Robust Prediction Model Research Articles

Multiple Regression Analysis and Non-Dominated Sorting Genetic Algorithm II Optimization of Machining Carbon-Fiber-Reinforced Polyethylene Terephthalate Glycol Parts Fabricated via Additive Manufacturing Under Dry and Lubricated Conditions

The present research deals with the processing of the additively manufactured Carbon-Fiber-Reinforced Polymer (CFRP) under dry and lubricated cutting conditions, focusing on the generated surface roughness. The cutting speed, feed, and depth of cut were selected as the continuous variables. A comparison between the generated surface roughness of the dry and the lubricated cuts revealed that the presence of coolant contributed towards reducing surface roughness by more than 20% in most cases. Next, a regression analysis was performed with the obtained measurements, yielding a robust prediction model, with the determination coefficient R2 being equal to 94.65%. It was determined that feed and the corresponding interactions contributed more than 45% to the model’s R2, followed by the depth of cut and the machining condition. In addition, the cutting speed was the variable with the least effect on the response. The Non-Dominated Sorting Genetic Algorithm 2 (NSGA-II) was employed to identify the front of optimal solutions that consider both minimizing surface roughness and maximizing Material Removal Rate (MRR). Finally, a set of extra experiments proved the validity of the model by exhibiting relative error values, between the measured and predicted roughness, below 10%.

Customer Churn Prediction Using Machine Learning

Customer attrition poses a major issue for the telecommunications sector, resulting in revenue decline and higher expenses for gaining new customers. Predicting churn accurately enables telecom companies to take proactive measures to retain customers. This study leverages specifically Random Forest, Decision Tree, and XGBoost, to develop a robust churn prediction model. These algorithms analyze customer behavioral data, including call duration, internet usage, billing history, and complaints, to identify potential churners. Decision Tree provides an interpretable model, Random Forest improves predictive accuracy through ensemble learning, and XGBoost enhances performance with gradient boosting and optimized handling of imbalanced datasets. The proposed model assists telecom companies in classifying customers based on their churn risk, enabling the implementation of targeted retention approaches like tailored discounts and rewards programs. By integrating advanced machine learning techniques, telecom service providers can enhance customer retention, minimize the churn rates, and improves the business sustainability. The study highlights the importance of data-driven decision-making in the telecom sector, demonstrating how predictive analytics can optimize customer relationship management and drive profitability. Key Words: Predicting telecom customer churn, machine learning techniques, Random Forest algorithm, Decision Tree method, XGBoost model, retention of customers, analytics for forecasting, gradient boosting techniques, telecom sector.

Advancements in Bearing Health Monitoring and Remaining Useful Life Prediction: Techniques, Challenges, and Future Directions

Abstract This paper presents a comprehensive review of recent advancements in bearing health monitoring and remaining useful life (RUL) prediction. It highlights key innovations in anomaly detection, health indicator (HI) construction, degradation modeling, and RUL estimation, examining developments across statistical, machine learning, and deep learning approaches while analyzing their strengths, limitations, and application contexts. Special emphasis is placed on the role of deep learning in capturing complex degradation patterns from multi-dimensional time series data and improving predictive accuracy in dynamic industrial settings. Additionally, this review explores multi-source data fusion techniques, which enhance anomaly detection robustness by integrating information from diverse sensor modalities. By identifying critical challenges and suggesting future research directions, this study aims to advance the development of robust and adaptive prediction models for intelligent maintenance in industrial applications.

Using a Neural Network Architecture for the Prediction of Neurologic Outcome for Out-of-Hospital Cardiac Arrests Using Hospital Level Variables and Novel Physiologic Markers

Out-of-hospital cardiac arrest (OHCA) is a major public health burden due to its high mortality rate, sudden nature, and long-term impact on survivors. Consequently, there is a crucial need to create prediction models to better understand patient trajectories and assist clinicians and families in making informed decisions. We studied 107 adult OHCA patients admitted at an academic Emergency Department (ED) from 2018–2023. Blood samples and ocular ultrasounds were acquired at 1, 6, and 24 h after return of spontaneous circulation (ROSC). Six classes of clinical and novel variables were used: (1) Vital signs after ROSC, (2) pre-hospital and ED data, (3) hospital admission data, (4) ocular ultrasound parameters, (5) plasma protein biomarkers and (6) sex steroid hormones. A base model was built using 1 h variables in classes 1–3, reasoning these are available in most EDs. Extending from the base model, we evaluated 26 distinct neural network models for prediction of neurological outcome by the cerebral performance category (CPC) score. The top-performing model consisted of all variables at 1 h resulting in an AUROC score of 0.946. We determined a parsimonious set of variables that optimally predicts CPC score. Our research emphasizes the added value of incorporating ocular ultrasound, plasma biomarkers, sex hormones in the development of more robust predictive models for neurological outcome after OHCA.

