Variable Reduction Research Articles

Behavioural changes to moderate heat load in grazing dairy cows under on-farm conditions

Heat stress poses an increasing risk to welfare, health and productivity of dairy cows, especially for cows on pasture. To apply timely mitigation strategies for grazing cows, simple indicators are needed that signal heat stress. We conducted an exploratory study on the behaviour of grazing dairy cows in relation to the environmental heat load on four commercial dairy farms in Switzerland with herd sizes ranging from 20 to 57 cows. In a scan-sampling procedure standing/lying, feeding/ruminating, low inter-individual distances, proximity to drinker, use of natural shade and insect infestation were observed during 30 days (5–9 days/per farm). Additionally, 10 focal cows per farm were equipped with accelerometers to analyse lying duration and locomotor activity during on average 46 days per farm. On one farm all cows (N = 57) were equipped with GPS devices which were used to calculate inter-individual distances among cows continuously during 69 days. Air temperature and relative humidity were recorded to calculate the temperature-humidity index (THI). For behaviours recorded in direct observations, a principal component analysis was performed for variable reduction. The first three principal components (PC) as well as the variables from automatic measurements were used as outcome variables in mixed effects models with daily maximum THI (THImax), time of day (continuous, in 10 min intervals) and their interaction as explanatory variables. The three PCs could be described as: “feeding and standing”, “proximity to drinker” and “standing in close proximity and seeking shade”. The daily pattern of these PC's differed by THImax (interaction time of day * THImax; all p < 0.01). On days with high THImax compared to days with lower THImax cows were seen more often close to the drinker in the morning, but not in the afternoon when they were observed standing close to each other and in the shade. On days with high THImax, cows also were lying less and increased their locomotor activity towards noon (interaction time of day * THImax; p < 0.001). Data from GPS devices confirmed the findings: On days with high THImax, cows reduced their inter-individual distances over the course of the day, while this was not observed on days with lower THImax (interaction time of day * THImax; p < 0.001). Insect infestation increased with higher THImax. We conclude that a distinct change in daily behavioural patterns, especially a reduction of lying behaviour, an increase in locomotor activity and a decrease in inter-individual distances could be used to monitor heat stress of dairy cows on pasture.

Construction and validation of the area level deprivation index for health research: A methodological study based on Nepal Demographic and Health Survey.

Area-level factors may partly explain the heterogeneity in risk factors and disease distribution. Yet, there are a limited number of studies that focus on the development and validation of the area level construct and are primarily from high-income countries. The main objective of the study is to provide a methodological approach to construct and validate the area level construct, the Area Level Deprivation Index in low resource setting. A total of 14652 individuals from 11,203 households within 383 clusters (or areas) were selected from 2016-Nepal Demographic and Health survey. The index development involved sequential steps that included identification and screening of variables, variable reduction and extraction of the factors, and assessment of reliability and validity. Variables that could explain the underlying latent structure of area-level deprivation were selected from the dataset. These variables included: housing structure, household assets, and availability and accessibility of physical infrastructures such as roads, health care facilities, nearby towns, and geographic terrain. Initially, 26-variables were selected for the index development. A unifactorial model with 15-variables had the best fit to represent the underlying structure for area-level deprivation evidencing strong internal consistency (Cronbach's alpha = 0.93). Standardized scores for index ranged from 58.0 to 140.0, with higher scores signifying greater area-level deprivation. The newly constructed index showed relatively strong criterion validity with multi-dimensional poverty index (Pearson's correlation coefficient = 0.77) and relatively strong construct validity (Comparative Fit Index = 0.96; Tucker-Lewis Index = 0.94; standardized root mean square residual = 0.05; Root mean square error of approximation = 0.079). The factor structure was relatively consistent across different administrative regions. Area level deprivation index was constructed, and its validity and reliability was assessed. The index provides an opportunity to explore the area-level influence on disease outcome and health disparity.

