Predictive Regression Research Articles

Combination of Regression and ARIMA Methods ( Reg – ARIMA ) Stock Price Prediction Model

Combination of Regression and ARIMA Methods ( Reg – ARIMA ) Stock Price Prediction Model

Spatial extrapolation of cadmium concentration in terrestrial mosses using multiple linear regression model predictions across French biogeographical regions.

The French Moss Survey employs forest mosses as indicators to monitor the deposition of atmospheric substances, notably focusing on cadmium (Cd), a known carcinogenic and contributor to respiratory illnesses. This comprehensive study encompasses 55 variables to understand Cd accumulation in terrestrial mosses in France. These variables include moss species, tree cover, biogeographical markers, land use area, proximity to road and rail networks, soil concentration of Cd and atmospheric concentration and deposition of Cd using a physical model. The response variable undergoes a complementary log-log transformation to constrain prediction values within the maximum Cd content in mosses. We have built a regression model to improve predictions, considering the impacts of covariates in France. This model retains biogeographical effects, leading to data segmentation into four distinct biogeographical zones: Atlantic, Continental, Mediterranean and Alpine. Subsequently, zone-specific regression models are explored to refine predictions and consider the impacts of covariates specific to each region, such as those related to railways and roads of the Mediterranean zone. Our biogeographical models effectively mitigate spatial correlation issues and yield accurate predictions, as evidenced by the leave-one-out cross-validation assessment. Compared to ordinary kriging map, the regression prediction maps highlight the contributions of certain covariates, such as the EMEP atmospheric transport model, to areas with high Cd concentrations. Furthermore, these maps exhibit new areas with high (resp. low) Cd concentrations due to high (resp. low) values of the covariates.

Development and validation of a nomogram for predicting early neurological deterioration in patients with moderate traumatic brain injury: a retrospective analysis

ObjectiveEarly neurological deterioration (END) greatly affects prognosis of moderate traumatic brain injury (TBI). This study aimed to develop and validate a nomogram to predict the occurrence of END in patients with moderate TBI.MethodsA total of 371 patients with moderate TBI were enrolled and divided into the training (n = 260) and validation (n = 111) groups at a ratio of 7:3. Univariate and multivariate logistic regression analyses were used to identify the significant factors for END, which were used to develop a nomogram. The discrimination of the nomogram was evaluated using area under the receiver operating characteristic curves (AUC), the calibration was evaluated using calibration curves and Hosmer-Lemeshow tests. Decision curve analysis (DCA) was used to evaluate the net benefit of the model for patients.ResultsIn the training group, multivariate logistic regression demonstrated that GCS score, epidural hematoma, intracerebral hemorrhage, fibrinogen, and D-dimer were independent risk factors for END in patients with moderate TBI. A nomogram was constructed using the logistic regression prediction model. The AUCs of the nomogram in the training and validation groups were 0.901 and 0.927, respectively. The calibration curves showed that the predicted probability was consistent with the actual situation in both the training and validation sets. DCA curves demonstrated significantly better net benefit with the model. Then a web-based calculator was generated to facilitate clinical application.ConclusionThe present study developed and validated a model to predict END in patients with moderate TBI. The nomogram that had good discrimination, calibration, and clinical utility can provide clinicians with an effective and accurate tool for evaluating the occurrence of END after moderate TBI.

Evaluating the trustworthiness of explainable artificial intelligence (XAI) methods applied to regression predictions of Arctic sea-ice motion.

Abstract Recent advances in explainable artificial intelligence (XAI) methods show promise for understanding predictions made by machine learning (ML) models. XAI explains how the input features are relevant or important for the model predictions. We train linear regression (LR) and convolutional neural network (CNN) models to make one-day predictions of sea-ice velocity in the Arctic from inputs of present-day wind velocity and previous-day ice velocity and concentration. We apply XAI methods to the CNN and compare explanations to variance explained by LR. We confirm the feasibility of using a novel XAI method (i.e. global layerwise relevance propagation (LRP)) to understand ML model predictions of sea-ice motion by comparing to established techniques. We investigate a suite of linear, perturbation-based, and propagation-based XAI methods in both local and global forms. Outputs from different explainability methods are generally consistent in showing that wind speed is the input feature with the highest contribution to ML predictions of ice motion, and we discuss inconsistencies in spatial variability of the explanations. Additionally, we show that CNN relies on both linear and non-linear relationships between the inputs and uses non-local information to make predictions. LRP shows that wind speed over land is highly relevant for predicting ice motion offshore. This provides a framework to show how knowledge of environmental variables (i.e. wind) on land could be useful for predicting other properties (i.e. sea-ice velocity) elsewhere.