Robust Cluster Prediction Across Data Types Validates Association of Sex and Therapy Response in GBM

Background: Previous studies have described sex-specific patient subtyping in glioblastoma. The cluster labels associated with these “legacy data” were used to train a predictive model capable of recapitulating this clustering in contemporary contexts. Methods: We used robust ensemble machine learning to train a model using gene microarray data to perform multi-platform predictions including RNA-seq and potentially scRNA-seq. Results: The engineered feature set was composed of many previously reported genes that are associated with patient prognosis. Interestingly, these well-known genes formed a predictive signature only for female patients, and the application of the predictive signature to male patients produced unexpected results. Conclusions: This work demonstrates how annotated “legacy data” can be used to build robust predictive models capable of multi-target predictions across multiple platforms.

Spatio-temporal analysis of surface water quality on urban tanks in Coimbatore

Effective monitoring and predicting surface water quality are vital for sustainable water resource control. Traditional in-situ techniques are regularly constrained by their time-consuming nature and restrained spatial coverage. This study seeks to develop a predictive version that combines physio-chemical water quality parameters with remote sensing indices derived from the Sentinel-2A dataset to enhance accuracy and spatial attainment. The research focuses on four urban tanks in Coimbatore namely Krishnampathy, Selvampathy, Kumaraswamy and Ukkadam Periyakulam. The physio-chemical parameters for assessing the water quality which include pH, Electrical Conductivity (EC), Total Dissolved Solids (TDS), Dissolved oxygen (DO), Calcium (Ca), Magnesium (Mg), Total hardness, Chloride (Cl-), Carbonate (Co3-) and Bicarbonate (HCo3-) have been measured, additionally for the detection of surface water extent using remote sensing indices namely Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Modified Normalized Difference Water Index (MNDWI), Water Ratio Index, Normalized Difference Chlorophyll Index (NDCI) indices which were extracted from sentinel-2A datasets. The parameters such as EC, TDS, Ca2+, Cl- and Total hardness show a high coefficient of determination (R2). Correlation and regression techniques have been employed to integrate these datasets, resulting in the development of a robust predictive model. By combining these two information sources, the model is constructed using stepwise regression analysis. The model's accuracy turned into established towards ground-truth records, showing large improvements whilst far away Remote sensing indices had been covered.

Evaluating the impact of ESICM 2023 guidelines and the new global definition of ARDS on clinical outcomes: insights from MIMIC-IV cohort data

BackgroundIn 2023, the European Society of Intensive Care Medicine (ESICM) recommended updated criteria for acute respiratory distress syndrome (ARDS). In 2024, Matthay et al. updated the global ARDS definition in AJRCCM, titled “A New Global Definition of Acute Respiratory Distress Syndrome.” However, the impact of this new definition on ARDS treatments is currently unknown.ObjectiveThis study aims to determine the effect of the new ARDS definition on patients with hypoxemic respiratory failure and study the heterogeneity of patients in the new definition to guide treatment.MethodsClinical consultation data from the Medical Information Mart for Intensive Care IV database were extracted using Structured Query Language based on the PostgreSQL tool (version 10.0). Data were analyzed using Python (version 3.9) and the deep learning framework Pytorch. Kaplan–Meier survival analysis was used to compare survival between the old and new definitions. A hierarchical clustering approach was applied to identify potential ARDS clinical subtypes.ResultsThe new definition diagnosed ARDS earlier and included individuals with lower mortality rates compared with the Berlin definition. Patients meeting the new definition but not the Berlin criteria exhibited a favorable response to non-invasive ventilation strategies (p = 0.009). The XGBoost classifier, trained to predict subphenotypes, achieved an AUC of 0.88 ± 0.02 on the training set. Additionally, mortality was significantly associated with patients with hypoxemia compared with survivors, particularly regarding respiratory parameters. Easily accessible metrics, such as respiratory rate and urea nitrogen (BUN), can help diagnose ARDS in high-risk populations in resource-limited settings.ConclusionsThe new ARDS definition offers advantages in earlier detection, more accurate grading, and more precise diagnosis in resource-limited settings compared with the Berlin definition. This study also established a robust prediction model for early ARDS identification, improving the patient prognosis and reducing the mortality rate.