Insights in Hierarchical Clustering of Variables for Compositional Data

R-mode hierarchical clustering is a method for forming hierarchical groups of mutually exclusive subsets of variables. This R-mode cluster method identifies interrelationships between variables which are useful for variable selection and dimension reduction. Importantly, the method is based on metric elements defined on the sample space of variables. Consequently, hierarchical clustering of compositional parts should respect the particular geometry of the simplex. In this work, the connections between concepts such as distance, cluster representative, compositional biplot, and log-ratio basis are explored within the framework of the most popular R-mode agglomerative hierarchical clustering methods. The approach is illustrated in a paleoecological study to identify groups of species sharing similar behavior.

Regulating phonon transport in silicon nanofilms by resonant nanopillars

Nanophononic materials have shown significant potential for advancing future thermoelectrics due to their ability to achieve low thermal conductivity. Recent studies have unveiled that thermal conductivity can be further diminished through the phenomenon of phonon resonance, which emerges from the incorporation of additional pillar structures. In this investigation, we delve into the realm of phonon resonance within silicon nanofilms (SiNFs) outfitted with resonant pillars, by employing classical molecular dynamics simulations. Our study reveals that SiNFs featuring resonant pillars exhibit a variable reduction in thermal conductivity contingent upon the thickness of the SiNFs. This reduction is notably influenced by the geometry of the pillars. Remarkably, thermal conductivity can be diminished by up to 20 % for SiNFs endowed with pillars on both sides, characterized by a pillar height of 0.8 nm and a duty cycle of 50 %. This reduction significantly surpasses the corresponding decrease observed in SiNFs equipped with unilateral pillars (13 %).Furthermore, the choice of an appropriate pillar height contributes to a more substantial drop in thermal conductivity. Specifically, a reduction of 14 % is observed for the model with a pillar height of 0.8 nm, while a height of 3.1 nm leads to a more pronounced reduction of 21 %. Intriguingly, we also identified that void defects present on the pillars have the potential to disrupt phonon resonance, thereby leading to an increase in thermal conductivity. The insights garnered from our findings offer valuable guidance for the strategic design of low-dimensional materials, capitalizing on the phenomenon of phonon resonance.

Risk factors associated with development of colitis in horses post-exploratory laparotomy.

Diagnosis of colitis has been shown to impact morbidity and mortality in hospitalised horses. There are no studies to date that describe the incidence of infectious colitis after exploratory laparotomy. To investigate risk factors associated with the development of colitis and infectious colitis post-exploratory laparotomy. Retrospective case-control. Medical records of equids admitted from 2011 to 2020 were reviewed. The primary outcome was a diagnosis of colitis following exploratory laparotomy. Bivariable associations between colitis and risk factors were assessed using the 2-sample t-test and Fisher's exact test. All risk factors were subjected to a backward elimination variable reduction algorithm within a logistic regression framework (p-value set to 0.05). Odds ratios and 95% confidence intervals were computed for the final model. A total of 504 equids were included in the study. Forty-two patients (8.3%) were diagnosed with postoperative colitis. Five patients were diagnosed with Salmonella spp. and two with Clostridioides difficile. The odds of postoperative colitis were higher among patients that had pelvic flexure enterotomy (OR = 3.7, 95% CI = 1.7-7.9, p = 0.001), postoperative leukopenia or leukocytosis (OR = 21.2, 95% CI = 9.7-46.7, p < 0.001), or plasma lactate 2.0-4.0 mmol/L (OR = 3.0, 95% CI = 1.3-6.7, p < 0.008). Patients diagnosed with colitis had a longer median length of hospitalisation (9 days; range 2-21) compared with patients without colitis (7 days; range 2-25). Patients with colitis had a survival to discharge rate similar to patients without colitis (95% compared to 93%). Risk factors for infectious colitis could not be determined due to variation in testing protocols in this retrospective study and the low number of positive cases. Colitis as a postoperative complication does not negatively impact survival to discharge but is associated with longer hospitalisation. Pelvic flexure enterotomy, postoperative leukopenia or leukocytosis, and increased plasma lactate were identified as significant risk factors associated with colitis.

