Regression Method Research Articles

Comparative Analysis of Regression Methods for Estimation of Remaining Useful Life of Lithium Ion Battery

Lithium batteries play a critical role in modern technological applications, including electric vehicles and portable electronic devices. Ensuring accurate estimation of their remaining useful life is essential to improve system efficiency and reliability. This study focuses on predicting the remaining useful life of lithium batteries using advanced regression methods. Data were collected from lithium battery charge-discharge cycles, encompassing key operational parameters such as voltage, current, and temperature. The analysis employed several regression models, including linear regression, lasso regression, and Ridge regression, to identify relationships between these parameters and battery life. The models were evaluated based on estimation accuracy, with Root Mean Square Error (RMSE) as the primary performance metric. The findings demonstrate that regression methods can effectively capture non-linear relationships between input variables and the remaining useful life, with lasso and Ridge regression showing superior performance in reducing prediction errors. These results underscore the potential of regression-based approaches in providing robust and reliable estimations of battery life. The conclusions highlight the importance of these models for developing predictive battery management systems, which can optimize battery performance and extend their operational lifespan across various applications. This research establishes a solid foundation for future studies on intelligent battery health monitoring and management.

What Is the Relationship Between Forest Footprint and Export of Forest Products? Evidence from Method of Moments Quantile Regression

This study investigates the long-run relationship between forest footprint, which shows the amount of forest area needed for pulp, industrial wood, firewood and timber, and forest products as an environmental indicator. Forest footprint, forest product exports, forest product production, forest areas, biomass consumption, and urbanization variables are used in the analyses with annual data for the period 2000–2017 for selected European Union (EU) countries. As a result of the cointegration analyses, there is a long-run relationship between the variables. According to the results of coefficient estimation, it is concluded that forest product exports and urbanization have a decreasing effect on forest ecological footprint, while forest area, forest product production, and biomass consumption have an increasing effect. According to the Method of Moments Quantile Regression (MMQR) estimation results, it is concluded that forest product exports have a decreasing effect on forest footprint in all quantiles in the analysis period. The production of forest products is determined as the variable with the highest negative impact on the forest’s ecological footprint. The effect of urbanization is calculated as positive, but it is the variable with the lowest impact together with forest area. Biomass consumption is found to significantly reduce the forest footprint. In view of the aforementioned findings, it is recommended that efforts be made to promote high-value added, sustainable, and environmentally friendly production processes in forest products exports. This is considered to be a key strategy to reducing the ecological footprint of forests.

An Updated and Unified Complete Earthquake Catalog of the Shillong Plateau and Adjoining Region

Abstract The Shillong Plateau (SP) and adjoining regions are one of the world’s most seismically active regions and to conduct a thorough seismological investigation and improved seismic hazard assessment, an updated and unified earthquake catalog is a prerequisite. In the present study, we compiled a homogeneous declustered earthquake catalog for the SP and adjoining regions by utilizing updated earthquake data from various international and national databases and pertinent literature for the period 825–2024, which consists of 6877 earthquakes. Employing the generalized orthogonal regression, orthogonal distance regression, standard least square regression, and inverted standard least square regression methods, 30 empirical relations have been derived to achieve magnitude homogeneity. All the reported magnitudes are converted into moment magnitude (Mw) using the derived regression relations. Subsequently, the declustering of the compiled homogenous earthquake catalog has been carried out using three techniques to distinguish dependent and independent earthquakes. Three approaches are used to assess the completeness of the nondeclustered and declustered earthquakes in terms of time and magnitude. The compiled updated and unified catalog in the present study contains 6877 earthquakes of Mw≥3.5 for the period 825–2024 obtained out of a total of 19,499 earthquakes reported from all sources. The completeness periods of the GK74 declustered earthquake catalog obtained by the cumulative visual inspection method for different magnitudes range from 14 to 218 yr and by Stepp’s method, it ranges from 15 to 215 yr. The spatial distribution of the magnitude of completeness (Mc), cumulative a-value, and b-value of the GK74 declustered earthquake catalog varies from 4.71 to 4.99, 5.64 to 10.17, and 0.68 to 1.63, respectively. The compiled updated and unified earthquake catalog is expected to be an essential input for regional seismicity, earthquake hazard, seismotectonic, and seismic hazard analysis of the SP and adjoining region.

