Stepwise Regression Algorithm Research Articles

Interpretation techniques to explain the output of a spatial land subsidence hazard model in an area with a diverted tributary

Due to the nature of black-box machine learning (ML) models used in the spatial modelling field of environmental and natural hazards, the interpretation of predictive model outputs is necessary. For this purpose, we applied four interpretation techniques consisting of interaction plot, permutation feature importance (PFI) measure, shapley additive explanation (SHAP) decision plot, and accumulated local effects (ALE) plot to explain and interpret the output of an ML model applied to map land subsidence (LS) in the Nazdasht plain, Hormozgan province, southern Iran. We applied a stepwise regression (SR) algorithm and five ML models (Cforest (as a conditional random forest), generalized linear model (GLM), multivariate linear regression (MLR), partial least squares (PLS) and extreme gradient boosting (XGBoost)) to select important features and to map the LS hazard, respectively. Thereafter, several interpretation techniques were used to explain the spatial ML hazard model output. Our findings revealed that a GLM model was the most accurate approach to map LS in our study area, and that 24.3% of the total study area had a very high susceptibility to the LS hazard. According to the interpretation techniques, land use, elevation, groundwater level and vegetation were the most important variables controlling the LS hazard and also the most important variables contributing to the model’s output. Overall, human activities, especially the diversion of the route of one of the main tributaries feeding the plain and the recharging of groundwater five decades ago, intensified the current LS occurrence. Therefore, management activities such as water spreading projects upstream of the plain can be useful to mitigate LS occurrence in the plain.

Does vegetation greening have a positive effect on global vegetation carbon and water use efficiency?

Terrestrial ecosystems have undergone significant changes as a result of climate change, profoundly affecting global carbon and water cycling processes. Notably, the synergistic changes in vegetation carbon use efficiency (CUE) and water use efficiency (WUE) and their response to patterns of climate change over the last 40 years are unknown. Therefore, in this study, global vegetation WUE and CUE were inverted using Gross primary productivity (GPP), Net primary productivity (NPP) and total evaporation (ET) data from 1981 to 2019 to reveal their temporal and spatial patterns of change through trend analysis and stability analysis. A stepwise regression algorithm was used to reveal the potential driving law of environmental factors on vegetation WUE and CUE. The results shows that (1) From 1981 to 2019, the global vegetation WUE and CUE showed in a relatively stable state, and the trends of WUE and CUE were -0.00004/year and 0.006 g C m−2 mm−1/year, respectively; (2) the greening of vegetation was the most important cause of the changes in WUE and CUE, and the driving force of rain and heat conditions on the CUE of vegetation was smaller than that of solar radiation and soil water, the regions where CO2 is the dominant factor affecting CUE and WUE are mainly in the north temperate zone; (3) the region of synergistic growth of WUE and CUE accounts for about 31.38 % of the global terrestrial area, and this pattern of change suggests that the global vegetation carbon sink potential is huge, and the popularization of vegetation planting patterns under the synergistic growth of CUE and WUE should be strengthened. The research has shown that vegetation greening is a key factor influencing changes in the WUE and CUE of vegetation, therefore, the implementation of ecological engineering will be an important step in combating climate change.

Molecular characterization, clinical value, and cancer–immune interactions of genes related to disulfidptosis and ferroptosis in colorectal cancer