HiRXN: Hierarchical Attention-Based Representation Learning for Chemical Reaction.

In recent years, natural language processing (NLP) techniques, including large language modeling (LLM), have contributed significantly to advancements in organic chemistry research. Chemical reaction representations provide a link between NLP models and chemistry prediction tasks and enable the translation of complex chemical processes into a format that NLP models can understand and learn from. However, previous representation methods fail to adequately consider the hierarchical and structural information inherent in chemical reactions. Here, we propose a tool named HiRXN to learn the comprehensive representation of chemical reactions based on their hierarchical structure. In order to significantly enhance feature engineering for machine learning (ML) models, HiRXN develops an effective tokenization method called RXNTokenizer to capture atomic microenvironment features with multiradius. Then, the hierarchical attention network is used to integrate information from atomic microenvironment-level and molecule-level to accurately understand chemical reactions. The experimental results show that HiRXN is capable of representing chemical reactions and achieves remarkable performance in terms of reaction regression and classification prediction tasks. A web server has been developed to provide a specialized service that accepts Reaction SMILES as input and provides predicted results. The Web site is accessible at http://bdatju.com.

Ensemble Machine Learning for Predictive Accuracy and Experimental Corroboration on transition metal-based electrodes for Supercapacitor

Machine learning (ML) proves highly effective in refining the electrochemical performance of electrode materials through precise tuning of compositional parameters. This study employs a data-driven ensemble learning approach, integrating classification and regression techniques to investigate the electrochemical behavior of transition metal-based electrodes with varying Ni:Co stoichiometry. A resample filter is introduced to the dataset to enhance predictive accuracy, yielding closely aligned classification and regression predictions. ML predictions discern that Ni-rich electrodes are apt for high-capacity, while Co-rich excels in high-rate applications. The characterizations of developed electrode materials XRD, FT-IR, EDS, and XPS confirmed the formation of desired compositions. The XPS analysis indicates that nickel and cobalt are in variable Ni2+/Ni3+ and Co2+/Co3+ oxidation states, enabling a reversible charge storage mechanism, while phosphorus is present in the pentavalent state. The TEM micrograph suggests that the nanoparticle size is extremely small, in the range of 1-3 nm, which is significant for providing a high surface-to-volume ratio, thereby enhancing specific capacitance (Csp). Experimental findings highlight N1C2 as the superior electrode, exhibiting impressive capacitance (2247.6 F g−1 at 3 A g−1), robust rate retention (78.6% from 3 to 20 A g−1), and substantial capacitance retention (91.3% after 10,000 charge-discharge cycles), which is in line with ML prediction. The experimental validation highlights the excellent agreement (percentage error: 2.48 to 8.46%) between the capacitance that was predicted, and the capacitance achieved experimentally. Predicted rate and cyclic retention curves align closely with experimental results, with minimal errors (0.03 to 19.3% and 3.95 to 19.64%, respectively). Lastly, the introduction of reverse structural and operational feature engineering contributes to the optimization of electrode material performance.

An Approach to Forecast Quality of Water Effectively Using Machine Learning Algorithms

