Stepwise Regression Research Articles

Correlation between corneal biomechanical and tomographic parameters in cataract patients

AIM: To investigate the relationship between preoperative corneal biomechanical properties and corneal tomographic properties in cataract patients. METHODS: The study consisted of 59 eyes of 30 participants who were diagnosed as cataract in Peking University Third Hospital between September 2019 and November 2019. Stepwise multivariable linear regression analysis was calculated to determine the relationship between corneal biomechanical parameters and tomographic parameters. The patients were classified into three groups of with the rule (WTR) astigmatism, against the rule astigmatism and oblique astigmatism. And the differences in corneal parameters among different groups were compared. RESULTS: There were significant differences in the first applanation time (A1T), the first applanation length (A1L), corneal velocity during the first applanation (Vin), the second applanation time (A2T), highest concavity (HC) radius, displacement amount (DA), DA ratio, stiffness parameter A1 (SPA1) and integrated radius (IR) between oblique astigmatism patients and the other two groups. Total corneal steep meridian (K2) was negatively associated with A1L, A1T and corneal velocity during the second applanation (Vout). Patients with higher anterior corneal curvature had lower HC radius and central corneal thickness (CCT; P=0.001 and 0.006, respectively), while the Ambrosio relational thickness to the horizontal profile (ARTh) was higher than those with lower anterior corneal curvature (P=0.009). CONCLUSION: The study reveals that the elasticity of corneal collagen fibers is greater, but the viscoelasticity of cornea is smaller in patients with oblique astigmatism. There is no significant difference in ARTh between patients with different types of astigmatism, that is, the corneal biomechanical specificity of oblique astigmatism group is probably not caused by corneal thickness. Moreover, we find patients with higher anterior corneal curvature has lower HC radius and CCT but higher ARTh than those with lower anterior corneal curvature.

Predictive role of neutrophil percentage-to-albumin ratio in acute fulminant myocarditis patients receiving extracorporeal membrane oxygenation.

Venoarterial extracorporeal membrane oxygenation induces an inflammatory response upon initiation. The neutrophil percentage-to-albumin ratio is a promising biomarker for predicting mortality in patients with systemic inflammation. This study aimed to investigate the association between neutrophil percentage-to-albumin ratio and in-hospital mortality in pediatric patients withacute fulminant myocarditis undergoing extracorporeal membrane oxygenation and to develop a PEACE model for predicting mortality in these patients. This retrospective study included pediatric patients diagnosed with acute fulminant myocarditis who underwent venoarterial extracorporeal membrane oxygenation between July 2015 and August 2022. Multivariable logistic regression analysis was used to investigate the independent association between the neutrophil percentage-to-albumin ratio and the risk of in-hospital mortality. In addition, we utilized least absolute shrinkage and selection operator regression to select predictive factors, ultimately developing a nomogram to predict outcomes in pediatric patients receiving venoarterial extracorporeal membrane oxygenation. A total of 125 patients eligible for analysis were included in this study, with an in-hospital mortality rate of 28.8%. Multivariable logistic regression revealed that the neutrophil percentage-to-albumin ratio was an independent risk factor for in-hospital mortality in venoarterial extracorporeal membrane oxygenation patients. Restricted cubic splines revealed a positive association between the two (Pnonlinearity = 0.84). Least absolute shrinkage and selection operator regression and backward stepwise logistic regression identified age, cardiopulmonary resuscitation, lactate levels, and the neutrophil percentage-to-albumin ratio as key predictive factors. Using these factors, a nomogram (PEACE model) was developed to predict in-hospital mortality in venoarterial extracorporeal membrane oxygenation patients. The area under the receiver operating characteristic curve was 0.83 [95% confidence interval (CI), 0.74-0.92], with the inclusion of the neutrophil percentage-to-albumin ratio significantly enhancing the model's predictive accuracy. The neutrophil percentage-to-albumin ratio may serve as a potential predictor for venoarterial extracorporeal membrane oxygenation in-hospital mortality in pediatric patients with acute fulminant myocarditis, suggesting that inflammatory responses are associated with patient prognosis. The PEACE model is superior in predicting the prognosis of pediatric patients supported by venoarterial extracorporeal membrane oxygenation, and can help in clinical decision making.