Pengaruh Dewan Komisaris Independen, Kepemilikan Institusional dan Corporate Social Responsibility terhadap Kinerja Keuangan

This study aims to examine the influence of the Independent Board of Commissioners, Institutional Ownership, and Corporate Social Responsibility on Financial Performance. The research focuses on manufacturing companies in the food and beverage sub-sector listed on the Indonesian Stock Exchange (IDX) during the 2019-2023 period. The sample was selected using purposive sampling, resulting in 26 companies as the research objects. Data analysis was conducted using multiple linear regression methods. The study results show that (1) The Independent board of commissioners has no significant impact on financial performance and shows a negative relationship. (2) Institutional ownership has a positive and significant effect on financial performance. Meanwhile, (3) Corporate Social Responsibility does not have a significant effect on financial performance and tends to exhibit a negative relationship.

Extended eigenvalue calibration method for Mueller matrix polarimetry with four photoelastic modulators.

The Mueller matrix polarimeter with four photoelastic modulator (4-PEM MMP), known for its high-speed and high-precision measurement capabilities, holds great potential in material characterization and nanoscale measurements. We propose an extended eigenvalue calibration method (eECM) to the 4-PEM MMP by introducing two temporal basis vectors to project the continuously modulated light intensity into a time-independent projection matrix, which can then be acquired by a nonlinear regression method. The eECM enables complete calibration of system parameters of the 4-PEM MMP in a model-free manner without any approximation, including the parameters associated with each PEM and the alignment errors in the azimuths of all the polarization elements. Moreover, it is unnecessary to precisely adjust the orientations of the polarization elements, and it does not require precise reference sample information either. Consequently, the eECM allows the 4-PEM MMP to achieve accurate Mueller matrix measurement under not only the usual system configurations but also other unusual system configurations in a simple and unified approach. Simulation results have verified the effectiveness and advantages of the eECM.

Nomogram for prognosis prediction in metastatic pancreatic cancer patients undergoing intra-arterial infusion chemotherapy: incorporating immune-inflammation scores and coagulation indicators

BackgroundPancreatic cancer is one of the most malignant tumors with an inferior prognosis. This study aims to determine the prognostic significance of immune-inflammatory scores and coagulation indices in patients with metastatic pancreatic cancer(MPC) and develop a predictive nomogram.MethodsThis study retrospectively analyzed the clinical data of 384 patients with MPC who underwent intra-arterial infusion chemotherapy (IAIC). Patients were randomly divided into training and validation cohorts. Firstly, the optimal cutoff values for continuous variables were obtained in the training cohort. Then, survival analysis was performed to evaluate the impact of immune-inflammatory scores neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), systemic immune-inflammation index (SII), and coagulation indicators prothrombin time (PT), fibrinogen (FIB), and D-dimer on the overall survival (OS) of patients. Next, univariate analysis was utilized to identify prognostic factors, and a stepwise regression method was employed for variable selection to construct a nomogram based on the Cox proportional hazards model. Additionally, the predictive performance of the nomogram was assessed by the concordance index (C-index), the area under the ROC curve (AUC), and calibration curves. Finally, patients were stratified into risk groups based on the total score of the nomogram.ResultsThe Kaplan–Meier survival curves indicated that immune-inflammatory scores NLR, PLR, SII, and coagulation indicators PT, FIB, and D-dimer were associated with OS. Through Cox regression analysis, a nomogram was ultimately constructed incorporating NLR, PLR, PT, alkaline phosphatase (ALP), carbohydrate antigen 125 (CA125), age, and ablation. The model demonstrated good discriminative ability, with a C-index of 0.722, and the AUC values at 6- and 12-month OS predictions were 0.828 and 0.851 in the training cohort, while in the validation cohort, the corresponding AUC values were 0.754 and 0.791, respectively. The calibration curves showed a good fit, confirming the stability of the model. A cutoff value of 353.3 was identified as optimal for risk stratification, with a statistically significant difference in OS between the high- and low-risk groups.ConclusionThe nomogram based on immune-inflammatory scores, coagulation indicators, and other clinicopathological factors can effectively predict the OS of patients with MPC.