BackgroundThis research strived to construct a new signature utilizing disulfidptosis-related ferroptosis (SRF) genes to anticipate response to immunotherapy, prognosis, and drug sensitivity in individuals with colorectal cancer (CRC).MethodsThe data for RNA sequencing as well as corresponding clinical information of individuals with CRC, were extracted from The Cancer Genome Atlas (TCGA) dataset. SRF were constructed with the help of the random forest (RF), least absolute shrinkage and selection operator (LASSO), and stepwise regression algorithms. To validate the SRF model, we applied it to an external cohort, GSE38832. Prognosis, immunotherapy response, drug sensitivity, molecular functions of genes, and somatic mutations of genes were compared across the high- and low-risk groups (categories). Following this, all statistical analyses were conducted with the aid of the R (version 4.23) software and various packages of the Cytoscape (version 3.8.0) tool.ResultsSRF was developed based on five genes (ATG7, USP7, MMD, PLIN4, and THDC2). Both univariate and multivariate Cox regression analyses established SRF as an independent, prognosis-related risk factor. Individuals from the high-risk category had a more unfavorable prognosis, elevated tumor mutational burden (TMB), and significant immunosuppressive status. Hence, they might have better outcomes post-immunotherapy and might benefit from the administration of pazopanib, lapatinib, and sunitinib.ConclusionIn conclusion, SRF can act as a new biomarker for prognosis assessment. Moreover, it is also a good predictor of drug sensitivity and immunotherapy response in CRC but should undergo optimization before implementation in clinical settings.

Calibration Methods for Low-Cost Particulate Matter Sensors Considering Seasonal Variability.

Many countries use low-cost sensors for high-resolution monitoring of particulate matter (PM2.5 and PM10) to manage public health. To enhance the accuracy of low-cost sensors, studies have been conducted to calibrate them considering environmental variables. Previous studies have considered various variables to calibrate seasonal variations in the PM concentration but have limitations in properly accounting for seasonal variability. This study considered the meridian altitude to account for seasonal variations in the PM concentration. In the PM10 calibration, we considered the calibrated PM2.5 as a subset of PM10. To validate the proposed methodology, we used the feedforward neural network, support vector machine, generalized additive model, and stepwise linear regression algorithms to analyze the results for different combinations of input variables. The inclusion of the meridian altitude enhanced the accuracy and explanatory power of the calibration model. For PM2.5, the combination of relative humidity, temperature, and meridian altitude yielded the best performance, with an average R2 of 0.93 and root mean square error of 5.6 µg/m3. For PM10, the average mean absolute percentage error decreased from 27.41% to 18.55% when considering the meridian altitude and further decreased to 15.35% when calibrated PM2.5 was added.

Predicting Fine Dead Fuel Load of Forest Floors Based on Image Euler Numbers

The fine dead fuel load on forest floors is the most critical classification feature in fuel description systems, affecting several parameters in the manifestation of wildland fires. An accurate determination of this fine dead fuel load contributes substantially to effective wildland fire prevention, monitoring, and suppression. This study investigated the viability of incorporating image Euler numbers into predictive models of fine dead fuel load via the taking photos method. Pinus massoniana needles and Quercus fabri broad leaves—typical fuel types in karst areas—served as the research subjects. Accurate field data were collected in the Tianhe Mountain forests, China, while artificial fine dead fuelbeds of differing loads were constructed in the laboratory. Images of the artificial fuelbeds were captured and uniformly digitized according to various conversion thresholds. Thereafter, the Euler numbers were extracted, their relationship with fuel load was analyzed, and this relationship was applied to generate three load-prediction models based on stepwise regression, nonlinear fitting, and random forest algorithms. The Euler number had a significant relationship with both P. massoniana and Q. fabri fuel loads. At low conversion thresholds, the Euler number was negatively correlated with fuel load, whereas a positive correlation was recorded when this threshold exceeded a certain value. The random forest model showed the best prediction performance, with mean relative errors of 9.35% and 14.54% for P. massoniana and Q. fabri, respectively. The nonlinear fitting model displayed the next best performance, while the stepwise regression model exhibited the largest error, which was significantly different from that of the random forest model. This study is the first to propose the use of image features to predict the fine fuel load on a surface. The results are more objective, accurate, and time-saving than current fuel load estimates, benefiting fuel load research and the scientific management of wildland fires.