FeNO as a biomarker of interstitial and fibrotic pulmonary sequelae in patients admitted for severe SARS-CoV-2 pneumonia.

Pulmonary fibrosis after severe SARS-CoV-2 pneumonia is a major sequela in surviving patients which requires evaluation. Fractional exhaled nitric oxide (FeNO) is a marker of airway inflammation, easy to obtain and available in most functional testing laboratories of pulmonology services. Our objective was to evaluate the capacity of FeNO as a biomarker of interstitial and fibrotic pulmonary sequelae in patients admitted for severe SARS-CoV-2 pneumonia. We recruited 335 patients admitted for severe pneumonia secondary to SARS-CoV-2 who were being followed up at the Diffuse Interstitial Lung Disease unit at Hospital Universitario Marqués de Valdecilla. FeNO levels were higher in patients with fibrotic interstitial sequelae: mean24.3 vs. 19.8 ppbs, p = 0.002, with an area under the curve (AUC) of 0.63; 95% confidence interval (CI) 0.57-0.69 and an optimal cut-off point of 11 ppb maximizing the weighted combination of Sensitivity and specificity. FeNO ranked 6th among the 18 variables studied using various methods (forward selection, backward elimination, and stepwise regression) in evaluating the predictive ability for fibrotic interstitial sequelae, and it was the 5 th most predictive variable after using the cut-off point of 11 ppb. The joint predictive ability of the overall model with the 6 more predictive variables was higher than 0.8: AUC (Use of systemic corticosteroids + peak C-reactive Protein at admission + Age + Endotracheal intubation + Diffusing Capacity for CO (DLCO) + FeNO as quantitative continuous) = 0.81; 95%CI (0.77-0.86). AUC of the same model with FeNO as dichotomous (11 ppb cut-off point) = 0.82; 95%CI (0.78-0.87). Our study shows an increase in FeNO in patients who, after admission for severe SARS-CoV-2 pneumonia, present fibrotic interstitial sequelae at the three-month follow-up, as one of the different predictive variables related to the presence of these sequelae.

Association of skeletal muscle strength and cardiorespiratory fitness with bone mineral density: a cross-sectional study

BackgroundThe high prevalence and increasing severity of osteoporosis have elevated it to a significant global public-health concern, imposing a substantial economic burden. Skeletal muscle strength and cardiorespiratory endurance serve as pivotal metrics in evaluating physical health. They play a vital role in mitigating the risks associated with bone density decline and the development of osteoporosis. This cross-sectional study was carried out among Chinese adults aged 30–60 years. Its aim is to investigate the associations between skeletal muscle strength, cardiorespiratory endurance, and bone density, thereby providing scientific evidence for formulating prevention and intervention strategies against osteoporosis.MethodA handgrip dynamometer was employed to measure the subjects’ grip strength, which served as an indicator for assessing their upper-limb muscle strength. Additionally, an isokinetic muscle-strength tester was utilized to determine the subjects’ lower–limb isokinetic muscle strength, thereby evaluating the strength of their lower–limb muscles. The exercise cardiopulmonary testing system was utilized to directly measure the subjects’ maximum oxygen uptake (VO₂ max) via a treadmill test. This crucial indicator was then employed to assess the subjects’ cardiorespiratory fitness (CRF). Moreover, the QCT bone density analysis system was used to measure the thoracolumbar bone density of the subjects, and their bone density levels were evaluated based on the T value. A multiple stepwise linear regression model was utilized to further examine the associations between the independent variables grip strength, quadriceps muscle strength, and maximum oxygen uptake and the dependent variable, the bone density T value, stratified by gender. A series of factors potentially influencing the results were adjusted for, such as age, weight, body mass index (BMI), smoking and drinking habits, as well as vitamin D and calcium levels.ResultsIn the final fully adjusted model, a significant positive correlation was detected between grip strength and the BMD T score (β = 0.03, p < 0.001). This correlation held significance in both women (β = 0.15, p < 0.001) and men (β = 0.07, p < 0.001). A significant correlation was observed between quadriceps muscle strength and the bone mineral density T score (β = 0.94, p < 0.001). Notably, this correlation was particularly pronounced in the female group, with a more significant relationship (β = 1.35, p < 0.001), whereas in the male group, the correlation was not significant (β = 0.42, p = 0.230). In addition, a significant correlation was identified between the maximum oxygen uptake and the bone density T value in the overall sample (β = 0.28, p = 0.009). Nevertheless, upon gender stratification, the correlation between the maximum oxygen uptake and bone density was not significant in women (p = 0.884), yet it was significant in men (β = 0.42, p = 0.009).ConclusionIn the 30–60 age group, a significant positive correlation was detected between skeletal muscle strength and bone density. Specifically, in women, lower limb muscle strength was more closely associated with bone density; however, this relationship was not significant in men. Moreover, the association between cardiorespiratory endurance and bone density varied by gender. It was not significant in women but demonstrated a significant positive correlation in the male group.

