Establish Prediction Models Research Articles

Impact of hepatic steatosis on the efficacy of antiviral treatment for chronic hepatitis B and the establishment of predictive model: a cohort study

AimMetabolic dysfunction-associated steatotic disease (MASLD) and chronic hepatitis B (CHB) are prevalent liver disorders. Ongoing discussions investigate the impact of MASLD on the therapeutic outcomes of CHB.MethodsA cohort of 320 CHB patients on antiviral therapy (including NAs and PEG IFNα) were included and categorized into CHB + MASLD (n = 125) and CHB group (n = 195). The treatment response rates, Kaplan–Meier survival analysis, and Cox regression were assessed between the two groups to investigate the impact of MASLD on antiviral responses in patients with CHB.ResultsAt weeks 24 and 48, the CHB + MASLD group displayed a higher HBsAg response rate than the CHB group (24 weeks: 11.5% vs. 3.8%, p = 0.026; 48 weeks: 24.4% vs. 8.4%, p = 0.001). The pgRNA response was also higher in the CHB + MASLD group at both time points (24 weeks: 30.9% vs. 19.7%, p = 0.163; 48 weeks: 48.8% vs. 28.3%, p = 0.049). Kaplan–Meier survival analysis revealed a shorter median time to HBsAg response at 48 weeks for the CHB + MASLD group (HR = 3.251, 40 weeks vs. 42.5 weeks, p = 0.002). This is particularly evident among individuals who are negative for HBeAg (48w: 24.2% vs 12.2%, p = 0.005). KM survival analysis demonstrated that the CHB + MASLD group was more likely to achieve HBsAg response (HR = 2.428, p = 0.039).COX regression analysis identified age (HR = 0.948, p = 0.005), antiviral regimen (NAs + PEG IFNα: HR = 5.33, p < 0.001; PEG IFNα: HR = 1.099, p = 0.93), baseline HBsAg level (HR = 0.648, p = 0.009), and MASLD presence (HR = 3.321, p = 0.002) as independent predictors for HBsAg response. Time-ROC analysis showed that these factors effectively predicted HBsAg decline (24 weeks: AUC = 0.902; 48 weeks: AUC = 0.890). The model demonstrated strong discriminative power, calibration, and clinical relevance.ConclusionIn CHB patients without significant liver fibrosis who receive antiviral therapy, concurrent MASLD enhances HBsAg response, particularly in HBeAg-negative patients. Factors like younger age, NAs with PEG IFNα therapy, lower initial HBsAg levels, and MASLD presence predict treatment success. Further investigations are required to elucidate the impact of diverse metabolic disorders on the advancement of liver fibrosis.Trial registrationRegistry and the registration No. Of the study/trial: ChiCTR23000 74064(2023-07-28).

Analysis of prognostic factors and establishment of a recurrence risk prediction model for papillary thyroid carcinoma based on BRAF stratification.

Predicting the likelihood of papillary thyroid carcinoma (PTC) recurrence is crucial for improving patient outcomes. The association between the BRAF V600E (BRAF) mutation and PTC recurrence remains controversial. Our goal was to determine prognostic features of PTC patients and construct models for predicting recurrence risk according to BRAF mutation status. A total of 811 PTC patients whose clinical information and survival data were available were included in this study. Independent prognostic variables of PTC identified by screening via LASSO-Cox regression analysis were then used to construct nomograms. The performance of the predictive models was assessed according to the C-index, ROC curve, validation curve, and decision curve analyses. Kaplan-Meier curves were used to analyze differences between patients grouped according to prognostic factors and relapse risk. Multivariate Cox regression analysis demonstrated that extrathyroidal extension (ETE), vascular tumor thrombus, and lymph node yield (LNY) were correlated with recurrence-free survival (RFS) in the BRAF mutation-negative group, while extranodal extension (ENE), number of metastatic lymph node (NMLN), pathological stage, and vascular tumor thrombus were correlated with RFS in the BRAF mutation-positive group. The mutation-stratified predictive models demonstrated better performance than the model without stratification, as indicated by the greater C-index values (0.880 vs. 0.859 vs. 0.753), AUC values (1-year AUC: 0.946 vs. 0.947 vs. 0.758; 3-year AUC: 0.889 vs. 0.871 vs. 0.760; 5-year AUC: 0.845 vs. 0.793 vs. 0.758), and net clinical benefit. The calibration curves at 1 year, 3 years, and 5 years showed good consistency. The bootstrap internal validation had good AUC values exceeding 0.8 and showed a well-fitting calibration curve. Significant differences in RFS were observed between the low-risk and high-risk groups (P < 0.001). Stratifying patients based on their BRAF mutation status can facilitate the development of better and more targeted postoperative management strategies. Nomograms for BRAF mutation positive and negative patients were developed to precisely and consistently predict recurrence risk in PTC patients.