Age estimation model for individual tree in natural Larix gmelinii forest based on random forest model

To accurately estimate the age of individual tree and to achieve full-cycle sustainable management of natural Larix gmelinii forest in Great Xing'an Mountains of northeastern China, we constructed individual tree age prediction model using stepwise regression and random forest algorithms based on 44 fixed plots data and 280 stan-dard tree cores obtained from the Pangu Forest Farm. We analyzed the influence of stand structure, site conditions, and competition index on the accuracy of model prediction. The model was evaluated by the coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE). The results showed that the random forest model had the highest prediction accuracy when number of decision trees was 1500 and number of node con-tention variables was 8. The random forest model had better accuracy and prediction ability than the stepwise regression model, with R2, RMSE and MAE of 0.5882, 9.9259 a, 8.1155 a. Diameter at breast height was the most important factor affecting age prediction (83.8%), followed by tree height (34.4%), elevation (17.9%), and basal area per hectare (17.5%). The random forest algorithm exhibited better adaptability and modeling effect on constructing a predictive model for individual tree age. This research contributed to improving the accuracy of growth and harvest estimation for L. gmelinii, and could provide a reference for other scientific studies related to tree age estimation in forests.

A Machine Learning–Based Readability Model for Gujarati Texts

This study aims to develop a machine learning–based model to predict the readability of Gujarati texts. The dataset was 50 prose passages from Gujarati literature. Fourteen lexical and syntactic readability text features were extracted from the dataset using a machine learning algorithm of the unigram parts of speech tagger and three Python programming scripts. Two samples of native Gujarati speaking secondary and higher education students rated the Gujarati texts for readability judgment on a 10-point scale of “easy” to “difficult” with the interrater agreement. After dimensionality reduction, seven text features as the independent variables and the mean readability rating as the dependent variable were used to train the readability model. As the students' level of education and gender were related to their readability rating, four readability models for school students, university students, male students, and female students were trained with a backward stepwise multiple linear regression algorithm of supervised machine learning. The trained model is comparable across the raters’ groups. The best model is the university students’ readability rating model. The model is cross-validated. It explains 91% and 88% of the variance in readability ratings at training and cross-validation, respectively, and its effect size and power are large and high.

Single cell sequencing revealed the mechanism of CRYAB in glioma and its diagnostic and prognostic value.

We explored the characteristics of single-cell differentiation data in glioblastoma and established prognostic markers based on CRYAB to predict the prognosis of glioblastoma patients. Aberrant expression of CRYAB is associated with invasive behavior in various tumors, including glioblastoma. However, the specific role and mechanisms of CRYAB in glioblastoma are still unclear. We assessed RNA-seq and microarray data from TCGA and GEO databases, combined with scRNA-seq data on glioma patients from GEO. Utilizing the Seurat R package, we identified distinct survival-related gene clusters in the scRNA-seq data. Prognostic pivotal genes were discovered through single-factor Cox analysis, and a prognostic model was established using LASSO and stepwise regression algorithms. Moreover, we investigated the predictive potential of these genes in the immune microenvironment and their applicability in immunotherapy. Finally, in vitro experiments confirmed the functional significance of the high-risk gene CRYAB. By analyzing the ScRNA-seq data, we identified 28 cell clusters representing seven cell types. After dimensionality reduction and clustering analysis, we obtained four subpopulations within the oligodendrocyte lineage based on their differentiation trajectory. Using CRYAB as a marker gene for the terminal-stage subpopulation, we found that its expression was associated with poor prognosis. In vitro experiments demonstrated that knocking out CRYAB in U87 and LN229 cells reduced cell viability, proliferation, and invasiveness. The risk model based on CRYAB holds promise in accurately predicting glioblastoma. A comprehensive study of the specific mechanisms of CRYAB in glioblastoma would contribute to understanding its response to immunotherapy. Targeting the CRYAB gene may be beneficial for glioblastoma patients.