CTA-based evaluation of carotid body size reveals associations with cardiovascular and metabolic conditions

Background: The carotid body (CB) detects blood oxygen changes and may play a role in metabolic diseases. Several studies have suggested a link between CB size and cardiovascular conditions. This study aimed to evaluate CB size using computed tomography angiography (CTA) and investigate its associations with cardiovascular and metabolic conditions. Methods: A retrospective analysis of 279 patients undergoing CTA of the cervical vasculature was conducted. The CB was identified as an enhancing structure at the carotid bifurcation, and its area was measured on axial images. Clinical data, including comorbidities and vascular risk factors, were collected. Statistical analyses included univariate and stepwise multiple linear regression to identify significant predictors of CB size. Results: The CB was identified in 163 patients (49.1% right, 50.9% left). The mean CB area was 3.183 mm 2 for the right side and 2.901 mm 2 for the left. Obstructive sleep apnea (OSA) and internal carotid artery (ICA) stenosis ⩾ 70% were significant predictors of increased CB size. In the final regression model, OSA was associated with a 1.049 mm 2 increase in CB area ( p = 0.027), whereas ICA stenosis ⩾ 70% and renin-angiotensin system inhibitor treatment were associated with increases of 0.528 mm 2 ( p = 0.036) and 0.494 mm 2 ( p = 0.037), respectively. CB hypertrophy was also associated with hypertension, obesity, and smoking in univariate analyses. Conclusions: This study highlights significant associations between CB hypertrophy and conditions such as OSA and ICA stenosis, suggesting that CB enlargement reflects the interplay between hypoxia, vascular pathology, and metabolic dysregulation. CTA may assess CB size as a cardiovascular biomarker.

간호대학생의 안전동기, 개인적 태도, 표준주의 안전환경이 환자안전관리 행위에 미치는 영향

Objectives The purpose of this study is to confirm the effects of safety motivation, personal attitude, and standardized safety environment on patient safety management behaviors of nursing students. Methods A questionnaire survey was conducted on 198 nursing college students attending 2 university. The data of this study were analyzed by means of SPSS WIN 20.0 program, mean and standard deviation, pearson's correlation coefficients and stepwise multiple regression analysis. Results Results of this study, safety motivation of nursing students was positively correlated with personal attitude(r=.261, p<.001), and patient safety management behavior was positively correlated with safety motivation (r=.768, p<.001) and personal attitude (r=.768, p<.001). In addition, safety motivation (β=.748, p<.001) was the predictive factor influencing safety management behavior of nursing students, and the explanatory power of safety motivation for patient safety management behavior was 58.8%. Conclusions Therefore, safety motivation is an important variable to enhance patient safety management behavior of nursing students, and it is suggested to develop and apply a nursing education program that can improve safety motivation for patient safety management.