Prediction model establishment of prognosis factors for distant metastasis of hepatocellular carcinoma based on the SEER database.

Distant metastasis in hepatocellular carcinoma (HCC) is an important indicator of poor patient prognosis. Identifying patients who are at high risk of metastasis early on is essential for creating personalized treatment plans, yet currently, there is a scarcity of effective predictive tools. To investigate the effects of different factors on distant metastasis in HCC patients and to establish a clinical prediction model for predicting distant metastasis in HCC patients. Our study retrospectively examined 22,318 patients diagnosed with confirmed HCC from the SEER database. Prognostic factors for developing distant metastases in HCC patients were identified by univariate and multivariate logistic regression analyses. Utilizing data from a multivariate logistic regression analysis, we created a nomogram. Its predictive precision was evaluated by analyzing the calibration curve, the area under the curve (AUC) of the receiver operating characteristic curve, decision curve assessment (DCA), and Kaplan-Meier (KM) curve analysis of overall survival. Finally,the nomogram was visualized with an online calculator. We identified six independent prognostic factors: ethnicity, marital status, tumor size, survival time, surgery, and radiotherapy. The nomogram constructed from these six factors showed good calibration, discrimination, and clinical application value after calibration curve analysis, receiver operating characteristic curve analysis and DCA curve analysis. Besides, KaplanMeier survival curves also demonstrated that this nomogram had predictive accuracy. In this research, a nomogram model was created to accurately predict distant metastasis risk in patients with HCC. This study provides guidance for optimizing individual therapies and making better clinical decisions.

Predicting the Zinc Content in Rice from Farmland Using Machine Learning Models: Insights from Universal Geochemical Parameters

Zinc (Zn) is an essential nutrient for the human body and is prone to deficiency. Supplementing Zn through zinc-enriched cereals is of great significance in addressing the widespread issue of zinc deficiency. However, there is no simple linear correlation between the soil zinc content and rice grain zinc content, which poses challenges for zoning zinc-enriched rice cultivation based on the soil Zn content. Therefore, accurately predicting the zinc content in rice grains is of great importance. To verify the robustness of the prediction model and expand its applicability, this study established a prediction model using 371 sets of previously collected and tested rice grain and root zone soil samples from the Pearl River Delta and Heyuan regions in Guangdong. The model was validated using the data from 65 sets of rice and root zone soil samples collected and analyzed in Zijin and Dongyuan counties, Heyuan, in 2023. The results show that zinc absorption by rice grains is controlled by multiple factors, primarily related to the soil S, P, CaO, Mn, TFe2O3, TOC, and SiO2/Al2O3 ratio. Both the artificial neural network model and random forest model demonstrated a good predictive performance across large regions. However, in the Heyuan region, the random forest model outperformed the artificial neural network model, with an R2 of 0.79 and an RMSE of 0.05 when the predicted data were compared against the measured BAFZn of the rice. This suggests that predicting the zinc content in rice grains based on the soil macro-elements (including oxides) and TOC is feasible, and, within certain regional boundaries, the prediction model is robust and widely applicable. This study provides valuable insights into the rational development of zinc-enriched rice in the Heyuan region and offers a useful reference for establishing prediction models of the beneficial element content in rice grains in areas with limited data.