Background:: The quality of water directly or indirectly impacts the health and environmental well-being. Data about water quality can be evaluated using a Water Quality Index (WQI). Computing WQI is a quick and affordable technique to accurately summarise the quality of water. Objective:: The objective of this study is to find strategies for data preparation to categorize a dataset on the water quality in two remote Indian villages in different geographic locations, to predict the quality of water, and to identify low-quality water before it is made accessible for human consumption. Methods:: To accomplish this task, four water quality features Nitrate, pH, Residual Chlorine, and Total Dissolved Solids which are crucial for human consumption, are considered to dictate the quality of water. Methods used in handling these features include five steps that are data preprocessing with min-max normalization, finding WQI, using feature correlation to identify parameter importance with WQI, application of supervised machine learning regression models such as Random Forest (RF), Multiple Linear Regression (MLR), Gradient Boosting (GB) and Support Vector Machine (SVM) for WQI prediction. Then, a variety of machine learning classification models, including K-Nearest Neighbour (KNN), Support Vector Classifier (SVC), and Multi-layer Perceptron (MLP), are ensembled with Logistic Regression (LR), acting as a meta learner, to create a stack ensemble model classifier to predict the Water Quality Class (WQC) more accurately. Results:: The examination of the testing model revealed that RF regression and MLR algorithms performed best in predicting the WQI with mean absolute error (MAE) of 0.003 and 0.001 respectively. Mean square error (MSE), root mean square error (RMSE), R squared (R2), and Explained Variance Score (EVS) findings are 0.002,0.005,0.988 and 0.998 respectively with RF while 0.001,0.031,0.999 and 0.999 respectively with MLR. Meanwhile, for predicting WQC, the stack model classifier showed the best performance with an Accuracy of 0.936, F1 score of 0.93, and Matthews Correlation Coefficient (MCC) of 0.893 for the dataset of Lalpura and Accuracy of 0.991, F1 Score of 0.991 and MCC of 0.981 respectively for the dataset of Heingang. Conclusion:: This study explores a method for predicting water quality that combines easy and feasible water quality measurements with machine learning. The stack model classifier performed best for multiclass classification, according to this study. To ensure that the highest quality of water is given throughout the year, information from this study will motivate researchers to look into the underlying root causes of the quality variations.

Predictive model for orbital target volume for reconstruction in hyperostotic sphenoid wing meningiomas.

To devise a predictive model for estimating the requisite volume of the orbit in patients poised for resection of hyperostotic spheno-orbital meningiomas. The predictive regression model was conceived through the retrospective analysis of perioperative radiological data from 25 patients who initially underwent surgery at the Burdenko Neurosurgery Center for hyperostotic spheno-orbital meningiomas grade I. The model quality metrics were evaluated utilizing the performance library in the R programming language, including the Akaike Information Criterion, Bayesian Information Criterion, adjusted R-squared, Root Mean Squared Error, and Sigma. An optimal model was discerned based on a comprehensive set of quality metrics. An assessment of the linear relationship between the dynamics of the orbital volume ratio, the absolute and relative dimensions of the orbital volume and the positional alteration of the eyeglobes measured preoperatively was performed using linear regression. The best performance out of the 175 models tested showed a linear model that incorporated the preoperative orbital volume ratios, accounting for the tumor's soft tissue component, and the extent of proptosis. The established linear correlation between the globe's position alteration and the volume index dynamics was employed to predict the "optimal" orbital volume. The developed predictive model suggests an "ideal" orbital volume based on data about the preoperative orbital volume ratios and the desired level of proptosis correction. A customized approach to repairing cranio-orbital bone defects in these patients is expected to consistently produce better functional and esthetic surgical results.

Assessment of the Rainfed Wheat Productivity in a Changing Climate in Irbid, Jordan Using Statistical Downscaling and Random Forest Regression Prediction Under RCP4.5 & 8.5 Pathways.

Jordan faces risks associated with climate change and endeavors to sustain resources through the SDGs. This study examines the impacts of climate change on rainfed wheat productivity in Irbid Governorate during the historical period (1994-2021) and simulated for the future (2024-2100) using climate models (SDGs 1, 2, 11, 12, 13, & 15). Monthly precipitation, mean temperature, and annual wheat yield data were collected for the period (1994-2021) and analyzed using the Mann-Kendall test with Sen’s slope to investigate the historical trends and elaborate Phi-k correlation coefficient to determine their association. Monthly precipitation and mean temperature projections were collected from CSIRO-MK3.6.0 & GFDL-ESM2M CMIP5 ensemble models under two concentration pathways: RCP4.5 and RCP8.5. The projections were biased and corrected by the dynamic bilinear interpolation method and Delta, EQM, and QM statistical downscaling to enhance the projections' reliability and performance in capturing the region's climate. Taylor diagram was utilized to choose the best model to represent observed data. Cumulative Distribution Function and Probability Density Function Curves were plotted to describe the changes over decades based on climate models. Future wheat yields were predicted using the Random Forest Regression model. The results revealed a non-significant increase in total annual precipitation of 1.8% and 13.8% and a rise in annual mean temperature of 5.4% and 2.7% for Irbid and Baqura stations, respectively during the baseline timeframe on which wheat yield increased by 21.3%. The CSIRO model outperformed the GFDL model with greater fidelity in simulating historical monthly precipitation and mean temperatures. The CSIRO-MK3.6.0 model indicates precipitation and temperature shifts for near and far future periods. Temperature increases are expected to have more severe impacts in the far future (2073-2100) while projected decreases in precipitation of 0.06 mm/day and 0.10 mm/day under RCP4.5 and RCP8.5 affect yield reductions by 11.42 kg/ha and 12.21 kg/ha. Temperature increases of 2.7°C and 4.1°C under RCP 4.5 & 8.5 correspond to yield decreases of 24.825 kg/ha and 33.395 kg/ha, respectively. The study highlights the need for enhanced adaptation strategies; cultivating resilient wheat varieties, promoting crop rotation, improving meteorological monitoring and risk response approaches, and providing timely weather forecasts to mitigate the adverse effects of climate change on wheat productivity and ensure sustainable agriculture in Jordan.