간호대학생의 그릿, 학업적 자기효능감, 전공만족이 진로준비행동에 미치는 영향

Objectives This study was conducted to confirm the effects of nursing students' grit, academic self-efficacy, and major satisfaction on career preparation behavior. Methods This study conducted a convenience sample of 300 nursing students at a university, and the analysis was conducted on 267 people after excluding missing values ​​from the questionnaire. The data collection period was from October 1 to October 31, 2024. The collected data were analyzed using the SPSS WIN 27.0 program, and the correlation between variables was analyzed using Pearson’s Correlation, and the predictive factor for career preparation behavior was analyzed using Stepwise multiple regression. Results First, academic self-efficacy showed a positive correlation with grit (r=.474, p<.01) and major satisfaction (r=.667, p<.01). Major satisfaction showed a positive correlation with grit (r=.377, p<.01). Career preparation behavior showed a positive correlation with grit (r=.275, p<.01), major satisfaction (r=.353, p<.01), and academic self-efficacy (r=.366, p<.01). Second, the predictive factors affecting career preparation behavior were major satisfaction (β=.184, p=.016) and academic self-efficacy (β=.188, p=.019). Grit was not a statistically significant predictive factor, and major satisfaction and academic self-efficacy explained 16.5% of career preparation behavior. Conclusions Through this study, we were able to identify the relationship between grit and academic self-efficacy, major satisfaction, and career preparation behaviors of nursing students, and analyze the influence of these on career preparation behaviors. Based on this, we were able to find out that there is a need to foster grit in nursing students who require passion and perseverance, thereby positively affecting their academic self-efficacy, major satisfaction, and career preparation behaviors.

간호대학생의 간호정보핵심역량과 비판적사고성향이 간호실무준비도에 미치는 영향

Objectives This study aimed to provide basic data necessary for developing a program to improve nursing practice readiness by assessing the influence of nursing information core competency and critical thinking disposition with nursing practice readiness among nursing students. Methods The study participants were 4th year students expected to graduate nursing university students in G province in korea. Data were collected by online self-report questionnaires from September 1 to September 30, 2023. Data analysis was done by using SPSS/WIN 26.0 program for Descriptive statics, Independent t-test, one-way ANOVA, Sheffe' test, Pearson's correlation coefficient, Stepwise multiple regression analysis. Results This study showed positive correlation between nursing practice readiness and nursing information core competency(r=.49, p<.001), critical thinking disposition(β=.54, p<.001). In addition, nursing information core competency(β=.14, p<.001), critical thinking disposition(β=.54, p<.001), satisfaction of nursing major(β=.13, p<.001), clinical practice experience(β=.20, p<.001) have a 49.0% explanatory power for the nursing practice readiness in nursing students. Conclusions In order to improve the nursing practice readiness of nursing students, it is necessary to develop a nursing practice readiness improving program and improve education methods based on strengthening nursing information core comptencies and critical thinking dispositions. In addition, it is necessary to improve major satisfaction through cooperative interaction between professors and students and to expand diverse clinical practice experiences.

Screening and Comprehensive Evaluation of Drought Resistance in Cotton Germplasm Resources at the Germination Stage

Drought stress has a significant impact on cotton growth, development, and productivity. This study conducted drought stress treatment and normal water treatment (control group) on 502 cotton accessions and analyzed data on eight phenotypic traits closely related to drought stress tolerance. The results showed that all indicators changed significantly under drought stress conditions compared to the control group, with varying degrees of response among different indicators. To comprehensively evaluate the drought resistance of cotton during the germination period, the values of drought resistance comprehensive evaluation (D-value), weight drought resistance coefficient (WDC-value), and comprehensive drought resistance coefficient (CDC-value) were calculated based on membership function analysis and principal component analysis. Cluster analysis based on the D-value divided the germplasm into five drought-resistant grades, followed by the selection of one extreme material, each from the strongly drought-resistant and strongly drought-sensitive groups. An evaluation model was established using stepwise regression analysis, including the following effective indicators: Relative Fresh Weight (RFW), Relative Hypocotyl Length (RHL), Relative Seeds Water Absorption Rate (RAR), Relative Germination Rate (RGR), Relative Germination Potential (RGP), and Relative Drought Tolerance Index (RDT). The validation of the D-value prediction model based on the Best Linear Unbiased Prediction (BLUP) showed that the results obtained from two independent biological replicates were highly consistent. The comprehensive evaluation system and screening indicators established in this study provide a reliable method for identifying drought tolerance during the germination period.

The use of the Odds Ratio Product (ORP) and self-reported data to detect comorbid insomnia and sleep apnea.