Identification of depressive symptoms in adolescents using machine learning combining childhood and adolescence features

BackgroundDepressive symptoms in adolescents can significantly affect their daily lives and pose risks to their future development. These symptoms may be linked to various factors experienced during both childhood and adolescence. Machine learning (ML) has attracted substantial attention in the field of adolescent depression; however, studies establishing prediction models have primarily considered childhood or adolescent features separately, resulting in a lack of analyses that incorporate factors from both stages.MethodsWe collected 39 features related to childhood and adolescence. Using the maximum relevance-minimum redundancy method and four ML algorithms, we determined the optimal feature subset for identifying depressive symptoms and constructed child-adolescent models. Stepwise logistic regression and four ML methods were employed to create demographic and combined models, respectively. The performance of each model was evaluated using a test set, and SHapley Additive exPlanations (SHAP) were utilized to interpret the models’ prediction results.ResultsThe proposed child-adolescent models exhibited superior performance on the test set than the demographic and combined models (AUC: 0.835–0.879 versus 0.530 and 0.840–0.876, respectively). The optimal feature subset included two childhood features (relationship quality with peers and parental absence) and four adolescence features (social trust, academic pressure, importance of the internet for entertainment, and positive parenting behaviour). These features were found to be more effective than demographic characteristics in distinguishing depressive symptoms in adolescents.ConclusionsThis study demonstrates the correlation between adolescence depressive symptoms and specific factors from both childhood and adolescence, as well as the potential of ML to predict it. These findings may serve as a reference for future intervention studies.

Time Series Data Augmentation for Energy Consumption Data Based on Improved TimeGAN.

Predicting the time series energy consumption data of manufacturing processes can optimize energy management efficiency and reduce maintenance costs for enterprises. Using deep learning algorithms to establish prediction models for sensor data is an effective approach; however, the performance of these models is significantly influenced by the quantity and quality of the training data. In real production environments, the amount of time series data that can be collected during the manufacturing process is limited, which can lead to a decline in model performance. In this paper, we use an improved TimeGAN model for the augmentation of energy consumption data, which incorporates a multi-head self-attention mechanism layer into the recovery model to enhance prediction accuracy. A hybrid CNN-GRU model is used to predict the energy consumption data from the operational processes of manufacturing equipment. After data augmentation, the prediction model exhibits significant reductions in RMSE and MAE along with an increase in the R2 value. The prediction accuracy of the model is maximized when the amount of generated synthetic data is approximately twice that of the original data.

Advanced Machine Learning to Predict Coronary Artery Disease Severity in Patients with Premature Myocardial Infarction.

Studies using machine learning to identify the target characteristics and develop predictive models for coronary artery disease severity in patients with premature myocardial infarction (PMI) are limited. In this observational study, 1111 PMI patients (≤55 years) at Tianjin Chest Hospital from 2017 to 2022 were selected and divided according to their SYNTAX scores into a low-risk group (≤22) and medium-high-risk group (>22). These groups were further randomly assigned to a training or test set in a ratio of 7:3. Lasso-logistic was initially used to screen out target factors. Subsequently, Lasso-logistic, random forest (RF), k-nearest neighbor (KNN), support vector machine (SVM), and eXtreme Gradient Boosting (XGBoost) were used to establish prediction models based on the training set. After comparing prediction performance, the best model was chosen to build a prediction system for coronary artery severity in PMI patients. Glycosylated hemoglobin (HbA1c), angina, apolipoprotein B (ApoB), total bile acid (TBA), B-type natriuretic peptide (BNP), D-dimer, and fibrinogen (Fg) were associated with the severity of lesions. In the test set, the area under the curve (AUC) of Lasso-logistic, RF, KNN, SVM, and XGBoost were 0.792, 0.775, 0.739, 0.656, and 0.800, respectively. XGBoost showed the best prediction performance according to the AUC, accuracy, F1 score, and Brier score. In addition, we used decision curve analysis (DCA) to assess the clinical validity of the XGBoost prediction model. Finally, an online calculator based on the XGBoost was established to measure the severity of coronary artery lesions in PMI patients. In summary, we established a novel and convenient prediction system for the severity of lesions in PMI patients. This system can swiftly identify PMI patients who also have severe coronary artery lesions before the coronary intervention, thus offering valuable guidance for clinical decision-making.