The Relationship Between the Optical Properties of Articular Cartilage and the Biomechanical Parameters of the Tissue.

This study investigates the relationship between the optical properties of articular cartilage with its biomechanical parameters. The absorption and reduced scattering coefficients of articular cartilage, and average maximum penetration depth, average maximum lateral spread, and average path length of photons were estimated by optical measurements and Monte Carlo simulation. The equilibrium and instantaneous moduli, and initial fibrillar network, strain-dependent fibrillar network, and nonfibrillar network moduli, and initial and deformation-dependent permeability of the tissue were estimated by multistep stress-relaxation measurement and fibril-reinforced poroelastic modeling. The relationship between the optical properties and biomechanical parameters was assessed using predictive regression modeling. A strong relationship was found between reduced scattering coefficients, averaged maximum penetration depth, averaged maximum lateral spread, and average path length of photons with equilibrium, instantaneous, and initial fibrillar network moduli. We attribute this relationship to the collagen fibers as the main contributor to the scattering and biomechanical properties of the tissue.

How to estimate the probability of a live birth after one or more complete IVF cycles? the development of a novel model in a single-center.

To develop a predictive tool in the form of a Nomogram based on the Cox regression model, which incorporates the impact of the length of treatment cycles on the outcome of live birth, to evaluate the probability of infertile couples having a live birth after one or more complete cycles of In Vitro Fertilization (IVF), and to provide patients with a risk assessment that is easy to understand and visualize. A retrospective study for establishing a prediction model was conducted in the reproductive center of Shenzhen Zhongshan Obstetrics & Gynecology Hospital (formerly Shenzhen Zhongshan Urology Hospital). A total of 4413 patients who completed ovarian stimulation treatment and reached the trigger were involved. 70% of the patients were randomly placed into the training set (n = 3089) and the remaining 30% of the patients were placed into the validation set (n = 1324) randomly. Live birth rate (LBR) and cumulative LBR (CLBR) were calculated for one retrieval cycle and the subsequent five frozen embryo transfer (FET) cycles. Proportional Hazards (PH) Assumption test was used for selecting the parameter in the predictive model. A Cox regression model was built based on the basis of training set, and ROC curves were used to test the specificity and sensitivity of the prediction model. Subsequently, the validation set was applied to verify the validity of the model. Finally, for a more intuitive assessment of the CLBR more intuitively for clinicians and patients, a Nomogram model was established based on predictive model. By calculating the scores of the model, the clinicians could more effectively predict the probability for an individual patient to obtain at least one live birth. In the fresh embryo transfer cycle, the LBR was 38.7%. In the first to fifth FET cycle, the optimal estimate and conservative estimate CLBRs were 59.95%, 65.41%, 66.35%, 66.58%, 66.61% and 56.81%, 60.84%, 61.50%, 61.66%, 61.68%, respectively. Based on PH test results, the potential predictive factors for live birth were insemination method, infertility factors, serum progesterone level (R = 0.043, p = 0.059), and luteinizing hormone level (R = 0.015, p = 0.499) on the day initiated with gonadotropin, basal follicle-stimulating hormone (R = -0.042, p = 0.069) and BMI (R = -0.035, p = 0.123). We used ROC curve to test the predictive power of the model. The AUC was 0.782 (p < 0.01, 95% CI: 0.764-0.801). Then the model was verified using the validation data. The AUC was 0.801 (p < 0.01, 95% CI: 0.774-0.828). A Nomogram model was built based on potential predictive factors that might influence the event of a live birth. The Cox regression and Nomogram prediction models effectively predicted the probability of infertile couples having a live birth. Therefore, this model could assist clinicians with making clinical decisions and providing guidance for patients. N/A.