To evaluate the utility of the Odds Ratio Product (ORP) in differentiating comorbid insomnia and sleep apnea (COMISA) from obstructive sleep apnea (OSA) and chronic insomnia (CIN). We retrospectively analyzed 9750 patients in four groups: 1) 1152 controls; 2) 2395 with CIN; 3) 2297 with OSA; and 4) 3906 with COMISA. CIN was defined as difficulty initiating/maintaining sleep with daytime fatigue/sleepiness occurring "often"/"always". OSA was defined as an apnea-hypopnea index >5 on polysomnography. ORP, computed every 3seconds from polysomnography, was analyzed alongside sleep metrics, comorbidities, and sleep habits. Associations were assessed using univariate multinomial logistic regression, followed by stepwise regression to identify independent predictors of COMISA versus OSA or CIN. Machine learning models classified COMISA, OSA, and CIN as distinct clinical groups. ORP-derived features showed stronger associations with COMISA than traditional sleep metrics (except N3 latency). Independent objective predictors of COMISA included male sex (OR = 1.31, 95% CI = [1.16, 1.47]), BMI (1.27, [1.25, 1.29]), N3 latency (1.21, [1.13, 1.29]), age (1.17, [1.16, 1.19]), peak ORP during spontaneous arousals (1.12, [1.01, 1.25]), and time in ORP decile 7 (1.10, [1.07, 1.13]). Subjective predictors included depression, hypertension, allergy, headache, sleep aid/alcohol use, sleepiness, and lower sleep duration. Machine learning achieved overall accuracy of 61.2% (p<.05), with sensitivity of 71% for COMISA, 65% for OSA, and 43% for CIN. ORP is a promising objective marker for COMISA, distinguishing it from OSA more effectively than sleep metrics but separating COMISA from CIN poorly.

Assessing Wellbeing in the Indian Leather Industry: Evidence from Kanpur's Factory Workers

The wellbeing of workers in the labour-intensive industries like leather manufacturing is vital for their economic stability. This study investigates the multidimensional wellbeing of workers engaged in the Kanpur's leather cluster, an important hub of India's export economy. The study uses primary data collected from randomly selected 103 workers. The survey conducted on factory workers captured the demographics and the five key dimensions of wellbeing: personal wellbeing, health and material wellbeing, workplace wellbeing, and relational wellbeing. Descriptive statistics and stepwise regression analysis were conducted on the survey data to identify the influence of socio-economic, workplace, and demographic factors on different dimensions and the overall wellbeing of the workers engaged in the leather cluster. The descriptive statistics show stark differences in the education levels and skill type amongst the respondents. Results of the stepwise regression analysis reveals gender, job security, job duration, age, and education as the critical determinants of overall wellbeing of the workers. In addition, the workers' skill type, weekly working hours, wages, and housing status also significantly influence different wellbeing dimensions. Importantly, the wage inequality and gender disparities exacerbate challenges, particularly in respect of women and unskilled workers. Thus, the findings underscore the negative impacts of informal employment and highlight the importance of supportive workplace environments. Policy interventions should, therefore, include promoting formal employment, strengthening health and safety measures, addressing wage inequality, and providing targeted supports. Incorporating wellbeing metrics into the industry evaluation mechanism can potentially ensure holistic labour assessments. Nevertheless, future research should explore these aspects in other clusters with more qualitative perspectives for deeper understanding of the underlying dynamics.

ECMO and impella increase stroke risk in acute myocardial infarction.