A study of the current status of the quality of teacher-child interactions in preschools based on mathematical statistics multiple regression modeling

Abstract In group teaching activities, teacher-child interaction occupies a dominant position, and its quality has a significant impact on children’s interpersonal communication, language expression, and problem-handling. This study first designed a relevant questionnaire to assess the current quality of teacher-child interactions in preschools and then developed a stepwise regression algorithm based on decision trees to investigate the factors that influence the quality of these interactions. Firstly, we remove a certain number of irrelevant variables from all the input variables to reduce interference. Next, we establish stepwise regression equations between the input variables and output variables selected by decision tree pruning. Finally, we eliminate the variables with covariance. The study found that the average score of the quality of teacher-child interactions in preschools is 4.2381, which is relatively low, and the highest score in the dimension of “classroom organization” is only 4.8908. The highest score in the dimension of “classroom organization” was only 4.8908. The average scores of “emotional support”, “classroom management,” and “educational support” in the dimensions of small class and life activities were 4.76746, 4.10562, 2.8455, and 5.562, respectively, 2.8455) and (5.01564, 4.99567, 3.75402) respectively. Teacher-child interaction quality = 4.012 + 0.145 × emotional expression, i.e., emotional expression has a significant positive effect on the quality of teacher-student interaction, and the independent variable, emotional expression, can significantly and positively predict the dependent variable teacher-child interaction quality. This paper provides a useful exploration for improving the professional abilities level of kindergarten teachers and the quality of their interactions with children.

The Impact of the Big Thinking Program on Students’ Mental Health in the Context of Three-Way Parenting

Abstract This paper combines the stepwise regression algorithm as well as the decision tree algorithm to design a decision tree stepwise regression algorithm model of the impact of the big Civics program on students’ mental health. This paper focuses on the use of a stepwise regression algorithm to eliminate the independent variables that do not have a significant impact on the dependent variable, and through the decision tree, ID3 method to test all the features, in-depth exploration of the degree of influence of the large courses of ideology and politics on students’ mental health. In order to be able to more deeply solve the degree of influence of the large course of thought and politics on students’ mental health, a oneway linear regression analysis is carried out on the basis of the relevant analysis data of students’ mental health. The results show that the histogram of the independent variable of the large course of ideology and the dependent variable of the total mean score is distributed in a normal curve, with a mean of 2.82E-16, a standard deviation of 0.999, and a correlation coefficient squared of 0.057, which indicates that the large course of ideology has a predictive effect on the mental health of students and that the students’ anorexia mental health is influenced by the large course of ideology by 0.057. This study shows that the large course of ideology is able to promote the physical and mental health development of students and contribute to the mental health of the students. This study shows that the large ideology and politics program can promote the healthy development of student’s physical and mental health, provide a direction for mental health education to become an effective carrier of “moral education”, and also provide a breakthrough point for the reform and innovation of students’ mental health education.

Single-cell sequencing analysis related to sphingolipid metabolism guides immunotherapy and prognosis of skin cutaneous melanoma.

We explore sphingolipid-related genes (SRGs) in skin melanoma (SKCM) to develop a prognostic indicator for patient outcomes. Dysregulated lipid metabolism is linked to aggressive behavior in various cancers, including SKCM. However, the exact role and mechanism of sphingolipid metabolism in melanoma remain partially understood. We integrated scRNA-seq data from melanoma patients sourced from the GEO database. Through the utilization of the Seurat R package, we successfully identified distinct gene clusters associated with patient survival in the scRNA-seq data. Key prognostic genes were identified through single-factor Cox analysis and used to develop a prognostic model using LASSO and stepwise regression algorithms. Additionally, we evaluated the predictive potential of these genes within the immune microenvironment and their relevance to immunotherapy. Finally, we validated the functional significance of the high-risk gene IRX3 through in vitro experiments. Analysis of scRNA-seq data identified distinct expression patterns of 4 specific genes (SRGs) in diverse cell subpopulations. Re-clustering cells based on increased SRG expression revealed 7 subgroups with significant prognostic implications. Using marker genes, lasso, and Cox regression, we selected 11 genes to construct a risk signature. This signature demonstrated a strong correlation with immune cell infiltration and stromal scores, highlighting its relevance in the tumor microenvironment. Functional studies involving IRX3 knockdown in A375 and WM-115 cells showed significant reductions in cell viability, proliferation, and invasiveness. SRG-based risk signature holds promise for precise melanoma prognosis. An in-depth exploration of SRG characteristics offers insights into immunotherapy response. Therapeutic targeting of the IRX3 gene may benefit melanoma patients.