Establishment and validation of predictive model of ARDS in critically ill patients

BackgroundAcute respiratory distress syndrome (ARDS) is a prevalent complication among critically ill patients, constituting around 10% of intensive care unit (ICU) admissions and mortality rates ranging from 35 to 46%. Hence, early recognition and prediction of ARDS are crucial for the timely administration of targeted treatment. However, ARDS is frequently underdiagnosed or delayed, and its heterogeneity diminishes the clinical utility of ARDS biomarkers. This study aimed to observe the incidence of ARDS among high-risk patients and develop and validate an ARDS prediction model using machine learning (ML) techniques based on clinical parameters.MethodsThis prospective cohort study in China was conducted on critically ill patients to derivate and validate the prediction model. The derivation cohort, consisting of 400 patients admitted to the ICU of the Peking University Third Hospital(PUTH) between December 2020 and August 2023, was separated for training and internal validation, and an external data set of 160 patients at the FU YANG People's Hospital from August 2022 to August 2023 was employed for external validation. Least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression were used to screen predictor variables. Multiple ML classification models were integrated to analyze and identify the best models. Several evaluation indexes were used to compare the model performance, including the area under the receiver-operating-characteristic curve (AUC) and decision curve analysis (DCA). SHapley Additive ex Planations (SHAP) is used to interpret ML models.Results400 critically ill patients were included in the analysis, with 117 developing ARDS during follow-up. The final model included gender, Lung Injury Prediction Score (LIPS), Hepatic Disease, Shock, and combined Lung Contusion. Based on the AUC and DCA in the validation group, the logistic model demonstrated excellent performance, achieving an AUC of 0.836 (95% CI: 0.762–0.910). For external validation, comprising 160 patients, 44 of whom developed ARDS, the AUC was 0.799 (95% CI: 0.723–0.875), significantly outperforming the LIPS score alone.ConclusionCombining the LIPS score with other clinical parameters in a logistic regression model provides a more accurate, clinically applicable, and user-friendly ARDS prediction tool than the LIPS score alone.

Risk factors analysis of hypokalemia after radical resection of esophageal cancer and establishment of a nomogram risk prediction model.

This study aimed to explore the risk factors of hypokalemia after radical resection of esophageal cancer (EC) and establish a nomogram risk prediction model to evaluate hypokalemia risk after esophagectomy. Thus, this study provides a reference for the clinical development of intervention measures. Clinical data of EC patients who underwent radical surgery from January 2020 to November 2022 in the First Affiliated Hospital of Guangxi Medical University were retrospectively collected. The relevant variables were screened using multivariate logistic regression analysis with IBM SPSS 25.0 and R 4.2.0 software, and a nomogram for predicting hypokalemia risk was established. The established nomogram was evaluated by receiver operating characteristic (ROC), calibration, and decision curves. The model was also internally validated by 1000 bootstrap resampling methods. After radical EC resection, the incidence rate of hypokalemia in 213 patients was 19.2% (41/213). The hemoglobin levels, total serum protein, serum albumin, calcium ion concentration, direct bilirubin, prothrombin time (PT), and activated partial thromboplastin time (APTT) were related (p < 0.05). The multivariate logistic analysis showed that the white blood cell count, serum albumin level, direct bilirubin, and operation time were risk factors for hypokalemia after radical EC resection (p < 0.05). The area under the ROC curve (AUC) was 0.764, demonstrating the good discriminative ability of the established nomogram for hypokalemia prediction. The calibration curve showed a good fit between the predicted and actual observed probabilities. The model maintained a high C-index in the internal validation (C-index = 0.758), supporting that the nomogram can be widely used for hypokalemia prediction. The prediction model for hypokalemia risk with individualized scores based on the patient's white blood cell count, serum albumin level, direct bilirubin, and operation time can screen out high-risk patients who might develop hypokalemia. It is of certain reference value for clinicians to screen and follow up with patients with emphasis and to formulate preoperative and postoperative intervention strategies.

Simplifying coronary artery disease risk stratification: development and validation of a questionnaire-based alternative comparable to clinical risk tools.