Detection of saffron adulteration with Crocus sativus style using NIR-hyperspectral imaging and chemometrics

Saffron is a valuable spice that is often adulterated. This study proposes using near-infrared hyperspectral imaging (NIR-HSI) and chemometrics as a fast and cost-effective method for detecting and quantifying adulteration in saffron stigmas. Adulterated saffron samples were prepared by adding Crocus sativus style to pure saffron stigmas in varying concentrations (20–90%). The spectral data were pre-treated using standard normal variate (SNV), and multiplicative scatter correction (MSC), while variable reduction was performed by Principal Component Analysis (PCA) and Partial Least Squares (PLS). Classification was done using Linear Discriminant Analysis (LDA), PLS-DA, Support Vector Machine (SVM), and Multi-layer Perceptron (MLP) models, while quantification was achieved by PLS, PCA, SVM, and MLP-based regression models. The HSI technique achieved correct classification rates of 95.6%–100% in discriminating authentic saffron from plant adulterants and adulterated saffron across all the models. Regression models to quantify the percentage style adulteration in saffron demonstrated excellent prediction abilities with almost all models achieving RPD (Residual Predictive Deviation) values of 3.0–5.4. The MLP model (1 hidden layer with 3 neurons) built from SNV pre-processed and PLS reduced data (15 LVs), showed exceptional predictive capabilities, with an R2p of 0.97, a Root Mean Squared Error of Prediction (RMSEP) of 4.3%, and an RPD of 5.4. The results demonstrate the potential of NIR-HSI and chemometrics for rapid and nondestructive detection and quantification of style in saffron stigmas.

Early Insolvency Prediction as a Key for Sustainable Business Growth

This research aimed to determine whether and how financial analysis combined with machine learning can support decision-making for sustainable business growth. This study was conducted using a sample of 100 Serbian companies whose bankruptcies were initiated between 2019 and 2021 to identify key factors that distinguish solvent from insolvent companies. Two neural networks (NNs) were trained and tested to predict these discriminating factors one year (Y-1) and two years (Y-2) before bankruptcy initiation. Initially, a total of 37 predictor variables were included, but prior to modeling, variable reduction was performed through VIF analysis and t-tests. The training dataset comprised 70% of the sample, while the remaining 30% was used for testing. Both NNs utilized a softmax activation function for the output layer and a hyperbolic tangent for the hidden layers. Two hidden layers were included, and training was conducted over 2000 epochs using the gradient descent algorithm for optimization. The research results indicate that poor cash management is the first sign of possible insolvency one year in advance. Additionally, the findings reveal that retained earnings management can serve as a reliable bankruptcy predictor two years in advance. The overall predictive accuracy of the NN models is 80.0% (Y-1) and 73.3% (Y-2) for the testing dataset. These findings demonstrate how selected ratios can support bankruptcy prediction, providing valuable insights for company proprietors, management, and external stakeholders.

High-dimensional local polynomial regression with variable selection and dimension reduction

High-dimensional local polynomial regression with variable selection and dimension reduction

Adaptive and variable model order reduction method for fracture modelling using explicit time integration

Damage modelling in modern material architectures using explicit time integration is a computationally expensive task. To reduce such expenses, model training methods based on simulation data is a suitable approach, for instance, model order reduction using proper orthogonal decomposition. However, such approaches require extensive offline training and a complete replacement of full-order modelling will hinder damage evolution that requires accurate modelling of local stress states. To overcome these problems, this works presents an adaptive reduced-order modelling framework without the need for offline training whilst having the ability to vary the level of approximations within a solution. This is achieved by adaptively calculating the reduced-bases and the hyper-reduction parameters after solving the full-order model for a few increments. Domain decomposition is enabled to separate damage modelling from reduced-order modelling and a computationally efficient method for history recovery is presented. To model the non-linearities, the reduced-order modelling parameters are updated as the solution progresses. The reduced-order modelling approximation switches to full-order modelling at regular intervals to achieve the linear momentum balance between the spatially same but dimensionally different domains. The performance of the developed method is demonstrated using benchmark composites damage modelling examples. In these examples, strong discontinuities are modelled using adaptively initiated cohesive segments and the overall framework requires minimal user intervention.