Culturing Limbal Epithelial Cells of Long-term Stored Corneal Donors (Organ Culture) In Vitro - A Stepwise Linear Regression Algorithm.

To assess various potential factors on human limbal epithelial cell (LEC) outgrowth in vitro using corneal donor tissue following long-term storage (organ culture) and a stepwise linear regression algorithm. Of 215 donors, 304 corneoscleral rings were used for our experiments. For digestion of the limbal tissue and isolation of the limbal epithelial cells, the tissue pieces were incubated with 4.0 mg/mL collagenase A at 37 °C with 95% relative humidity and a 5% CO2 atmosphere overnight. Thereafter, limbal epithelial cells were separated from limbal keratocytes using a 20-µm CellTricks filter. The separated human LECs were cultured in keratinocyte serum-free medium medium, 1% penicillin/streptomycin (P/S), 0.02% epidermal growth factor (EGF), and 0.3% bovine pituitary extract (BPE). The potential effect of donor age (covariate), postmortem time (covariate), medium time (covariate), size of the used corneoscleral ring (360°, 270°180°, 120°, 90°, less than 90°) (covariate), endothelial cell density (ECD) (covariate), gender (factor), number of culture medium changes during organ culture (factor), and origin of the donor (donating institution and storing institution, factor) on the limbal epithelial cell outgrowth was analyzed with a stepwise linear regression algorithm. The rate of successful human LEC outgrowth was 37.5%. From the stepwise linear regression algorithm, we found out that the relevant influencing parameters on the LEC growth were intercept (p < 0.001), donor age (p = 0.002), number of culture medium changes during organ culture (p < 0.001), total medium time (p = 0.181), and size of the used corneoscleral ring (p = 0.007), as well as medium time × size of the corneoscleral ring (p = 0.007). The success of LEC outgrowth increases with lower donor age, lower number of organ culture medium changes during storage, shorter medium time in organ culture, and smaller corneoscleral ring size. Our stepwise linear regression algorithm may help us in optimizing LEC cultures in vitro.

Identification of memory B-cell-associated miRNA signature to establish a prognostic model in gastric adenocarcinoma

BackgroundMemory B cells and microRNAs (miRNAs) play important roles in the progression of gastric adenocarcinoma (GAC), also known as stomach adenocarcinoma (STAD). However, few studies have investigated the use of memory B-cell-associated miRNAs in predicting the prognosis of STAD.MethodsWe identified the marker genes of memory B cells by single-cell RNA sequencing (scRNA-seq) and identified the miRNAs associated with memory B cells by constructing an mRNA‒miRNA coexpression network. Then, univariate Cox, random survival forest (RSF), and stepwise multiple Cox regression (StepCox) algorithms were used to identify memory B-cell-associated miRNAs that were significantly related to overall survival (OS). A prognostic risk model was constructed and validated using these miRNAs, and patients were divided into a low-risk group and a high-risk group. In addition, the differences in clinicopathological features, tumour microenvironment, immune blocking therapy, and sensitivity to anticancer drugs in the two groups were analysed.ResultsFour memory B-cell-associated miRNAs (hsa-mir-145, hsa-mir-125b-2, hsa-mir-100, hsa-mir-221) with significant correlations to OS were identified and used to construct a prognostic model. Time-dependent receiver operating characteristic (ROC) curve analysis confirmed the feasibility of the model. Kaplan‒Meier (K‒M) survival curve analysis showed that the prognosis was poor in the high-risk group. Comprehensive analysis showed that patients in the high-risk group had higher immune scores, matrix scores, and immune cell infiltration and a poor immune response. In terms of drug screening, we predicted eight drugs with higher sensitivity in the high-risk group, of which CGP-60474 was associated with the greatest sensitivity.ConclusionsIn summary, we identified memory B-cell-associated miRNA prognostic features and constructed a novel risk model for STAD based on scRNA-seq data and bulk RNA-seq data. Among patients in the high-risk group, STAD showed the highest sensitivity to CGP-60474. This study provides prognostic insights into individualized and precise treatment for STAD patients.