Coronary artery disease (CAD) comprises one of the leading causes of morbidity and mortality both in the European population and globally. All established clinical risk stratification scores and models require blood lipids and physical measurements. The latest reports of the European Commission suggest that attracting health professionals to collect these data can be challenging, both from a logistic and cost perspective, which limits the usefulness of established models and makes them unsuitable for population-wide screening in resource-limited settings, i.e., rural areas. Therefore, the aim of this study was to develop and externally validate a questionnaire-based risk stratification model on a population scale at minimal cost, i.e., the Questionnaire-Based Evaluation for Estimating Coronary Artery Disease (QUES-CAD) to stratify the 10-year incidence of coronary artery disease. Cox proportional hazards (CoxPH) and Cox gradient boosting (CoxGBT) models were trained with 10-fold cross-validation using combinations of ten questionnaire variables on the White population of the UK Biobank (n=448,818) and internally validated the models in all ethnic minorities (n=27,433). The Lifelines cohort was employed as an independent external validation population (n=97,770). Additionally, we compared QUES-CAD's performance, containing only questionnaire variables, to clinically established risk prediction tools, i.e., Framingham Coronary Heart Disease Risk Score, American College of Cardiology/American Heart Association pooled cohort equation, World Health Organization cardiovascular disease risk charts, and Systematic Coronary Risk Estimation 2 (SCORE2). We conducted partial log-likelihood ratio (PLR) tests and C-index comparisons between QUES-CAD and established clinical prediction models. In the external validation set, QUES-CAD exhibited C-index values of CoxPH: 0.692 (95% Confidence Interval [CI]: 0.673-0.71) and CoxGBT: 0.699 (95% CI: 0.681-0.717) for the male population and CoxPH: 0.771 (95% CI: 0.748-0.794) and CoxGBT: 0.759 (95% CI: 0.736-0.783) for the female population. The addition of measurement-based variables and variables that require a prior medical examination (i.e., insulin use, number of treatments/medications taken, prevalent cardiovascular disease [other than CAD, and stroke diagnosed by a doctor]) and the further addition of biomarkers/other measurements (i.e., high-density lipoprotein [HDL] cholesterol, total cholesterol, and glycated haemoglobin) did not significantly improve QUES-CAD's performance in most instances. C-index comparisons and PLR tests showed that QUES-CAD performs and fits the data at least as well as the clinical prediction models. QUES-CAD performs comparably to established clinical prediction models and enables a population-wide identification of high-risk individuals for CAD. The model developed and validated herein relies solely on ten questionnaire variables, overcoming the limitations of existing models that depend on physical measurements or biomarkers. University Medical Center Groningen.

Machine learning-based prediction and model interpretability analysis for algal growth affected by microplastics.

Microplastics (MPs), the plastic debris smaller than 5mm, are ubiquitous in waterbodies and have been shown to be toxic to aquatic organisms, especially to microalgae. The aim of this study is to use machine learning models to predict the effects of MPs on algal growth and to evaluate the relative importance of different features (MP properties, algal characteristics, and experimental conditions) through model interpretability analysis. Based on literature search, 408 samples were collected as inputs for the models. Three integrated machine learning algorithms, Random Forest (RF), Categorical Boosting (CatBoost), and Light Gradient Boosting Machine (LightGBM), were used to construct classification prediction models for algal growth. Our results show that the LightGBM model yields the best performance, with a total accuracy rate of 0.8305 and a Kappa value of 0.7165. The model interpretability analysis indicates that "Exposure time", "MP concentrations", and "MP size" are the most influential features affecting algal growth. For "Exposure time", which belongs to experimental conditions, 72-216h of MP exposure was found to exert the greatest effects on algal growth. The impact of MPs on algal growth increases with increasing MP concentrations over the range of 0 to 300mg/L. Smaller MPs exert more effects on algal growth, and MPs are more likely to inhibit algal growth when the ratio of algal cell size to MP size is higher. Our study successfully established prediction models for evaluating the effects of various MP properties on algal growth. This study also provides insights into the prediction of MP toxicity in organisms.

Experimental Investigation and Modeling of Surface Roughness in BTA Deep Hole Drilling with Vibration Assisted.

The surface roughness of hole machining greatly influences the mechanical properties of parts, such as early fatigue failure and corrosion resistance. The boring and trepanning association (BTA) deep hole drilling with axial vibration assistance is a compound machining process of the tool cutting and the guide block extrusion. At the same time, the surface of the hole wall is also ironed by the axial large amplitude and low-frequency vibration of the guide block. The surface-forming mechanism is very complicated, making it difficult to obtain an effective theoretical analytical model of the surface roughness of the hole wall through kinematic analysis. In order to achieve accurate prediction of the surface quality of the hole wall, the chip-breaking mechanism and the hole wall formation mode of BTA deep hole vibration drilling were analyzed. The influence of drilling spindle speed, feed, amplitude, and vibration frequency on the surface roughness of the hole wall during BTA deep hole vibration drilling was illustrated by a single-factor experiment. A four-factor and three-level test scheme was designed by using the Box-Behnken design (BBD) experimental design method. A surface roughness prediction model for hole wall machining was established based on the response surface methodology. The accuracy of the prediction model was analyzed through ANOVA, and the complex correlation coefficient of the model was 0.9948, indicating that the prediction model can better reflect the mapping relationship between vibration drilling parameters and surface roughness. After optimization analysis and experimental verification, the obtained vibration drilling parameters can achieve smaller surface roughness. The error between the predicted value of the model and the experimental measurement value is 8.65%. The established prediction model is reliable and can accurately predict the surface roughness of the hole wall of BTA deep hole axial vibration drilling, providing a theoretical basis for the surface quality control of the machining hole wall. It can be applied to process optimization in practical production.