Optimizing steel coil production schedules under continuous casting and hot rolling

In continuous steel casting operations, heats of molten steel are alloyed and refined in ladles, continuously cast and cut into slabs, and hot-rolled into coils. We present a mixed-integer program that produces a daily casting schedule and that is solved using state-of-the-art software for a 100% direct-charge steel mill; two casters concurrently produce slabs, which are rolled into coils at a single hot rolling mill. This model minimizes penalties incurred by violating plant best practices while strictly adhering to safety and logical constraints to manage risk associated with manufacturing incidents. An efficient formulation, combined with variable reduction and cutting planes, expedites solutions for small instances containing hundreds of variables and thousands of constraints by factors of at least two or three (and sometimes even 100); instances an order of magnitude larger along both problem dimensions suggest solutions that reduce costs incurred using plant best practices by as much as 40%.

A Sparse Local Binary Pattern extraction algorithm applied to event sensor data for object classification

Recently, new sensors with active pixels were brought to market. These sensors export local variations of light intensity in the form of asynchronous events with low latency. Since the data output format is a stream of addressable events and not a complete image of light intensities, new algorithms are required for known problems in the field of Computer Vision, such as segmentation, VO, SLAM, object, and scene recognition. There are some proposed methodologies for object recognition using conventional methods, convolutional neural networks, and third-generation neural networks based on spikes. However, convolutional neural networks and spike neural networks require specific hardware for processing, hard to miniaturize. Also, several traditional Computer Vision operators and feature descriptors have been neglected in the context of event sensors and could contribute to lighter methodologies in object recognition. This paper proposes an algorithm for local binary pattern extraction in sparse structures, typically found in this context. This paper also proposes two methodologies using local binary patterns to captures with event-based sensors for object recognition. The first methodology exploits the known motion performed by the sensor, while the second is motion agnostic. It is demonstrated experimentally that the LBP operator is a fast and light alternative that enables variable reduction using PCA in some cases. The experiments also show that it is possible to reduce the final feature vector for classification by up to 99,73% when compared to conventional methods considered state-of-the-art while maintaining comparable accuracy.

Imperfect Preventive Maintenance Optimization with Variable Age Reduction Factor and Independent Intervention Level

Maintenance policies are crucial for ensuring the reliability, safety, and longevity of a system, as well as reducing the risk of accidents. Preventive maintenance (PM) is an effective strategy to keep equipment and systems in good working order by fixing potential issues before they cause downtime or safety hazards. However, optimizing the time intervals between PM activities is essential for minimizing the overall maintenance cost. This paper proposes an innovative approach that considers the intervention level of maintenance activities as an independent variable of PM times. The approach provides greater flexibility in creating maintenance plans, as it considers practical aspects that may impact maintenance activities beyond the time interval between PMs. The proposed approach uses a reliability model that incorporates imperfect preventive maintenance and a variable improvement factor based on age reduction. The improvement factor of each preventive maintenance activity (PMA) is defined based on the intervention level of the activity itself, which is determined by the number of tasks performed, execution time, and the number of items replaced in the maintenance plan. The proposed maintenance strategy determines not only the optimal times for PMAs and the respective intervention level but also the optimal number of maintenance activities that minimize the total maintenance cost along a fixed and user-defined planning horizon. The effectiveness and precision of the approach have been demonstrated through a series of numerical examples and a comprehensive case study involving three heat exchangers situated within the hydroelectric power plant.

Efficiency Determinants in Healthcare: A Systematic Review With an Integrated Canonical Correlation Analysis-Data Envelopment Analysis/Assurance Region Model.