Using Wearable Accelerometers to Develop a Vertical Ground Reaction Force Prediction Model during Running: A Sensitivity Study

The estimation of vertical ground reaction forces (VGRFs) during running is necessary to understand running mechanisms. For this purpose, the use of force platforms is fundamental. However, to extend the study of VGRFs to real conditions, wearable accelerometers are a promising alternative to force platforms, whose use is often limited to the laboratory environment. The objective of this study was to develop a VGRF model using wearable accelerometers and a stepwise regression algorithm. Several models were developed and validated using the VGRFs and acceleration signals collected during 100 stances performed by one participant. The validated models were tested on eight participants. In a sensitivity study, the strongest correlations were observed at cut-off frequencies of ≤25 Hz and in models developed with 30 to 90 stances. After the validation phase, the 10 best models had, on average, low relative differences (≤10%) in the estimation of discrete VGRF parameters, i.e., the passive peak (εpp=6.26%), active peak (εap=2.22%), and loading rate (εlr=2.17%). The results indicate that the development of personalized models is more suitable for achieving the best estimates. The proposed methodology opens many perspectives for monitoring VGRFs under real conditions using a limited number of wearable sensors.

Development and validation of a risk prediction model for cage subsidence after instrumented posterior lumbar fusion based on machine learning: a retrospective observational cohort study.

Interbody cage subsidence is a common complication after instrumented posterior lumbar fusion surgery, several previous studies have shown that cage subsidence is related to multiple factors. But the current research has not combined these factors to predict the subsidence, there is a lack of an individualized and comprehensive evaluation of the risk of cage subsidence following the surgery. So we attempt to identify potential risk factors and develop a risk prediction model that can predict the possibility of subsidence by providing a Cage Subsidence Score (CSS) after surgery, and evaluate whether machine learning-related techniques can effectively predict the subsidence. This study reviewed 59 patients who underwent posterior lumbar fusion in our hospital from 2014 to 2019. They were divided into a subsidence group and a non-subsidence group according to whether the interbody fusion cage subsidence occurred during follow-up. Data were collected on the patient, including age, sex, cage segment, number of fusion segments, preoperative space height, postoperative space height, preoperative L4 lordosis Angle, postoperative L4 lordosis Angle, preoperative L5 lordosis Angle, postoperative PT, postoperative SS, postoperative PI. The conventional statistical analysis method was used to find potential risk factors that can lead to subsidence, then the results were incorporated into stepwise regression and machine learning algorithms, respectively, to build a model that could predict the subsidence. Finally the diagnostic efficiency of prediction is verified. Univariate analysis showed significant differences in pre-/postoperative intervertebral disc height, postoperative L4 segment lordosis, postoperative PT, and postoperative SS between the subsidence group and the non-subsidence group (p < 0.05). The CSS was trained by stepwise regression: 2 points for postoperative disc height > 14.68 mm, 3 points for postoperative L4 segment lordosis angle >16.91°, and 4 points for postoperative PT > 22.69°. If the total score is larger than 0.5, it is the high-risk subsidence group, while less than 0.5 is low-risk. The score obtains the area under the curve (AUC) of 0.857 and 0.806 in the development and validation set, respectively. The AUC of the GBM model based on the machine learning algorithm to predict the risk in the training set is 0.971 and the validation set is 0.889. The AUC of the avNNet model reached 0.931 in the training set and 0.868 in the validation set, respectively. The machine learning algorithm has advantages in some indicators, and we have preliminarily established a CSS that can predict the risk of postoperative subsidence after lumbar fusion and confirmed the important application prospect of machine learning in solving practical clinical problems.