Multi-modal prediction of extracorporeal support-a resource intensive therapy, utilizing a large national database.

Extracorporeal membrane oxygenation (ECMO) is among the most resource-intensive therapies in critical care. The COVID-19 pandemic highlighted the lack of ECMO resource allocation tools. We aimed to develop a continuous ECMO risk prediction model to enhance patient triage and resource allocation. We leveraged multimodal data from the National COVID Cohort Collaborative (N3C) to develop a hierarchical deep learning model, labeled "PreEMPT-ECMO" (Prediction, Early Monitoring, and Proactive Triage for ECMO) which integrates static and multi-granularity time series features to generate continuous predictions of ECMO utilization. Model performance was assessed across time points ranging from 0 to 96hours prior to ECMO initiation, using both accuracy and precision metrics. Between January 2020 and May 2023, 101400 patients were included, with 1298 (1.28%) supported on ECMO. PreEMPT-ECMO outperformed established predictive models, including Logistic Regression, Support Vector Machine, Random Forest, and Extreme Gradient Boosting Tree, in both accuracy and precision at all time points. Model interpretation analysis also highlighted variations in feature contributions through each patient's clinical course. We developed a hierarchical model for continuous ECMO use prediction, utilizing a large multicenter dataset incorporating both static and time series variables of various granularities. This novel approach reflects the nuanced decision-making process inherent in ECMO initiation and has the potential to be used as an early alert tool to guide patient triage and ECMO resource allocation. Future directions include prospective validation and generalizability on non-COVID-19 refractory respiratory failure, aiming to improve patient outcomes.

Analysis of risk factors and establishment of prediction model for lower extremity deep vein thrombosis after lumbar fusion surgery

BackgroundLower extremity deep vein thrombosis (LEDVT) is a common complication after orthopedic surgery. Currently, a reliable assessment tool is lacking to evaluate the risk of postoperative LEDVT in patients undergoing lumbar fusion surgery. This study aims to explore the risk factors for LEDVT formation after lumbar fusion surgery and establish a predictive model for it.MethodsData of patients admitted for multi-center spinal surgery from May 2022 to September 2023 were retrospectively collected. Patients were divided into DVT and non-DVT groups based on the occurrence of LEDVT after surgery. Potential risk factors were initially identified through intergroup comparative analysis and single-factor logistic regression, which were considered candidate indicators. LASSO regression was applied to select candidate indicators, and the filtered variables were included in a multivariable logistic regression model. Nomogram and dynamic nomogram were constructed to visualize the model, and the model was subsequently validated.ResultsFactors including weakened lower extremity muscle strength, intraoperative blood loss, walking impairment, and Venous reflux/ Varicose veins were included in the multivariable logistic regression model. The results showed that the model had an area under the receiver operating characteristic curve of 0.870, 0.777 and 0.750 for the training set, internal validation set, and external validation set, respectively. Nomograms and web-based dynamic nomograms were created based on the multivariable logistic regression model. The model exhibited good performance in calibration curves and decision analysis.ConclusionThe study identified weakened lower extremity muscle strength, intraoperative blood loss, walking impairment, and Venous reflux/ Varicose veins as risk factors for LEDVT formation following lumbar fusion surgery. The predictive tool established based on the logistic regression model demonstrated good performance and can be considered for assessing the risk of LEDVT formation after lumbar fusion surgery.

MinLinMo: a minimalist approach to variable selection and linear model prediction

Generating prediction models from high dimensional data often result in large models with many predictors. Causal inference for such models can therefore be difficult or even impossible in practice. The stand-alone software package MinLinMo emphasizes small linear prediction models over highest possible predictability with a particular focus on including variables correlated with the outcome, minimal memory usage and speed. MinLinMo is demonstrated on large epigenetic datasets with prediction models for chronological age, gestational age, and birth weight comprising, respectively, 15, 14 and 10 predictors. The parsimonious MinLinMo models perform comparably to established prediction models requiring hundreds of predictors.