There is limited data on the risk of stroke in patients with acute myocardial infarction (AMI) treated with temporary mechanical circulatory support devices. Retrospective data were extracted from the U.S. National Inpatient Sample between October 2015 and December 2020, involving hospitalizations for AMI. The final cohort comprised: 4,370,069 hospitalizations with medical therapy only, 136,005 with intra-aortic balloon pump (IABP) only, 41,560 with Impella only, and 10,695 with extracorporeal membrane oxygenation (ECMO) utilized during hospitalization. The overall stroke rates for patients receiving medical therapy only, IABP only, Impella only, and ECMO were 3.39%, 3.56%, 4.54%, and 13.14%, respectively. Specifically, ischemic stroke rates were 2.93%, 3.17%, 3.96%, and 9.91%, and hemorrhagic stroke rates were 0.69%, 0.59%, 0.87%, and 4.77% for the respective groups. In stepwise forward Cox regression analysis, ECMO use was associated with the highest adjusted odds ratio (aOR) for overall stroke (aOR 3.04, 95% CI [2.66-3.48]), followed by Impella alone (aOR 1.79, 95% CI [1.61-2.00]) and atrial fibrillation (aOR 1.34, 95% CI [1.31-1.38]). However, IABP use showed no significant association with increased stroke risk in either univariate or multivariate analyses. While IABP use is not associated with an increased risk of ischemic or hemorrhagic stroke, ECMO and Impella use are linked to a higher stroke risk, particularly for ECMO-treated AMI.

Calibration of DEM Parameters for Multi-Component Chinese Cuisine

With the industrialization and standardization of Chinese cuisine, accurate discrete element simulation parameters are essential for analyzing the flow and conveying behavior of dishes. This study focused on standardized Kung Pao Chicken and employed the Hertz–Mindlin (JKR) model to develop a discrete element model suitable for cohesive, multi-component Chinese cuisine. The triaxial dimensions of diced chicken, peanuts, and scallions were measured to construct the model. Physical experiments were conducted to obtain basic parameters. The main parameters of the constitutive model were determined using a stepwise regression fitting method. For inter-material contact parameters that are difficult to measure directly, key model parameters were calibrated by fitting simulated repose angle results to experimental measurements. The calibrated parameters enabled high simulation accuracy, with repose angle errors below 0.05%, confirming the model’s reliability. This study provides a theoretical foundation for the simulation and design of automated conveying systems tailored to Chinese cuisine.

Comprehensive Analysis of Soil Physicochemical Properties and Optimization Strategies for “Yantai Fuji 3” Apple Orchards

Based on an integrated analysis, this study summarized the current status of soil quality in Yantai apple orchards, developed a multivariate regulation model for key soil physicochemical properties, and proposed optimized fertilization strategies to improve soil quality in the region. The study analyzed the physicochemical properties of the topsoil (0–30 cm) in 19 representative apple orchards across Yantai, including indicators like pH, organic matter (OM), major nutrient ions, and salinity indicators, using standardized measurements and multivariate statistical methods, including descriptive statistics analysis, frequency distribution analysis, canonical correlation analysis, stepwise regression equation analysis, and regression fit model analysis. The results demonstrated that in apple orchards across the Yantai region, reductions in pH were significantly mitigated under the combined increased OM and exchangeable calcium (Ca). Exchangeable potassium (EK) rose in response to the joint elevation of OM and available nitrogen (AN), and AN was also positively influenced by EK, while OM also exhibited a promotive effect on Olsen phosphorus (OP). Furthermore, Ca increased with higher pH. AN and EK jointly contributed to the increases in electrical conductivity (EC) and chloride ions (Cl), while elevated exchangeable sodium (Na) and soluble salts (SS) were primarily driven by EK. Accordingly, enhancing organic and calcium source fertilizers is recommended to boost OM and Ca levels, reduce acidification, and maintain EC within optimal limits. By primarily reducing potassium’s application, followed by nitrogen and phosphorus source fertilizers, the supply of macronutrients can be optimized, and the accumulation of Na, Cl, and SS can be controlled. Collectively, the combined analysis of soil quality status and the multivariate regulation model clarified the optimized fertilization strategies, thereby establishing a solid theoretical and practical foundation for recognizing the necessity of soil testing and formula fertilization, the urgency of improving soil quality, and the scientific rationale for nutrient input management in Yantai apple orchards.