Instead of considering many variables for the accurate measurement of healthcare efficiency, working with the select few variables that really affect efficiency will provide more accurate efficiency scores. In addition, calculating the efficiency by weighting the inputs and outputs according to their effect and severity levels will give more realistic results. In this article, a three-step hybrid system with a two-stage CCA (canonical correlation analysis)-DEA/AR (data envelopment analysis/assurance region) model is proposed to obtain results of health efficiency. Healthcare efficiency studies conducted between 2000 and 2020 were reviewed. In this examination of the input and output variables used in the DEA of 63 previous studies, the 6 inputs and 5 outputs preferred by previous researchers were determined. Afterward, the health efficiency scores of countries represented in the research were calculated with weight-restricted DEA, and CCA was used for a priori statistical analysis in determining the weights. Thus, in this analysis of the preferred outputs and inputs with the help of CCA to estimate the relationship between multiple input and output sets, the variables that had no effect were eliminated and the ones that had an effect were included in DEA/AR with their degree of effectiveness. For the model proposed here, three inputs and three outputs were identified by following a five-item variable reduction procedure. The numbers of doctors and nurses were identified as the most effective inputs, and infant mortality rates were found to be the most effective outputs. Therefore, health efficiency scores obtained with the proposed CCA-DEA/AR model and the basic DEA are presented together. A review of the results found fewer health-efficient countries with the weight-restricted DEA. This is proof that weighting the variables into the DEA increases the discriminating power of the method. By applying the proposed model, healthcare administrators can analyze healthcare efficiency accurately and thus improve efficiency by transferring limited resources to the right places according to deficiencies or surpluses identified by the model's inputs. Resources can be allocated at both private and public hospitals in a way that increases healthcare efficiency outputs.

Prevalence of hyponatremia and associated factors in patients with chronic kidney disease: the Fukuoka Kidney Disease Registry (FKR) study.

Hyponatremia is a common and important electrolyte disorder. However, the prevalence and factors associated with hyponatremia in patients with chronic kidney disease (CKD) are unknown. We studied the factors associated with hyponatremia (< 135mEq/L) in CKD patients registered in the Fukuoka Kidney Disease Registry (FKR) study using a logistic regression model variable selected using the variable reduction method. We analyzed the baseline characteristics of 4367 participants with CKD (age, 64 ± 16years; male, 56.1%). Hyponatremia was detected in 2.0% of the patients at baseline, and multivariate logistic analysis showed that the independent factors for hyponatremia were body mass index (odds ratio [OR] 0.91; 95% confidence interval [CI] 0.85-0.97), prescription of benzodiazepine (OR 2.31; 95% CI 1.39-3.86), blood hemoglobin level (OR 0.76; 95% CI 0.65-0.88), and serum C-reactive protein level (OR 1.27; 95% CI 1.04-1.54). The cross-sectional analysis using baseline data from the FKR study revealed independent factors associated with hyponatremia in patients with decreased kidney function. Longitudinal analyses of the FKR cohort are needed to evaluate the effects of these factors on the prognosis of hyponatremia in patients with CKD.

Hemes on a string: insights on the functional mechanisms of PgcA from Geobacter sulfurreducens

Microbial extracellular reduction of insoluble compounds requires soluble electron shuttles that diffuse in the environment, freely diffusing cytochromes, or direct contact with cellular conductive appendages that release or harvest electrons to assure a continuous balance between cellular requirements and environmental conditions. In this work, we produced and characterized the three cytochrome domains of PgcA, an extracellular triheme cytochrome that contributes to Fe(III) and Mn(IV) oxides reduction in Geobacter sulfurreducens. The three monoheme domains are structurally homologous, but their heme groups show variable axial coordination and reduction potential values. Electron transfer experiments monitored by NMR and visible spectroscopy show the variable extent to which the domains promiscuously exchange electrons while reducing different electron acceptors. The results suggest that PgcA is part of a new class of cytochromes - microbial heme-tethered redox strings - that use low-complexity protein stretches to bind metals and promote intra- and intermolecular electron transfer events through its cytochrome domains.

Effect of monitoring adherence to regular inhaled corticosteroid (ICS) alone or in combination with a long-acting β2-agonist (LABA) using electronic methods on asthma outcomes: a narrative systematic review

ObjectivesTo evaluate through a systematic review the effectiveness of electronic methods in monitoring adherence to regular inhaled corticosteroids (ICS) alone or in combination with long-acting β2-agonists (LABAs) and their effect...

The balanced maximally diverse grouping problem with integer attribute values

The paper considers the assignment of items to groups according to their attribute values such that the groups are as balanced as possible. Although the problem is in general NP-hard, we prove that it can be solved in pseudo-polynomial time if attribute values are integer. We point out a relation to partition and more general to multi-way number partitioning. Furthermore, we introduce a mixed-integer programming (MIP) formulation, a variable reduction technique, and an efficient lower bound for the objective value. Our computational results show that the lower bound meets the optimal objective value in the most of our instances of realistic size. Hence, the MIP solves instances with several thousand items within seconds to optimality.