The ability of CD4dimCD8+ T cells to distinguish between Epstein-Barr virus-associated hemophagocytic lymphohistiocytosis and pediatric infectious mononucleosis.

Epstein-Barr virus (EBV)-related hemophagocytic lymphohistiocytosis (EBV-HLH) and infectious mononucleosis (IM) are characterized by fever, hepatomegaly, and splenomegaly, but HLH has a 50% lethality rate. Therefore, this study aimed to compare the laboratory findings in differentiating EBV-HLH children from IM children who have fever, hepatomegaly, or splenomegaly. A total of 131 IM patients and 29 EBV-HLH pediatric patients with fever, hepatomegaly, or splenomegaly were enrolled in our study. The clinical traits and laboratory findings were analyzed, and a predictive regression analysis was also performed. EBV-HLH patients had a lower set of diagnostic markers, which included fibrinogen (FIB), white blood cells (WBC), hemoglobin (Hb), and platelet (PLT), compared to IM patients. Triglyceride (TG) and ferritin were elevated obviously in EBV-HLH patients compared to IM patients. CD4dimCD8+ T cells are highly activated T cells. EBV-HLH patients experienced a significant decrease in the absolute number and ratio of CD4dimCD8+T cells (CD4dimCD8+T#, CD4dimCD8+T%) compared to IM patients in our study. The AUC of CD4dimCD8+ T # in the diagnosis of EBV-HLH was 0.920 with a sensitivity of 86.2% and a specificity of 90.1%. A logistic regression analysis was developed to improve sensitivity. The results revealed that lower levels of Hb, FIB, and CD4dimCD8+T cells were risk factors for EBV-HLH. The regression model for separating EBV-HLH from EBV-IM had an AUC of was 0.996 with a sensitivity of 100% and a specificity of 95.3%. The ratio and absolute number of CD4dimCD8+T cells were decreased in IM and EBV-HLH patients. EBV-HLH can be identified in IM patients with fever, hepatomegaly, or splenomegaly by detecting Hb, FIB, and the absolute number of CD4dimCD8+T cells.

Professional Baseball Pitchers with Increased In-Pitch Stance Hip Flexion Excursion Demonstrate Higher Shoulder Internal Rotation Torque and Elbow Varus Torque With No Appreciable Ball Velocity Benefit.

Relationships between the stance leg's lumbopelvic control and throwing arm kinetics have been established in baseball pitchers; however, no study to date has analyzed the role hip flexion excursion may play in generation of throwing arm kinetics/ball velocity. To discern a plausible relationship between lead and stance hip flexion excursion with throwing arm kinetics, full body kinematics, and ball velocity in professional (PRO) baseball pitchers. PRO (n=283) pitchers instructed to throw 8-12 fastball pitches were evaluated with 3D-motion capture (480 Hz). PRO pitchers were divided into quartiles based on "high" and "low" lead and stance hip flexion excursion with ANOVA analysis between subgroups. Multi-regression models examined the association between lead and stance hip excursion, controlling for anthropometric parameters, with ball velocity and throwing arm kinetics. The quartile divisions noted no difference for ball speed between subgroups(p>0.05). PRO pitchers with Hi Lead Hip, Hi Stance Hip flexion (Q1) had significantly higher shoulder internal rotation torque (5.1 vs 4.8%BWxBH, p=0.03) and elbow varus torque (5.0 vs 4.7%BWxBH, p=0.02) compared to pitchers with Hi Lead Hip, Low Stance Hip flexion(Q3). Similarly, pitchers with Lo Lead Hip, Hi Stance Hip flexion (Q2) had significantly greater shoulder internal rotation torque (5.2 vs 4.8, 4.9%BWxBH, pmax=0.02) and elbow varus torque(5.1 vs 4.7, 4.8%BWxBH, pmax=0.03) compared to Q3 and pitchers with Lo Lead Hip, Lo Stance Hip flexion (Q4). When controlling for anthropometric parameters, the multi-regression model noted stance hip flexion excursion was moderately predictive for elbow varus torque (p<0.001, B=0.010 [0.004, 0.016], β=0.20), while weakly predictive for elbow medial force (p=0.018, B=0.064 [0.014, 0.115], β=0.148), shoulder anterior force (p=0.026, B=0.066 [0.008, 0.123], β=0.130), and shoulder internal rotation torque (p=0.002, B=0.009 [0.003, 0.015], β=0.180). PRO pitcher subgroups with high stance hip flexion (Q1, Q2) consistently demonstrated higher elbow varus torque and shoulder internal rotation torque compared to pitchers with low stance hip flexion(Q3, Q4); these results were corroborated with a moderate and weakly predictive regression analysis when controlling for confounders, respectively. This suggests professional pitchers with increased stance hip flexion may be at higher risk for increased throwing arm kinetics, as a surrogate for injury risk, with no appreciable ball velocity benefit noted. This may be an area of improvement in identifying & correcting maladaptive pitching patterns for players, pitching coaches, & athletic trainers.