Predicting medication wastage using machine learning based on patient beliefs

ObjectivesMedication wastage is a critical issue impacting the sustainability of subsidised healthcare systems in Southeast Asia due to financial and resource constraints. This study aimed to develop a machine learning (ML) model to predict medication wastage by analysing patient demographics, health conditions and beliefs about medicines, using Malaysia as a case study.MethodsA cross-sectional survey was conducted involving 734 patients across six public healthcare facilities in Malaysia. Data on demographics, medication history and beliefs about medicines were collected using validated questionnaires. Multiple ML regression models were evaluated to predict medication wastage, with performance assessed based on root mean squared error (RMSE).ResultsThe XGBoost model achieved the best performance with the lowest RMSE of 4.67, outperforming other models (RMSE range:4.68–5.10). It also performed best using only seven features selected by sequential backward elimination method using LR, making it practical for clinical implementation. Key predictors of medication wastage included beliefs about medicines, age, ethnicity, region and monthly income.ConclusionThis study is the first to apply ML to address medication wastage in a Southeast Asian context, filling a critical research gap. The proposed model provides a foundation for developing targeted interventions to reduce medication wastage and supports policymakers and healthcare providers in optimising the allocation of subsidised medications. The insights are broadly applicable to other countries with similar healthcare resource challenges.

Prediction of spontaneous preterm birth in pregnant women using machine learning.

Spontaneous preterm birth (sPTB) is a significant global health concern, contributing to adverse outcomes for both pregnant women and newborns. Early identification of women with risk of sPTB is essential for mitigating these negative effects and improving maternal and neonatal health outcomes. The aim of this study is to explore the feasibility of using machine learning to predict sPTB risk and to analyze the contribution of variables. All data were collected retrospectively. Prediction models were developed using eight different machine learning algorithms combined with six variable selection methods. The models' predictive performance was evaluated using area under the receiver operating characteristic curve (AUROC), area under the precision recall curve (AUPRC), accuracy, sensitivity, F1-score, positive predictive value, and negative predictive value. A total of 1122 pregnant women, of whom 187 had preterm birth and 935 had term birth, were enrolled. The model by combining the categorical boosting algorithm and backward elimination had the best predictive performance with the highest AUROC (0.8762) and AUPRC (0.7061), and the Brier score was 0.12 on the test set. The top 5 variables for predicting sPTB risk in this study were free triiodothyronine, albumin/globulin, thyroglobulin antibody, total thyroxine, red cell volume distribution width. The machine learning model may help identify pregnant women at high risk of sPTB, and individual risk factor analysis could provide reference for clinical decision. However, as some key variables are not part of routine laboratory tests during pregnancy worldwide, the model's generalizability and clinical applicability require further study.

Development and validation of a machine learning-based nomogram for survival prediction of patients with hilar cholangiocarcinoma after curative-intent resection

Hilar cholangiocarcinoma (hCCA), a rare cancer of the biliary system, has a poor prognosis. This study aimed to investigate the risk factors affecting the survival of patients with hCCA after curative-intent resection and establish a survival predictive model. Clinical data from 340 hCCA patients who underwent curative-intent resection at the First Affiliated Hospital of Xi’an Jiaotong University between 2010 and 2021 were collected. The patients were randomly assigned to a training set and a testing set in a 7:3 ratio. Risk factors selection was performed by five machine learning (ML) algorithms, including Least Absolute Shrinkage and Selection Operator (LASSO) Regression, Forward Stepwise Cox regression, Boruta feature selection, Random Forest and eXtreme Gradient Boosting (XGBoost). A nomogram was constructed based on identified risk factors. The independent risk factors for the postoperative survival in hCCA patients included positive margin, lymph node metastasis, low total lymph node count (TLNC) and poor tumor differentiation. In the training and testing sets, the consistency index (C-index) of ML-based nomogram was 0.731 (95% CI: 0.684–0.753) and 0.714 (95% CI: 0.661–0.775), while the 3-year AUC of the nomogram was 0.784 (95% CI: 0.724–0.844) and 0.770 (95% CI: 0.763–0.867), respectively. The calibration curves for the nomogram showed good concordance. Based on the decision curve analysis, the nomogram had a good clinical application value, outperforming both the TNM staging system and the Bismuth-Corlette classification. Furthermore, patients were stratified into three groups with varying risks of overall survival (OS): the low-risk, middle-risk and high-risk group according to the nomogram, with statistically significant differences observed among these groups (p < 0.001). The ML-based nomogram provided a personalized prognostic prediction model for hCCA patients after surgical resection.