Predictive risk model of mild cognitive impairment in patients with malignant haematological diseases after haematopoietic stem cell transplantation

ObjectiveThis study is to develop and validate a robust risk prediction model for mild cognitive impairment (MCI) in patients with malignant haematological diseases after haematopoietic stem cell transplantation (HSCT).MethodsIn this study, we analysed the clinical data of the included patients. Logistic regression analysis was used to identify independent risk factors for cognitive impairment after HSCT in patients with malignant haematological diseases, and a risk prediction model was constructed. Multiple cohorts of patients with haematological malignancies after HSCT (282 cases) from the Affiliated Hospital of Xuzhou Medical University and the First People’s Hospital of Yancheng City between April 2019 and February 2022, and patients from the Affiliated Huaian No.1 People’s Hospital of Nanjing Medical University between March 2022 and July 2023 were used for external validation. Logistic regression analysis was performed to develop the predictive model. The predictive value and consistency of the model were evaluated using the area under the curve (AUC) and calibration method, respectively. Decision curve analysis (DCA) was performed to access the utility of the model.ResultsApproximately half (52.26%) of the patients in the study developed mild cognitive impairment (MCI). Older age, allogeneic HSCT, anxiety, graft-versus-host disease, and longer hospital stay were associated with a higher risk of developing MCI. ROC curve analysis confirmed the sound performance of the predictive model and external validation, with AUC of 0.897 and 0.789 respectively. The direction of the calibration curves of the training and validation sets is closer to the diagonal (ideal curve), indicating good model consistency; the DCA curves also show that the model has good predictive ability and stability.ConclusionsWe conclude that it is possible to predict mild cognitive impairment with readily available, mostly pretransplant predictors. The accuracy of the risk prediction models can be improved for use in clinical practice, possibly by adding pretransplant patient-reported functioning and comorbidities.

Machine learning-based prediction for incidence of endoscopic retrograde cholangiopancreatography after emergency laparoscopic cholecystectomy: A retrospective, multicenter cohort study.

Laparoscopic cholecystectomy is the preferred treatment for symptomatic cholelithiasis and acute cholecystitis, with increasing applications even in severe cases. However, the possibility of postoperative endoscopic retrograde cholangiopancreatography (ERCP) to manage choledocholithiasis or biliary injuries poses significant clinical challenges. This study aimed to develop a predictive model for ERCP incidence following emergency laparoscopic cholecystectomy using advanced machine learning techniques. We conducted a retrospective cohort study using the Tokushukai Medical Database, which includes data from 42 hospitals. The study population consisted of adult patients undergoing emergency laparoscopic cholecystectomy. We used four machine learning models-logistic regression, random forest, gradient-boosting decision trees (GBDTs), and multilayer perceptrons on a dataset divided into training/validation and testing groups. We also calculated Shapley additive explanation values for GBDTs to identify variables with larger feature importance. Of 9,695 patients from July 2010 to June 2020, 8,854 met the inclusion criteria. The incidence of postoperative ERCP was 5.7% (362/6,377) and 6.4% (158/2477) in the training/validation and testing datasets, respectively. The GBDT demonstrated superior performance, with the highest predictive capacity for postoperative ERCP. Significant predictors identified included common bile duct dilatation on CT or ultrasound, serum albumin, and lactate dehydrogenase levels, which showed larger feature importance. This study successfully developed a robust predictive model for ERCP following emergency laparoscopic cholecystectomy.

Analysis and Prediction of Grouting Reinforcement Performance of Broken Rock Considering Joint Morphology Characteristics