Improving the efficiency of thyroid cancer diagnostics in difficult diagnostic conditions using molecular diagnostic methods

Purpose of the study. To develop a diagnostic calculator for assessing the expression of profile microRNAs (miRNAs) that are significant for the oncogenesis of thyroid cancer, to introduce it into clinical practice and to evaluate the effectiveness of the proposed method for early diagnosis of malignant neoplasms of the thyroid gland in uncertain diagnostic situations after fine-needle aspiration biopsy (FNAB).Material and methods. The previously developed «Method for Diagnosing Malignant Neoplasms of the Thyroid Gland» (patent RU2820815 C1) was tested on 284 patients of the dispensary group with nodular formations of the thyroid gland according to the results of ultrasound and an uncertain conclusion after FNAB. The number of patients with papillary thyroid carcinoma was 83 (29,2%), follicular carcinoma 43 (15,1%) and with benign thyroid pathology (follicular adenoma) – 158 (55,6%) people. The expression of miRNA-146b and –574–3p in cytological samples of thyroid nodules after FNAB was assessed using the real-time PCR method.Results. In thyroid cancer patients, the expression of miRNA-146b was higher (p=0,007) compared to patients with benign pathology, while the expression activity of miRNA-574–3p was lower (p=0,013). To assess the index of reciprocal paired dysregulation of miRNA-146b and miRNA-574–3p expression in thyroid cells, it is recommended to determine the ratio of the expression values of the corresponding molecules. If the ratio index of miRNA-146b and miRNA-574–3p expression is higher than 4.5, then a conclusion is made about a malignant neoplasm of the thyroid gland. The diagnostic coefficient calculated by the developed formula using the logistic regression method also allows diagnosing malignant neoplasm of the thyroid gland when compared with the cutoff level. The proposed methods are highly informative.Conclusion. Evaluation of the expression of miRNA-146b and miRNA-574–3p in the cells of thyroid nodules obtained by FNAB allows improving the diagnosis of malignant pathology in mutation-negative thyroid nodules with uncertain FNAB results.

INFLUENCERS AND CONTENT MARKETING AS THE MAIN DRIVERS OF PURCHASE DECISIONS AT BAKSO O'BOSS, BENGKULU CITY

This study aims to identify influencers and content marketing as drivers of consumer purchasing decisions at Bakso O'Boss in Bengkulu City. The study uses a quantitative approach with multiple linear regression methods to analyze data obtained from consumers. The results of the analysis show that partially the influencer variable has a positive and significant influence on purchasing decisions, as evidenced by the value> ttable (2,730>1.73406)and Sig level<alpha0.05 = (0.001 < 0.05). The content marketing variable also has a positive and significant influence on purchasing decisions with a value of> ttable (4.453>1.73406) and Sig<alpha0.05 = (0.012 < 0.05).Simultaneously, influencers and content marketing have a significant effect on purchasing decisions, with a calculated F value (16.304) > F table (3.55) and significance (0.000 < 0.05). This study concludes that collaboration between influencers and effective content marketing strategies can increase product appeal and drive consumer purchasing decisions.

HDAC9-mediated deacetylation of CALML6 promotes excessive proliferation of glomerular mesangial cells in IgA nephropathy.