Estimating Some Stand Parameters Using Sentinel-1 and Sentinel-2 Satellite Images in Pure Black and Scots Pine Stands: A Case Study from Türkiye

ABSTRACT This study modeled various stand parameters (stand diameter, number of trees, basal area, volume, density) in pure Black and Scots pine stands in the Kastamonu Region of Türkiye using remote sensing data. A total of 146 Black pine and 96 Scots pine sample plots were analyzed. Stand parameters were calculated based on data obtained from field measurements. Data from Sentinel-1 (backscattering, polarization ratio, texture) and Sentinel-2 (reflectance, vegetation index, texture) were used, aggregated as the mean and sum of pixels for the sample plots. Correlation analysis identified relationships between stand parameters and remote sensing data, while stepwise regression analysis developed estimation models. In Black pine stands, the best models used the sum of Sentinel-2 pixels for stand diameter ( R d 2 = 0.170) and stand density ( R d 2 = 0.397), the sum of Sentinel-1 pixels for the number of trees ( R d 2 = 0.396) and the mean of Sentinel-2 pixels for stand volume ( R d 2 = 0.095). For Scots pine stands, the best models used the mean of Sentinel-2 pixels for stand diameter ( R d 2 = 0.314) and the sum of Sentinel-2 pixels for the number of trees ( R d 2 = 0.277), basal area ( R d 2 = 0.344), stand volume ( R d 2 = 0.123), and stand density ( R d 2 = 0.432). It achieved its best results in estimating stand density, showing the potential of remote sensing for forest inventory.

Enhancing Parkinson's disease prediction using meta-heuristic optimized machine learning models.

Parkinson's disease is a progressive neurological disorder affecting movement and cognition. Early detection is crucial but challenging with traditional methods. This study applies meta-heuristic optimization to enhance machine learning prediction models. A Parkinson's dataset with demographic, lifestyle, medical, clinical, and cognitive features was analyzed using three feature selection techniques: Whale Optimization Algorithm, Artificial Bee Colony Optimization, and Backward Elimination (BE). Random Forest (RF) models were optimized using Artificial Ant Colony Optimization for hyperparameter tuning. The optimized RF model with BE achieved 93% accuracy and 97% AUC, outperforming K-Nearest Neighbors, Support Vector Machines, Logistic Regression, XGBoost, and Stacked Ensemble models. Optimization reduced tuning time from 133 to 18 minutes. A comparison with traditional approaches and negative controls validated the results, though clinical validation remains essential before deployment. Meta-heuristic optimization significantly improves Parkinson's prediction performance and efficiency.

Risk factors of urinary tract infections with sodium-glucose cotransporter-2 inhibitors in heart failure

Sodium-glucose cotransporter-2 (SGLT-2) inhibitors, while providing cardiorenal benefits in heart failure, have been associated with an elevated risk of urinary tract infections (UTIs). This study aimed to identify independent risk factors for UTIs in HF patients receiving SGLT-2 inhibitor therapy. In this multicenter retrospective cohort study, 110 heart failure patients treated with SGLT-2 inhibitors were included, among whom 41 developed UTIs. Comparative analyses between UTI and non-UTI groups were performed for demographic, clinical, and laboratory variables. Statistically significant factors (P < .05) in univariate logistic regression were subsequently entered into a multivariate model using backward stepwise elimination to adjust for potential confounders. Multivariate analysis identified 3 independent predictors of UTIs: female (odds ratio [OR] = 8.87, 95% confidence interval [CI]: 2.24–33.81; P = .002), elevated urinary ketones (OR = 10.59, 95% CI: 1.49, 75.44; P = .019), and prolonged bedridden status (OR = 46.96, 95% CI: 4.03, 547.35; P = .002). Notably, glycosuria severity did not significantly correlate with UTI risk in adjusted models. The identified risk factors – female, ketonuria, and immobility – challenge the conventional hypothesis linking SGLT-2 inhibitor-associated glycosuria to UTIs. Instead, these findings emphasize patient-specific vulnerabilities, particularly immune-metabolic dysregulation and functional decline, as primary drivers of infection risk. Clinicians should prioritize individualized monitoring strategies in high-risk subgroups to optimize therapeutic safety.