In tunnel engineering, joint shear slip caused by external disturbances is a key factor contributing to landslides, instability of surrounding rock masses, and related hazards. Therefore, accurately characterizing the macromechanical properties of joints is essential for ensuring engineering safety. Given the significant influence of rock joint morphology on mechanical behavior, this study employs the frequency spectrum fractal dimension (D) and the frequency domain amplitude integral (Rq) as quantitative descriptors of joint morphology. Using Fourier transform techniques, a reconstruction method is developed to model joints with arbitrary shape characteristics. The numerical model is calibrated through 3D printing and direct shear tests. Systematic parameter analysis validates the selected quantitative indices as effective descriptors of joint morphology. Furthermore, multiple machine learning algorithms are employed to construct a robust predictive model. Machine learning, recognized as a rapidly advancing field, plays a pivotal role in data-driven engineering applications due to its powerful analytical capabilities. In this study, six algorithms—Random Forest (RF), Support Vector Regression (SVR), BP Neural Network, GA-BP Neural Network, Genetic Programming (GP), and ANN-based MCD—are evaluated using 300 samples. The performance of each algorithm is assessed through comparative analysis of their predictive accuracy based on correlation coefficients. The results demonstrate that all six algorithms achieve satisfactory predictive performance. Notably, the Random Forest (RF) algorithm excels in rapid and accurate predictions when handling similar training data, while the ANN-based MCD algorithm consistently delivers stable and precise results across diverse datasets.

Automatic Production Planning and Scheduling of Size

Automatic production planning and scheduling are critical for optimizing manufacturing processes in modern industries. This approach integrates advanced algorithms, machine learning, and real-time data analysis to streamline resource allocation, minimize production bottlenecks, and enhance overall efficiency. By automating traditionally manual planning tasks, the system improves accuracy and responsiveness to dynamic production demands. Challenges such as implementation costs, integration with existing systems, and data reliability are addressed through scalable architectures and robust predictive models. This study investigates innovative methodologies to achieve seamless coordination, examining their impact on productivity, cost reduction, and adaptability in competitive markets Key Words: optics, photonics, light, lasers, templates, journals

Development of immune-derived molecular markers for preeclampsia based on multiple machine learning algorithms

Preeclampsia (PE) is a major pregnancy-specific cardiovascular complication posing latent life-threatening risks to mothers and neonates. The contribution of immune dysregulation to PE is not fully understood, highlighting the need to explore molecular markers and their relationship with immune infiltration to potentially inform therapeutic strategies. We used bioinformatics tools to analyze gene expression data from the Gene Expression Omnibus (GEO) database using the GEOquery package in R. Differential expression analysis was performed using the DESeq2 and limma packages, followed by analysis of variance to identify immune-related differentially expressed genes (DEGs). Several machine learning algorithms, including least absolute shrinkage and selection operator (LASSO), bagged trees, and random forest (RF), were used to select immune-related signaling genes closely associated with the occurrence of PE. Our analysis identified 34 immune source–related DEGs. Using the identified PE- and immune source–related genes, we constructed a diagnostic forecasting model employing several ML algorithms. We identified six types of statistically significant immune cells in patients with PE and discovered a strong relationship between biomarkers and immune cells. Moreover, the immune-derived hub genes for PE exhibited strong binding capabilities with drugs, such as alitretinoin, tretinoin, and acitretin. This study presents a robust prediction model for PE that integrates multiple machine learning–derived immune-related biomarkers. Our results indicate that these biomarkers may outperform previously reported molecular signatures in predicting PE and provide insights into the mechanisms underlying immune dysregulation in PE. Further validation in larger cohorts could lead to their clinical application in PE prediction and treatment.

Machine Learning-Based Radiomics Analysis for Identifying KRAS Mutations in Non-Small-Cell Lung Cancer from CT Images: Challenges, Insights and Implications.

Kirsten Rat Sarcoma viral oncogene homolog (KRAS) is a frequently occurring mutation in non-small-cell lung cancer (NSCLC) and influences cancer treatment and disease progression. In this study, a machine learning (ML) pipeline was applied to radiomic features extracted from public and internal CT images to identify KRAS mutations in NSCLC patients. Both datasets were analyzed using parametric (t test) and non-parametric statistical tests (Mann-Whitney U test) and dimensionality reduction techniques. Afterwards, the proposed ML pipeline was applied to both datasets using a five-fold cross-validation on the training set (70/30 train/test split) before being validated on the other dataset. The results show that the radiomic features are significantly different (Mann-Whitney U test; p < 0.05) between the two datasets, despite the use of identical feature extraction methods. Model transferability is therefore difficult to achieve, which became evident during external testing (F1 score = 0.41). Oversampling, undersampling, clustering and harmonization techniques were applied to balance and harmonize the datasets, but did not improve the classification of KRAS mutation presence. In general, due to only a single moderate result (highest test F1 score = 0.67), the accuracy of KRAS prediction is not sufficient for clinical application. In future work, the complexity of KRAS mutation might be addressed by taking submutations into consideration. Larger multicentric datasets with balanced tumor stages, including multi-scanner datasets, seem to be necessary for building robust predictive models.