This study seeks to investigate the fundamental molecular processes through which histone deacetylase 9 (HDAC9) governs the proliferation of glomerular mesangial cells in the context of immunoglobulin A nephropathy (IgAN) and to identify novel targets for clinical research on IgAN. Data from high-throughput RNA sequencing for IgAN were procured from the Gene Expression Omnibus database to assess the expression profiles and clinical diagnostic significance of histone deacetylase family proteins (HDACs). Blood samples from 20 IgAN patients were employed in RT-qPCR analysis, and the spearman linear regression method was utilized to analyze the clinical correlation. The proliferation of glomerular mesangial cells (GMCs) under the influence of HDAC9 was examined using the 5-ethynyl-2'-deoxyuridine (EdU) assay. Proteins interacting with HDAC9 were predicted utilizing the STRING database. Immunoprecipitation and protein immunoblotting employing anti-acetylated lysine antibodies were conducted to determine the acetylation status of calmodulin-like protein 6 (CALML6). Analysis of the GSE141295 dataset revealed a significant upregulation of HDAC9 expression in IgAN and the results of RT-qPCR demonstrated a substantial increase in HDAC9 expression in IgAN patients. Receiver operating characteristic (ROC) analysis indicated that the area under the curve (AUC) value for HDAC9 were 0.845 and Spearman correlation analysis showed that HDAC9 expression was positively correlated with blood levels of blood urea nitrogen (BUN) and serum creatinine (Crea). The EdU cell proliferation assay indicated that HDAC9 facilitated the excessive proliferation of GMCs. The STRING database and recovery experiments identified CALML6 as a downstream effector of HDAC9 in controlling abnormal GMC multiplication. Co-immunoprecipitation assays demonstrated that HDAC9 modulates CALML6 expression through acetylation modification. HDAC9 is markedly upregulated in IgAN, and it mediates the excessive proliferation of GMCs by regulating the deacetylation of CALML6.

Detecting Influential Observations in Single-Index Fréchet Regression

Regression with random data objects is becoming increasingly common in modern data analysis. Unfortunately, this novel regression method is not immune to the trouble caused by unusual observations. A metric Cook’s distance extending the original Cook’s distances to regression between metric-valued response objects and Euclidean predictors is proposed. The performance of the metric Cook’s distance is demonstrated in regression across four different response spaces in an extensive experimental study. Two real data applications involving the analysis of distributions of COVID-19 transmission in the State of Texas and the analysis of the structural brain connectivity networks are provided to illustrate the utility of the proposed method in practice.

Multimodal data-based longitudinal prognostic model for predicting atrial fibrillation recurrence after catheter ablation in patients with patent foramen ovale and paroxysmal atrial fibrillation

BackgroundClinical studies on atrial fibrillation (AF) recurrence after catheter ablation in patients diagnosed with patent foramen ovale (PFO) and paroxysmal AF (PAF) are scarce. Here, we aimed to develop a nomogram model utilizing multimodal data for the risk stratification of AF recurrence following catheter ablation in individuals diagnosed with PFO and new-onset PAF.MethodsPatients with PFO and PAF who underwent catheter ablation at the Renmin Hospital of Wuhan University from January 2018 to June 2020 were consecutively enrolled. The identification of potential risk factors was conducted using the regression method known as least absolute shrinkage and selection operator. Subsequently, multivariate COX regression analysis was conducted to determine the independent risk factors, after which a nomogram scoring system was developed. The nomogram's performance was assessed via various statistical measures, including receiver operating characteristic curve analysis, calibration curve, and decision curve analysis (DCA).ResultsThe dataset was partitioned into the development cohort (n = 102) and the validation cohort (n = 43) using a 7:3 ratio. The constructed nomogram included four clinical variables: age, diabetes mellitus, lipoprotein (a), and right ventricular diameter. The area under the curve values of the development and validation cohorts at 1, 2, and 3 years post-catheter ablation were 0.911, 0.812, and 0.786 and 0.842, 0.761, and 0.785, respectively. Additionally, the nomogram demonstrated a significant correlation between the predicted and actual outcomes in the development and validation cohorts, indicating its excellent calibration. Lastly, the DCA findings suggested that the model had notable clinical applicability in predicting the likelihood of AF recurrence within 1, 2, and 3 years after catheter ablation.ConclusionThe incorporation of multimodal data in a nomogram visualization tool facilitates the concise representation of multimodal data, thereby enhancing the comprehension of the clinical status of patients with PFO and PAF following catheter ablation and providing accurate risk stratification at 1, 2, and 3 years post-treatment.Trial registration: This trial was registered in the Chinese Clinical Trial Registry.(ChiCTR2300072320).Graphical

Leveraging machine learning for enhanced reservoir permeability estimation in geothermal hotspots: a case study of the Williston Basin