R software for QSAR analysis in phytopharmacological studies.

In recent decades, quantitative structure-activity relationship (QSAR) analysis has become an important method for drug design and natural product research. With the availability of bioinformatic and cheminformatic tools, a vast number of descriptors have been generated, making it challenging to select potential independent variables that are accurately related to the dependent response variable. The objective of this study is to demonstrate various descriptor selection procedures, such as the Boruta approach, all subsets regression, the ANOVA approach, the AIC method, stepwise regression, and genetic algorithm, that can be used in QSAR studies. Additionally, we performed regression diagnostics using R software to test parameters such as normality, linearity, residual histograms, PP plots, multicollinearity, and homoscedasticity. The workflow designed in this study highlights the different descriptor selection procedures and regression diagnostics that can be used in QSAR studies. The results showed that the Boruta approach and genetic algorithm performed better than other methods in selecting potential independent variables. The regression diagnostics parameters tested using R software, such as normality, linearity, residual histograms, PP plots, multicollinearity, and homoscedasticity, helped in identifying and diagnosing model errors, ensuring the reliability of the QSAR model. QSAR analysis is vital in drug design and natural product research. To develop a reliable QSAR model, it is essential to choose suitable descriptors and perform regression diagnostics. This study offers an accessible, customizable approach for researchers to select appropriate descriptors and diagnose errors in QSAR studies.

Construction of the XGBoost model for early lung cancer prediction based on metabolic indices

BackgroundLung cancer is a malignant tumour, and early diagnosis has been shown to improve the survival rate of lung cancer patients. In this study, we assessed the use of plasma metabolites as biomarkers for lung cancer diagnosis. In this work, we used a novel interdisciplinary mechanism, applied for the first time to lung cancer, to detect biomarkers for early lung cancer diagnosis by combining metabolomics and machine learning approaches.ResultsIn total, 478 lung cancer patients and 370 subjects with benign lung nodules were enrolled from a hospital in Dalian, Liaoning Province. We selected 47 serum amino acid and carnitine indicators from targeted metabolomics studies using LC‒MS/MS and age and sex demographic indicators of the subjects. After screening by a stepwise regression algorithm, 16 metrics were included. The XGBoost model in the machine learning algorithm showed superior predictive power (AUC = 0.81, accuracy = 75.29%, sensitivity = 74%), with the metabolic biomarkers ornithine and palmitoylcarnitine being potential biomarkers to screen for lung cancer. The machine learning model XGBoost is proposed as an tool for early lung cancer prediction. This study provides strong support for the feasibility of blood-based screening for metabolites and provide a safer, faster and more accurate tool for early diagnosis of lung cancer.ConclusionsThis study proposes an interdisciplinary approach combining metabolomics with a machine learning model (XGBoost) to predict early the occurrence of lung cancer. The metabolic biomarkers ornithine and palmitoylcarnitine showed significant power for early lung cancer diagnosis.

Багатокласовий класифікатор на основі бінарних логістичних регресій, одержаних за принципами МГУА