Artificial intelligence in breast cancer survival prediction: a comprehensive systematic review and meta-analysis.

Breast cancer (BC), as a leading cause of cancer mortality in women, demands robust prediction models for early diagnosis and personalized treatment. Artificial Intelligence (AI) and Machine Learning (ML) algorithms offer promising solutions for automated survival prediction, driving this study's systematic review and meta-analysis. Three online databases (Web of Science, PubMed, and Scopus) were comprehensively searched (January 2016-August 2023) using key terms ("Breast Cancer", "Survival Prediction", and "Machine Learning") and their synonyms. Original articles applying ML algorithms for BC survival prediction using clinical data were included. The quality of studies was assessed via the Qiao Quality Assessment tool. Amongst 140 identified articles, 32 met the eligibility criteria. Analyzed ML methods achieved a mean validation accuracy of 89.73%. Hybrid models, combining traditional and modern ML techniques, were mostly considered to predict survival rates (40.62%). Supervised learning was the dominant ML paradigm (75%). Common ML methodologies included pre-processing, feature extraction, dimensionality reduction, and classification. Deep Learning (DL), particularly Convolutional Neural Networks (CNNs), emerged as the preferred modern algorithm within these methodologies. Notably, 81.25% of studies relied on internal validation, primarily using K-fold cross-validation and train/test split strategies. The findings underscore the significant potential of AI-based algorithms in enhancing the accuracy of BC survival predictions. However, to ensure the robustness and generalizability of these predictive models, future research should emphasize the importance of rigorous external validation. Such endeavors will not only validate the efficacy of these models across diverse populations but also pave the way for their integration into clinical practice, ultimately contributing to personalized patient care and improved survival outcomes. https://www.crd.york.ac.uk/prospero/, identifier CRD42024513350.

Modelling the present potential habitat distribution of the near-threatened endemic species Silene marizii: implications for conservation

Species Distribution Models (SDMs) have been increasingly used in biodiversity conservation tasks, especially for rare and threatened species. In this study, we selected Silene marizii, a near-threatened endemic species distributed in the central and north-western Iberian Peninsula. We used a total of 88 occurrences, reduced to 67 after accounting for spatial correlation, obtained from field surveys, herbaria and GBIF. As potential predictors, we used 34 variables (23 bioclimatic, 7 edaphic, and 4 topographic) in Maxent. The Maxent model performed well, demonstrating high predictive accuracy. The predictors that contributed most to the model were precipitation seasonality, precipitation amount of the driest month, pH index, weight percentage of the sand particles, mean monthly precipitation amount of the coldest quarter, mean diurnal air temperature range and mean daily mean air temperature of the driest quarter. Our results suggested the following areas for further exploration of new populations and, if necessary, for reintroductions and translocation efforts: northern half of Portugal, southern Galicia, the northwest of Zamora province and several mountain ranges in the Central System. In conclusion, robust predictive models are now fundamental for conducting more efficient fieldwork, ultimately improving conservation and management plans for rare and threatened species such as S. marizii.

The value of a metabolic and immune-related gene signature and adjuvant therapeutic response in pancreatic cancer.

Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive malignancy characterized by a dismal prognosis. Treatment outcomes exhibit substantial variability across patients, underscoring the urgent need for robust predictive models to effectively estimate survival probabilities and therapeutic responses in PDAC. Metabolic and immune-related genes exhibiting differential expression were identified using the TCGA-PDAC and GTEx datasets. A genetic prognostic model was developed via univariable Cox regression analysis on a training cohort. Predictive accuracy was assessed using Kaplan-Meier (K-M) curves, calibration plots, and ROC curves. Additional analyses, including GSAE and immune cell infiltration studies, were conducted to explore relevant biological mechanisms and predict therapeutic efficacy. An 8-gene prognostic model (AK2, CXCL11, TYK2, ANGPT4, IL20RA, MET, ENPP6, and CA12) was established. Three genes (AK2, ENPP6, and CA12) were associated with metabolism, while the others were immune-related. Most genes correlated with poor prognosis. Validation in TCGA-PDAC and GSE57495 datasets demonstrated robust performance, with AUC values for 1-, 3-, and 5-year OS exceeding 0.7. The model also effectively predicted responses to adjuvant therapy. This 8-gene signature enhances prognostic accuracy and therapeutic decision-making in PDAC, offering valuable insights for clinical applications and personalized treatment strategies.