Geothermal energy is a large, renewable, and clean source of energy from the earth in the form of heat. Exploring the deeper layers of the Williston Basin has revealed favorable reservoir temperatures, particularly in the western areas where high heat flows are prevalent. The quality of a geothermal hotspot hinges on the reservoir quality index (RQI), which is determined by the accuracy of calculating the field reservoir permeability. The primary goal of this study is to apply machine learning techniques to accurately calculate the field permeability, which is important for optimizing the RQI. To enhance accuracy, we initially applied various clustering algorithms, including the density-based spatial clustering of applications with noise (DBSCAN), K-means, K-median, and hierarchical clustering methods, to delineate hydraulic flow units (HFU) within the reservoir using porosity, permeability and water saturation core data. Subsequently, regression models including supervised ML regression methods such as neural networks, support vector machine (SVM) regression, Gaussian process regression (GPR), ensemble regression, linear regression, and decision trees were employed for each flow unit to establish correlations and calculate field permeability with each of these models validated using cross-validation. In comparison to the other clustering methods, the hierarchical clustering method showed the best performance by showing a strong correlation between the actual and predicted permeability values. Overall, the SVM and GPR regression methods were observed to show consistent results with the training and testing datasets, with the SVM regression technique yielding higher R-squared values through regression across the different clustering techniques. In addition, cross-plots were employed to successfully delineate the Red River formation into distinct regions, aiding in the definition of formation lithology and the estimation of field water saturation. Our study showcases an integrated approach to predicting reservoir permeability, considering limited core data. ML emerges as an effective tool for characterizing the Red River formation as a geothermal hotspot in North Dakota, showcasing the potential for sustainable energy exploration and utilization which reduces the reliance on extensive coring in order to enhance geothermal exploration accuracy.

The paradox of resource-richness: unraveling the effects on financial markets in natural resource abundant economies

In the contemporary global landscape, understanding the nexus between financial inclusion and natural resource abundance is crucial, especially for resource-rich nations. This study uses diagnostic tests and method of moments quantile regression to examines this interplay across Australia, Brazil, Canada, China, India, Russia, and the United States. We find that achieving financial inclusion is significantly challenging for countries that heavily rely on natural resources. Diversified income sources and equitable wealth distribution are essential to mitigate these challenges. Additionally, we identify a positive correlation between economic development and financial inclusion, highlighting the mutually reinforcing relationship between growth and inclusivity. Our research also reveals a notable link between adopting renewable energy and improving financial inclusion, suggesting that environmental responsibility and financial accessibility are intertwined. Foreign direct investment has nuanced impacts on financial inclusion, adding depth to our understanding. Overall, stable income from natural resources and diversified economic development emerge as key promoters of financial inclusion. These insights advocate for regionally specific policies and lay a solid foundation for future research and informed policymaking that address financial inclusion challenges and advance sustainable development.Graphical abstract

Determinants of foreign direct invesment in eight ASEAN countries

This research aims to analyze the influence of economic growth, labor, regulatory quality, political stability, and carbon emissions on Foreign Direct Investment (FDI) in eight member countries of the Association of Southeast Asian Nations (ASEAN). With strong economic growth, a qualified workforce, good regulations, stable politics, and low environmental degradation, ASEAN member countries have great potential for investment, which generates long-term returns. This makes the ASEAN region interesting to analyze the factors that influence FDI in the region. Using the Seemingly Unrelated Regression (SUR) method and secondary data from the World Bank, this study combines cross-sectional and time-series data to provide a comprehensive overview. The research findings indicate that variables such as labor, regulatory quality, political stability, and carbon emissions have a significant impact on foreign direct investment in ASEAN countries. These findings underscore the importance of these factors in determining the direction and volume of foreign investment. This study contributes significantly to understanding the dynamics of foreign investment in the ASEAN region and highlights the importance of efforts to improve political stability, regulatory quality, and manage carbon emissions to drive economic growth through foreign capital inflows. The implications of these findings can serve as a policy foundation for governments and stakeholders in ASEAN to enhance the attractiveness of foreign investment.