Prediction of tricuspid regurgitation regression after mitral valve transcatheter edge-to-edge repair using three-dimensional transesophageal echocardiography

Abstract Aims We aimed to identify three-dimensional echocardiographic predictors of tricuspid regurgitation (TR) regression in patients with functional TR of moderate or greater severity undergoing mitral valve transcatheter edge-to-edge repair to optimize patient selection and improve clinical outcomes. Methods and Results This retrospective study analyzed 61 patients (mean age 81.3±7.6 years; 55.7% males) who underwent mitral valve transcatheter edge-to-edge repair. Two-dimensional transthoracic echocardiography was performed pre- and 1-month post-procedurally, while three-dimensional transesophageal echocardiography was performed pre-procedurally. We collected data on clinical variables, medications, and detailed echocardiographic measurements to evaluate procedural outcomes. TR severity was semi-quantitatively assessed and categorized. At the 1-month follow-up, TR severity had regressed in 43% of patients. A lower prevalence of atrial fibrillation, smaller left atrial volume index, and smaller right atrial area were significantly associated with TR regression. Multivariate analysis revealed the tricuspid valve annulus perimeter, area, and area change as significant predictors of post-procedure TR regression; tricuspid valve annulus perimeter was the strongest predictor among the three indicators (area under the receiver operating characteristic curve, 0.84 [95% confidence interval: 0.75–0.94], p&amp;lt;0.001). Receiver operating characteristic curve analysis indicated that tricuspid valve annulus perimeter cut-off of ≤13.75 cm was the best predictor of post-procedure TR regression. Additionally, tricuspid valve annulus area ≤13.55 cm² and annulus area change ≥17.5% were predictors of post-procedure TR regression. Conclusion In patients with relatively severe mitral regurgitation with a non-dilated tricuspid annulus and significant change in tricuspid valve annulus area, mitral valve transcatheter edge-to-edge repair may lead to TR regression.

Biofilm Formation Capabilities of Lactobacillus Species Isolated from Selected Fermented Food Products Using a Statistical Approach

Background: This study investigates the biofilm formation capabilities of Lactobacillus species isolated from fermented cassava and corn products. Understanding biofilm formation is crucial for evaluating the probiotic potential of these species, as biofilm-forming ability influences their survival and functionality in host environments. Methods: Nine bacterial isolates, including Lactobacillus fermentum, L. ghanensis, L. delbrueckii, L. plantarum, Lactococcus lactis, L. reuteri, Lysinibacillus sphaericus, Bacillus cereus, and B. pacificus, were assessed for biofilm production using the microtiter plate assay. After crystal violet staining, optical density (OD) values were measured at 570 nm spectrophotometrically. Based on OD values, isolates were classified into four categories: no biofilm, weak, moderate, and strong biofilm formation. Statistical analyses, including two-stage least squares regression, were employed to evaluate biofilm formation trends and predictors. Results: The predictive regression model was highly significant (R² = 0.987, F = 122.618, p &lt; 0.0001). Biofilm formation strength varied, with the highest mean percentage observed in the moderate group (31.29%), followed by weak (27.41%), strong (20.46%), and no biofilm (20.05%). Among the isolates, Lactobacillus fermentum exhibited the highest rate of strong biofilm formation (46.1%), while Lysinibacillus sphaericus showed none. Moreover, The highest biofilm formation was observed at 37°C (31.29%), followed by 25°C (27.41%), and 45°C (20.46%). Similarly, biofilm formation was highest at pH 6.5 (30.41%), followed by pH 7.5 (25.39%) and pH 4.5 (20.05%). Lactobacillus fermentum exhibited the highest strong biofilm formation (46.1%) at 37°C and pH 6.5. Conclusion: Biofilm formation in Lactobacillus species is species-specific and environmentally influenced by temperature and pH. Lactobacillus fermentum demonstrated strong biofilm formation, making it a promising candidate for probiotic applications.