Worst-case analysis of heuristic approaches for the temporal bin packing problem with fire-ups

We consider the temporal bin packing problem with fire-ups (TBPP-FU), a branch of operations research recently introduced in multi-objective cloud computing. In this scenario, any item is equipped with a resource demand and a lifespan meaning that it requires the bin capacity only during that time interval. We then aim at finding a schedule minimizing a weighted sum of the total number of bins required and the number of switch-on processes (so-called fire-ups) caused during operation. So far, research on the TBPP-FU has mainly focused on exact approaches and their improvement by valid cuts or variable reduction techniques. Although these studies have revealed the problem considered here to be very difficult to cope with, theoretical contributions to heuristic solution methods have not yet been presented in the available literature. Hence, in this article we investigate the worst-case behavior of some approximation algorithms, ranging from classic online algorithms to a more sophisticated look-ahead heuristic specifically designed for the TBPP-FU. In addition, we theoretically study three heuristics the ideas of which are inspired by solution methods for generalized bin packing problems in the field of logistics. As a main contribution, we constructively show that the feasible solutions obtained by all these approaches can be arbitrarily bad. By doing so, we (i) identify a new open problem in cutting and packing, and (ii) establish another previously unknown difference between the classical TBPP and the extended problem with fire-ups, rendering the latter the more difficult problem even from a heuristic point of view.

Predicting Species Occurrence of Litsea leytensis Merr. in the Provinces of Laguna and Quezon, Philippines

Litsea leytensis Merr. is an endemic and premium hardwood species in the Philippines used for wood carving. It is a threatened species that survival in the wild is impossible with persistent causal factors. The study estimated the species probability of occurrence in Laguna and Quezon based on local knowledge, published records of species distribution, and environmental variables using the Maximum Entropy (MaxEnt) model. Six pre-models were generated: one climatic model, four partial models of variable groups combined with climatic variables, and one full model with 31 original variables. The Final model had 19 highly correlated variables after variable selection and reduction using Jackknife and multi-collinearity tests. Analysis of variable importance revealed that L. leytensis’ occurrence was mostly determined by climatic (62.0%), edaphic (21.76%), and anthropogenic variables (9.53%), while topographic (5.15%) and vegetation-related variables (1.34%) had lesser contributions. The Area Under the Receiver Operating Characteristic Curve (AUC) and True Skill Statistics (TSS) measured model accuracy; and the Final model performed best at AUC = 0.9489 and TSS = 0.7175. Modeling species probability of occurrence could help key sectors in Laguna and Quezon to formulate appropriate conservation strategies for L. leytensis as a reforestation and industrial tree plantation species.

Development of a blood proteins-based model for bronchopulmonary dysplasia prediction in premature infants

BackgroundBronchopulmonary dysplasia (BPD) is the most common chronic pulmonary disease in premature infants. Blood proteins may be early predictors of the development of this disease.MethodsIn this study, protein expression profiles (blood samples during their first week of life) and clinical data of the GSE121097 was downloaded from the Gene Expression Omnibus. Weighted gene co-expression network analysis (WGCNA) and differential protein analysis were carried out for variable dimensionality reduction and feature selection. Least absolute shrinkage and selection operator (LASSO) were conducted for BPD prediction model development. The performance of the model was evaluated by the receiver operating characteristic (ROC) curve, calibration curve, and decision curve.ResultsThe results showed that black module, magenta module and turquoise module, which included 270 proteins, were significantly correlated with the occurrence of BPD. 59 proteins overlapped between differential analysis results and above three modules. These proteins were significantly enriched in 253 GO terms and 11 KEGG signaling pathways. Then, 59 proteins were reduced to 8 proteins by LASSO analysis in the training cohort. The proteins model showed good BPD predictive performance, with an AUC of 1.00 (95% CI 0.99-1.00) and 0.96 (95% CI 0.90-1.00) in training cohort and test cohort, respectively.ConclusionOur study established a reliable blood-protein based model for early prediction of BPD in premature infants. This may help elucidate pathways to target in lessening the burden or severity of BPD.