Introduction. The issue of accuracy improvement in classification tasks is always topical, and various approaches have been developed, applied in accordance with the peculiarities of the problem formulation and properties of the feature space. Among the most effective models, classifiers based on multiple logistic regressions have proved themselves. Purpose. The aim of the paper is to develop an algorithm for solving multiclassification problems on the basis of binary logistic models built by the stepwise multiple logistic regression algorithm of the Stepwise type, improved according to the principles of the method of group accounting of arguments. Methods. The paper proposes a modification of the stepwise algorithm for creating binary multivariate logistic regressions Stepwise, where it is proposed to optimize the algorithm parameters in accordance with the principles of the method of group consideration of arguments: significance levels by the logarithmic likelihood ratio test for inclusion and exclusion of model arguments. The choice of optimal parameters is realized in accordance with an external criterion that takes into account the balance of classification accuracy of training and test samples and the balance of class classification accuracy. Subsequently, the binary class models obtained by the one-versus-all principle are combined into a multiclass classifier that returns the answer according to the maximum likelihood of the class. The comparison of classification models obtained by the classical Stepwise algorithm and the one proposed in the robot is carried out on the medical data of the publicly available Internet resource Kaggle. Conclusion. The paper substantiates and demonstrates the advantages of classifiers based on logistic multivariate regressions optimized according to the principles of the method of group consideration of arguments relative to the classical version of the Stepwise algorithm. The effective application of the algorithm in solving multiclass classification problem is shown.

A prediction model of liver fat fraction and presence of non-alcoholic fatty liver disease (NAFLD) among patients with overweight or obesity.

The global prevalence of non-alcoholic fatty liver disease (NAFLD) has attained a level of 25.24%. The prevalence of NAFLD in China has exhibited an upward trajectory in parallel with the increasing incidence of obesity over the preceding decade. In order to comprehensively assess hepatic lipid deposition in individuals with overweight or obesity, we have devised a pioneering prognostic formula that capitalizes on clinical parameters. To this end, we have conducted a cross-sectional cohort study involving 149 overweight or obese subjects. Magnetic resonance imaging proton density fat fraction (MRI-PDFF) has been employed to evaluate the extent of liver fat accumulation. Through univariate analysis, we have identified potential factors, and the definitive elements in the prediction model were selected utilizing the forward stepwise regression algorithm. The Shang Hai Steatosis Index (SHSI) incorporates alanine aminotransferase (ALT), aspartate aminotransferase (AST), fasting insulin, and 1-h postload glycaemic levels, thereby furnishing the capability to predict NAFLD with an area under the receiver operator characteristic (AUROC) of 0.87. By establishing a threshold value of 10.96, determined through Youden's index, we have achieved a sensitivity of 69.57% and a specificity of 88.24%. The Spearman correlation coefficient between liver fat fraction ascertained by MRI-PDFF and that predicted by the SHSI equation amounts to 0.74. Consequently, the SHSI equation affords a dependable means of predicting the presence of NAFLD and liver fat accumulation within the overweight and obese population.

Development and Assessment of Seasonal Rainfall Forecasting Models for the Bani and the Senegal Basins by Identifying the Best Predictive Teleconnection

The high variability of rainfall in the Sahel region causes droughts and floods that affect millions of people every year. Several rainfall forecasting models have been proposed, but the results still need to be improved. In this study, linear, polynomial, and exponential models are developed to forecast rainfall in the Bani and Senegal River basins. All three models use Atlantic sea surface temperature (SST). A fourth algorithm using stepwise regression was also developed for the precipitation estimates over these two basins. The stepwise regression algorithm uses SST with covariates, mean sea level pressure (MSLP), relative humidity (RHUM), and five El Niño indices. The explanatory variables SST, RHUM, and MSLP were selected based on principal component analysis (PCA) and cluster analysis to find the homogeneous region of the Atlantic with the greatest predictive ability. PERSIANN-CDR rainfall data were used as the dependent variable. Models were developed for each pixel of 0.25° × 0.25° spatial resolution. The second-order polynomial model with a lag of about 11 months outperforms all other models and explains 87% of the variance in precipitation over the two watersheds. Nash–Sutcliffe efficiency (NSE) values were between 0.751 and 0.926 for the Bani River basin and from 0.175 to 0.915 for the Senegal River basin, for which the lowest values are found in the driest area (Sahara). Results showed that the North Atlantic SST shows a more robust teleconnection with precipitation dynamics in both basins.