Salmon Consumption Behavior Prediction Based on Bayesian Optimization and Explainable Artificial Intelligence

Predicting seafood consumption behavior is essential for fishing companies to adjust their production plans and marketing strategies. To achieve accurate predictions, this paper introduces a model for forecasting seafood consumption behavior based on an interpretable machine learning algorithm. Additionally, the Shapley Additive exPlanation (SHAP) model and the Accumulated Local Effects (ALE) plot were integrated to provide a detailed analysis of the factors influencing Shanghai residents’ intentions to purchase salmon. In this study, we constructed nine regression prediction models, including ANN, Decision Tree, GBDT, Random Forest, AdaBoost, XGBoost, LightGBM, CatBoost, and NGBoost, to predict the consumers’ intentions to purchase salmon and to compare their predictive performance. In addition, Bayesian optimization algorithm is used to optimize the hyperparameters of the optimal regression prediction model to improve the model prediction accuracy. Finally, the SHAP model was used to analyze the key factors and interactions affecting the consumers’ willingness to purchase salmon, and the Accumulated Local Effects plot was used to show the specific prediction patterns of different influences on salmon consumption. The results of the study show that salmon farming safety and ease of cooking have significant nonlinear effects on salmon consumption; the BO-CatBoost nonlinear regression prediction model demonstrates superior performance compared to the benchmark model, with the test set exhibiting RMSE, MSE, MAE, R2 and TIC values of 0.155, 0.024, 0.097, 0.902, and 0.313, respectively. This study can provide technical support for suppliers in the salmon value chain and help their decision-making to adjust their corporate production plan and marketing activities

Effect of nozzle arrangement on heat dissipation performance and mechanical efficiency loss of bionic texture gear

Abstract Friction and wear are key issues in gear meshing. Lubrication, therefore, is used to reduce heat. The position and direction of the flow of the lubrication determined by the way the nozzles are injecting the fluid and by the texture on the flanks of the teeth, both have an influence on the heat transfer. Furthermore, the geometric characteristics of the tooth texture influences the friction behaviour directly. This article presents investigations on the impact of the tooth texture on wear. A proposal to prepare the tooth flanks with arc shaped grooves is made presenting the manufacturing process. Different measurements to determine the tooth flank texture and the friction are performed. Measurements showed that the arc groove gear had lower friction coefficients, wear depth, surface roughness Ra, and maximum tooth profile peak height Rq compared to conventional gears. The dynamic process of gear meshing is simulated by FEM analysis to understand the physics of heat flow and friction in detail，which revealed that the arc groove gear had reduced sliding distance, contact pressure, and wear depth, along with a higher convective heat transfer coefficient. Based on the measurement results of orthogonal test, two different nozzle arrangements are predicted: one is linear regression prediction (nozzle arrangement A), and the other is nonlinear particle swarm optimization neural network prediction (nozzle arrangement B). By measuring the torque loss value under the two nozzle arrangement conditions, it is found that the nozzle arrangement B is the best arrangement. The specific parameters are: injection distance is 56.0674 mm, injection angle is 8.4527°, nozzle diameter is 3.3955 mm, and pinion speed is 3000r / min. Under this condition, the mechanical efficiency loss is reduced by 94.48 %.

Risk Prediction Model of Peristomal Skin Complications Among Patients with Colorectal Cancer and an Ostomy: A Cross-sectional Study in Shanghai, China.