Numerical Simulation and Bayesian Optimization CatBoost Prediction Method for Characteristic Parameters of Veneer Roller Pressing and Defibering

Defibering equipment is employed in the production of scrimber for the purpose of wood veneer rolling, cutting, and directional fiber separation. However, the current defibering equipment exhibits a notable degree of automation deficiency, relying more on manual operation and empirical methods for process control, which impedes the stability of the defibering equipment and the defibering quality. This study presented an in-depth finite element analysis of the roller-pressing process for veneer defibering equipment, and a prediction method incorporating numerical simulation and ensemble learning was proposed through data collection and feature selection. The objective is to integrate this method into the intelligent decision-making system of the equipment, with the aim of improving the productivity of the equipment and effectively stabilizing the product quality. The simulation process and the analysis of the results in ABAQUS 2020 revealed that the roller gap and roller velocity of the defibering equipment, as well as the geometrical parameters of the veneer, have a significant influence on the defibering effect. Combining these factors, 702 simulation experiments were devised and executed, and a database was constructed based on the model-building parameters and simulation outcomes. The strain and stress observed in the simulation results served to represent the veneer force and veneer deformation. The CatBoost algorithm was used to establish prediction models for the key parameters of the defibering effect, and the Bayesian Optimization and 5-fold cross-validation techniques enabled the strain and stress prediction models to achieve coefficients of determination of 0.98 and 0.97 for the training and test datasets, respectively. Shapley Additive Explanation was used to provide insight into the contribution of each feature, thereby guiding the selection of feature parameters and simplifying the model. The results show that the scheme can effectively determine the core process parameters of the defibering equipment and then provide a practical control strategy for intelligent online control.

Establishment of a decision tree prediction model for the treatment of intracranial aneurysms using temperature-sensitive embolic agents based on geometric features

Establishment of a decision tree prediction model for the treatment of intracranial aneurysms using temperature-sensitive embolic agents based on geometric features

Quality prediction and process optimization in abrasive waterjet cutting of ultra‐thick carbon fiber reinforced polymer

AbstractSurface roughness and delamination extent are key indicators for assessing the surface quality of machined carbon fiber reinforced polymer (CFRP) laminates. This study conducted an in‐depth investigation on the uncontrollable surface quality of abrasive water jet (AWJ) machining of ultra‐thick CFRP. Considering the sensitivity of the laminate structure to the jet impact angle, an optimized response surface methodology was used to analyze the effects of process parameters on surface roughness. Additionally, the differences in surface roughness on both sides of the kerf under non‐orthogonal cutting were examined. Based on these analyses, a surface roughness prediction model was developed. Furthermore, the study comprehensively analyzed the effects of material mechanical properties and process parameters on delamination damage. Using dimensional analysis, a predictive model for the maximum delamination length, considering jet energy dissipation, was established. The novelty of this research is to reveal the difference in surface quality on both sides of the kerf under non‐orthogonal cutting, and to establish prediction models for surface roughness and delamination damage based on the energy dissipation characteristics of ultra‐thick CFRP AWJ machining. This research provides essential modeling foundations for enhancing the controllability of AWJ machining quality for ultra‐thick CFRP complex components.Highlights Surface roughness and delamination damage impact AWJ use for ultra‐thick CFRP. Non‐orthogonal cutting causes surface quality differences on kerf sides. A model for kerf surface roughness at any jet impact angle was established. The delamination mechanism in internal path cutting was analyzed. The delamination length model for ultra‐thick CFRP laminates was established.

Genetic risk stratification and risk factors of early menopause in women: a multi-center study utilizing polygenic risk scores.