Safe Haven Assets In Stock Investments On The Indonesian Stock Exchange: An Empirical Study During The Covid-19 Pandemic

Safe haven assets are secure investment during times of economic downturn. This study aims at analysing the roles of three safe haven assets, Bitcoin, gold, and US dollars, during the COVID-19 pandemic. By conducting a time series analysis of returns on stocks, gold, Bitcoin, and US dollars between 2020 and 2022 using Vector Auto Regression (VAR) method, this study found the empirical evidence that suggests a negative correlation between gold and IDX, and between US dollars and IDX. Therefore, it can be conluded that gold and US dollars are safe haven assets in the IDX market. Meanwhile, despite being less resilient than gold and US dollars, Bitcoin’s indepence from stocks during bearish markets makes it a safe haven asset.

ANALYSIS OF HEMODIALYSIS SERVICE UTILIZATION BY JKN PARTICIPANTS IN THE SPECIAL REGION OF YOGYAKARTA: AN ANALYSIS OF BPJS KESEHATAN CLAIM DATA IN 2023

Chronic kidney disease (CKD) has become a serious global health problem with significant socio-economic implications. This study aims to analyze the determinants of the utilization of hemodialysis services for National Health Insurance (JKN) participants in the Special Region of Yogyakarta (DIY) in 2023. The study employs a cross-sectional design to analyze secondary data from the BPJS Kesehatan claim database in 2020, focusing on JKN participants receiving dialysis services at Advanced Referral Health Facilities (FKRTL). Data extraction is followed by analysis using STATA software, which includes univariate and bivariate analyses, multicollinearity and overdispersion tests, model comparison via Akaike Information Criterion (AIC), and multivariate analysis through Poisson regression and Binomial Negative methods. A total of 3,220 participants (0.089%) were enrolled in the study. The highest distribution of participants was in Sleman Regency (36.49%), followed by Bantul (29.53%), and Yogya City (25.96%). From the results of the binomial regression analysis, it was found that the age factor had a significant negative influence (p=0.000) with a coefficient of -0.0046 and an IRR of 0.9954, which means that every increase in age by one year will decrease the frequency of visits by 0.46%. The prevalence of diabetes (3.1%) and hypertension (10.2%) which exceed the national average indicates a high risk factor for CKD in this region is high. The age factor and location of health facilities in SLEMAN Regency significantly affected the frequency, while the segment group, gender, and nursing class rights did not show significant influence.

Reducing bias in source-free unsupervised domain adaptation for regression.

Due to data privacy and storage concerns, Source-Free Unsupervised Domain Adaptation (SFUDA) focuses on improving an unlabelled target domain by leveraging a pre-trained source model without access to source data. While existing studies attempt to train target models by mitigating biases induced by noisy pseudo labels, they often lack theoretical guarantees for fully reducing biases and have predominantly addressed classification tasks rather than regression ones. To address these gaps, our analysis delves into the generalisation error bound of the target model, aiming to understand the intrinsic limitations of pseudo-label-based SFUDA methods. Theoretical results reveal that biases influencing generalisation error extend beyond the commonly highlighted label inconsistency bias, which denotes the mismatch between pseudo labels and ground truths, and the feature-label mapping bias, which represents the difference between the proxy target regressor and the real target regressor. Equally significant is the feature misalignment bias, indicating the misalignment between the estimated and real target feature distributions. This factor is frequently neglected or not explicitly addressed in current studies. Additionally, the label inconsistency bias can be unbounded in regression due to the continuous label space, further complicating SFUDA for regression tasks. Guided by these theoretical insights, we propose a Bias-Reduced Regression (BRR) method for SFUDA in regression. This method incorporates Feature Distribution Alignment (FDA) to reduce the feature misalignment bias, Hybrid Reliability Evaluation (HRE) to reduce the feature-label mapping bias and pseudo label updating to mitigate the label inconsistency bias. Experiments demonstrate the superior performance of the proposed BRR, and the effectiveness of FDA and HRE in reducing biases for regression tasks in SFUDA.

Rheumatoid arthritis and the risk of fracture: A Mendelian randomization study.