To analyze the risk factors for peristomal skin complications (PSCs) in patients with colorectal cancer and an ostomy, construct a prediction model, and verify its effectiveness. In this cross-sectional study, researchers recruited 265 patients with an ostomy at the stoma clinic of a tertiary hospital, from May 2022 to August 2023. Patients were divided into two groups: complications group (n = 81) and no complications group (n = 184). Researchers constructed a logistic regression prediction model using univariate and multivariate analyses. From May 2023 to August 2023, a validation group of 135 patients with an ostomy was selected for external validation of the model. The incidence of PSCs was 30.57% in the modeling group and 30.37% in the validation group. The predictor variables were preoperative and postoperative health education, tumor therapy within 3 months, preoperative stoma positioning, stoma height, stoma type, and excrement state. The Hosmer-Lemeshow test yielded a P value of .513, the area under the ROC curve was 0.872, the Youden index was 0.561, and the sensitivity and specificity were 0.827 and 0.734. For external validation, the Hosmer-Lemeshow test yielded a P value of .835, the area under the ROC curve was 0.887, and the sensitivity and specificity were 0.905 and 0.720, respectively. The prediction model demonstrates good predictive efficacy and can serve as a reference for clinical caregivers in identifying patients at high risk of PSCs.

Symbolic regression for strength prediction of eccentrically loaded concrete-filled steel tubular columns

Concrete-filled steel tube (CFST) columns are widely employed in high-rise buildings, long-span bridges, and seismic-resistant structures due to their superior load-bearing capacity, structural efficiency, and resilience under extreme loading conditions. This study uses symbolic regression with structural design code provisions to predict the eccentric strength of concrete filled-steel tubular columns with circular shape (CCFST) and rectangular shape (RCFST). Previous studies have used two distinct approaches for estimating eccentric strength: explainable models based on theoretical derivations and black-box models derived from machine learning (ML) methods. This study proposes a hybrid model derived from the design code standards, with performance enhanced by the symbolic regression technique. This model is based on a comprehensive experimental database of 464 tests for CCFST columns and 313 tests for RCFST columns under eccentric loading from various research papers. The developed code-based symbolic regression (C-SR) displays both robust and interpretable, demonstrating high prediction accuracy with mean values of the prediction-to-actual ratios of 1.006 and 0.997 and coefficient of variation (CoV) values of 0.117 and 0.098 for CCFSTs and RCFSTs, respectively, while providing explainable mathematical expressions that align with the mechanical principles of code provisions. The developed C-SR model is benchmarked against EC4 and AISC360 standards and evaluated against the various ML techniques, demonstrating acceptable performance. The results highlight the C-SR model’s effectiveness in providing reliable predictions and valuable insights for practical engineering applications.

Individualised Therapy for Obstructive Sleep Apnoea: Predictive Models and Anatomical Phenotyping of Mandibular Advancement Devices Responses.

This non-randomised clinical study aimed to identify the phenotypic characteristics that distinguish responders from non-responders. Additionally, it sought to establish a predictive model for treatment response to obstructive sleep apnoea (OSA) using mandibular advancement devices (MAD), based on the analysed phenotypic characteristics. This study, registered under identifier NCT05596825, prospectively analysed MAD treatment over 6 years using two-piece adjustable appliances according to a standardised protocol. Two response definitions aligned with the latest International Consensus Statement on OSA severity were established. Logistic regression and CHAID models integrated baseline clinical, anthropometric, cephalometric anatomical, soft tissue characteristics and physiological upper airway variables. A total of 112 patients completed the study: 64 responders and 48 non-responders according to response definition 1, and 81 responders and 31 non-responders according to response definition 2. Responders to MAD treatment had lower body mass index (BMI), neck and waist circumference, Epworth Sleepiness Scale scores, apnoea-hypopnea index (AHI), snoring intensity on the Visual Analog Scale, CPAP pressure, and higher T90% and minSaO2. Patients exhibiting greater anatomical imbalance, smaller airway volume, smaller minimum cross-sectional area (CSAmin) and longer airway length demonstrated a poorer response to treatment. Airway length, initial T90% and anterior facial height collectively formed a highly predictive logistic regression model for response definition 1. Jarabak's ratio, gonial angle, CSAmin, airway length, initial BMI and baseline AHI constituted a highly predictive model for the second response definition. Furthermore, the CHAID regression tree established cutoff values for the variables that form the predictive models.