospective study. The cross-sectional analysis categorized reticulocyte count into quartiles, considering the sample distribution. Logistic regression models examined the connection between reticulocyte count and MASLD. In the prospective analysis, Cox analysis was utilized to investigate the association. Results: Our study findings indicate a significant association between higher reticulocyte count and an elevated risk of MASLD in both the cross-sectional and prospective analyses. In the cross-sectional analysis, the adjusted odds ratios (ORs) of MASLD increased stepwise over reticulocyte count quartiles (quartile 2: OR 1.22, 95% CI 1.17-1.28, p < 0.001; quartile 3: OR 1.44; 95% CI 1.38-1.51, p < 0.001; quartile 4: OR 1.66, 95% CI 1.59-1.74, p < 0.001). The results of prospective analyses were similar. Conclusion: Increased reticulocyte count was independently associated with a higher risk of MASLD. This discovery offers new insights into the potential of reticulocytes as biomarkers for MASLD.

A Deficit of Massive White Dwarfs in Gaia Astrometric Binaries

The third data release of Gaia introduced a large catalog of astrometric binaries, out of which about 3200 are likely main-sequence stars with a white dwarf (WD) companion. These binaries are typically found with orbital separations of ∼1 au, a separation range that was largely unexplored due to observational challenges. Such systems are likely to have undergone a phase of stable mass transfer while the WD progenitor was on the asymptotic giant branch. Here we study the WD mass distribution of a volume-complete sample of binaries with K/M dwarf primaries and orbital separations of ∼1 au. We find that the number of massive WDs relative to the total number of WDs in these systems is smaller by an order of magnitude compared to their occurrence among single WDs in the field. One possible reason can be an implicit selection of the WD mass range if these are indeed post-stable-mass-transfer systems. Another reason can be the lack of merger products in our sample compared to the field, due to the relatively tight orbital separations of these systems. In addition, we find that about 14% of these systems have distant tertiary companions within 1 pc.