This study aims to evaluate the utility of polygenic risk scores (PRS) in women with early menopause (EM) and to investigate the clinical characteristics and risk factors associated with EM based on genetic risk. Genotyping data and clinical data from women with EM and women with normal age of menopause retrieved from UK Biobank were used for early menopause risk prediction model establishment. Subsequently, 99 women diagnosed with EM and 1027 control women underwent PGT-M were recruited for model validation from across eight hospitals in China. According to PRS percentiles, these participants were further classified into high risk and intermediate risk groups. Characteristics among women at different risk levels were compared, and risk factors with early menopause were also statistical analyzed. The proportion of women at high risk in EM and control groups; Characteristics with significant difference among women at different risk levels; risk factors associated with EM. The proportion of high-risk women in the EM group was significantly higher than that in control women underwent PGT-M (Group PGT-M) (OR = 3.78), and that in women with normal age menopause from UK Biobank (Group UKB) (OR = 5.11). Notably, the women with high risk of EM exhibited distinct characteristics compared to women with the intermediate-risk of EM, and identified several risk factors associated with EM. We established a PRS model to serves as a valuable instrument for EM risk prediction. The exploratory analysis revealed that women with high risk of EM exhibited a higher height, suggesting EM related genetic loci may also influence growth and development level. Several risk factors were found to be potentially associated with EM, such as excessive familial contentment, COVID-19 vaccination, staying up late, and the husband's engagement in smoking and alcohol abuse.

Establishment of risk prediction model for pneumonia infection in elderly severe patients and analysis of prevention effect of 1M3S nursing plan under early warning mode

To construct a risk prediction model for elderly severe patients with pneumonia infection, and analyze the prevention effect of 1M3S nursing plan under early warning mode. Firstly, 180 elderly severe patients admitted to the department of intensive care unit (ICU) of the Second Affiliated Hospital of Xingtai Medical College from September 2020 to September 2021 were enrolled. Their clinical data were collected and retrospectively analyzed, and they were divided into infected group and non-infected group according to whether they developed severe pneumonia. The risk factors affecting severe pneumonia in elderly severe patients were screened by univariate and multifactorial analysis methods, and the risk prediction model was constructed. The predictive efficiency of the model was analyzed by receiver operator characteristic curve (ROC curve). Then the risk prediction model was applied to prospectively include 60 high-risk elderly patients with severe pneumonia admitted from December 2021 to August 2022. The patients were randomly divided into study group and control group by envelope method, with 30 cases in each group. Both groups were given routine nursing. On this basis, the study group adopted 1M3S nursing scheme [standardized nursing management (1M), improving nursing skills (S1), optimizing nursing service (S2), ensuring nursing safety (S3)] in the early warning mode for intervention. Acute physiology and chronic health evaluation II (APACHE II) and Murray lung injury score were compared between the two groups before intervention and 7 days after intervention. Among 180 elderly severe patients, 34 cases were infected with pneumonia (18.89%). The proportion of patients with Glasgow coma scale (GCS) ≤ 8, duration of mechanical ventilation > 7 days, use of antibiotics, poor oral hygiene, hospital stay > 15 days and albumin ≤ 30 g/L in the infected group were significantly higher than those in the non-infected group. Multivariate Logistic regression analysis showed that duration of mechanical ventilation > 7 days, use of antibiotics, GCS score≤ 8, hospital stay > 15 days, albumin ≤ 30 g/L and poor oral hygiene were all independent risk factors for severe pneumonia in elderly severe patients. The odds ratio (OR) values were 3.180, 3.394, 1.108, 1.881, 1.517 and 2.512 (all P < 0.05). ROC curve analysis showed that the area under the ROC curve (AUC) of the prediction model to predict severe pneumonia in elderly severe patients was 0.838, 95% confidence interval was 0.748-0.927, sensitivity and specificity were 81.25% and 72.57%, respectively, and the Youden index was 0.538. (2) There was no significantly difference in general data between the study group and the control group, which was comparable. After intervention, the APACHE II score and Murray lung injury score of the two groups were significantly decreased, and the APACHE II score and Murray lung injury score of the study group were significantly lower than those of the control group (APACHE II score: 3.15±1.02 vs. 3.81±0.25, Murray lung injury score: 5.01±1.12 vs. 6.55±0.21, both P < 0.01). There are many risk factors affecting the development of severe pneumonia in elderly severe patients. The risk prediction model based on duration of mechanical ventilation > 7 days, hospital stay > 15 days, GCS score≤ 8, albumin ≤ 30 g/L, poor oral hygiene and history of combined antibacterial use has high predictive efficacy. The intervention of 1M3S nursing scheme in the early warning mode can effectively reduce the risk of severe pneumonia in elderly severe patients, and significantly improve the pathophysiological status.