Observational epidemiological studies indicate a higher fracture incidence in rheumatoid arthritis (RA) patients compared to the general population. However, the causal relationship between RA and fracture risk, particularly traumatic and osteoporotic fractures, is not well established. We performed Mendelian randomization (MR) analysis to evaluate the causal relationship between RA and fracture risk. We performed a MR analysis using summary statistics from genome-wide association studies to investigate the causal association between RA and the risk of traumatic fractures at 9 sites and 3 types of osteoporotic fractures. The primary analysis used inverse-variance weighting, supplemented by MR-Egger regression and other methods to assess causal relationships and sensitivity analyses, including heterogeneity and pleiotropy assessments, using R software with appropriate packages. The inverse-variance weighting results demonstrated a causal relationship between genetically predicted RA and an elevated risk of fractures, particularly traumatic fractures of the long bones and osteoporotic fractures, including fractures of shoulder and upper arm (odds ratio [OR] = 1.041, 95% confidence interval [CI]: 1.020-1.062, P = 9.06e-05), fractures of forearm (OR = 1.026, 95% CI: 1.007-1.044, P = .006), fracture of femur (OR = 1.036, 95% CI: 1.009-1.064, P = .009), fractures of lower leg, including joint (OR = 1.031, 95% CI: 1.016-1.047, P = 6.38e-05), fractures of rib(s), sternum, and thoracic vertebrae (OR = 1.041, 95% CI: 1.018-1.064, P = 4.08e-04), osteoporotic with pathological features (OR = 1.128, 95% CI: 1.071-1.188, P = 5.54e-06), postmenopausal osteoporotic with pathological features (OR = 1.060, 95% CI: 1.002-1.123, P = .044), and drug-induced osteoporotic with pathological features (OR = 1.255, 95% CI: 1.124-1.400, P = 5.02e-05). This study highlights the genetic causal link between RA and an increased risk of traumatic and osteoporotic fractures, presenting a new direction for future exploration of the mechanisms underlying RA-related fractures.

A hybrid Harris Hawks Optimization with Support Vector Regression for air quality forecasting

This paper proposes a hybridized model for air quality forecasting that combines the Support Vector Regression (SVR) method with Harris Hawks Optimization (HHO) called (HHO-SVR). The proposed HHO-SVR model utilizes five datasets from the environmental protection agency’s Downscaler Model (DS) to predict Particulate Matter () levels. In order to assess the efficacy of the suggested HHO-SVR forecasting model, we employ metrics such as Mean Absolute Percentage Error (MAPE), Average, Standard Deviation (SD), Best Fit, Worst Fit, and CPU time. Additionally, we contrast our methodology with recently created models that have been published in the literature, such as the Grey Wolf Optimizer (GWO), Salp Swarm Algorithm (SSA), Henry Gas Solubility Optimization (HGSO), Barnacles Mating Optimizer (BMO), Whale Optimization Algorithm (WOA), and Manta Ray Foraging Optimization (MRFO). In particular, the proposed HHO-SVR model outperforms other approaches, establishing it as the optimal model based on its superior results.

Explainable machine learning framework for cataracts recognition using visual features

Cataract is the leading ocular disease of blindness and visual impairment globally. Deep neural networks (DNNs) have achieved promising cataracts recognition performance based on anterior segment optical coherence tomography (AS-OCT) images; however, they have poor explanations, limiting their clinical applications. In contrast, visual features extracted from original AS-OCT images and their transform forms (e.g., AS-OCT-based histograms) have good explanations but have not been fully exploited. Motivated by these observations, an explainable machine learning framework to recognize cataracts severity levels automatically using AS-OCT images was proposed, consisting of three stages: visual feature extraction, feature importance explanation and selection, and recognition. First, the intensity histogram and intensity-based statistical methods are applied to extract visual features from original AS-OCT images and AS-OCT-based histograms. Subsequently, the SHapley Additive exPlanations and Pearson correlation coefficient methods are applied to analyze the feature importance and select significant visual features. Finally, an ensemble multi-class ridge regression method is applied to recognize the cataracts severity levels based on the selected visual features. Experiments on a clinical AS-OCT-NC dataset demonstrate that the proposed framework not only achieves competitive performance through comparisons with DNNs, but also has a good explanation ability, meeting the requirements of clinical diagnostic practice.