Data Driven-Based Machine Learning Modelling and Empirical Correlations for Predicting Snow-Covered Area in the Swat Region, Pakistan

In recent decades, global and regional climate change has emerged as a significant challenge with potential catastrophic consequences, including hurricanes, floods, sea level rise, and temperature shifts. Snow-covered area (SCA) serves as a crucial climatic parameter reflecting climate changes, yet accurately determining SCA proves to be a challenging and time-consuming task. This study aims to develop robust prediction models for SCA by employing three machine learning (ML) approaches using readily available climatic data from Swat, Pakistan, spanning two decades. The climate data encompass precipitation, daily maximum/minimum temperatures, and SCA measurements. Three ML methods—artificial neural networks (ANN), functional networks (FN), and adaptive neuro-fuzzy inference systems (ANFIS)—were employed to model SCA. Accuracy measures, including coefficient of determination (R2), average absolute percentage error (AAPE), and root mean squared error (RMSE) were utilized to evaluate model performance. All three ML models exhibited superior performance, with high R2 values and low errors. Accuracy indicators of the ANN model are better than FN and ANFIS models, yielding the highest R2 (0.956) and minimum RMSE and AAPE values (0.61 and 0.91). ANFIS demonstrated slightly better performance than FN, with RMSE, AAPE, and R2 values of 0.65, 1.1, and 0.950, respectively. FN yielded RMSE, AAPE, and R values of 1.14, 1.72, and 0.85, respectively. Additionally, two empirical correlations were derived from the optimized FN and ANN models for SCA prediction using the same input variables. This study underscores the efficacy of ML techniques in accurately and consistently predicting SCA parameters, offering valuable insights into climate change and its consequences.

Pre-Operative Anemia is an Unsuspecting Driver of Machine Learning Prediction of Adverse Outcomes after Lumbar Spinal Fusion.

Pre-operative risk assessment remains a challenge in spinal fusion operations. Predictive modeling provides data-driven estimates of post-surgical outcomes, guiding clinical decisions and improving patient care. Moreover, automated machine learning models are both effective and user-friendly, allowing healthcare professionals with minimal technical expertise to identify high-risk patients who may need additional pre-operative support. This study investigated the use of automated machine learning models to predict discharge disposition, length of hospital stay, and readmission post-surgery by analyzing pre-operative patient electronic medical record data and identifying key factors influencing adverse outcomes. Retrospective cohort study. The sample includes electronic medical records of 3,006 unique surgical events from 2,855 patients who underwent lumbar spinal fusion surgeries at a single institution. The adverse outcomes assessed were discharge disposition (non-home facility), length of hospital stay (extended stay), and readmission within 90 days post-surgery. We employed several inferential and predictive approaches, including the automated machine learning tool TPOT2 (Tree-based Pipeline Optimization Tool-2). TPOT2, which uses genetic programming to select optimal machine learning pipelines in a process inspired by molecular evolution, constructed, optimized and identified robust predictive models for all outcomes. Feature importance values were derived to identify major pre-operative predictive features driving optimal models. Adverse outcome rates were 25.9% for discharge to non-home facilities, 23.9% for extended hospital stay, and 24.7% for readmission within 90 days post-surgery. TPOT2 delivered the best-performing predictive models, achieving balanced accuracies ((Sensitivity [true positive rate] + Specificity [true negative rate)]) / 2) of 0.72 for discharge disposition, 0.72 for length of stay, and 0.67 for readmission. Notably, preoperative hemoglobin emerged as a consistently strong predictor in best-performing models across outcomes. Patients with severe anemia (hemoglobin <80g/dL) demonstrated higher associations with all adverse outcomes and common comorbidities associated with frailty (e.g., hypertension, type II diabetes, and chronic pain). Additional patient variables and comorbidities, including body mass index, age, and mental health status, influencing post-surgical outcomes were also highly predictive. This study demonstrates the effectiveness of automated machine learning in predicting post-surgical adverse outcomes and identifying key pre-operative predictors associated with such outcomes. While factors like age, BMI, insurance type, and specific comorbidities showed notable effects on outcomes, preoperative hemoglobin consistently emerged as a significant predictor across outcomes, suggesting its critical role in pre-surgical assessment. These findings underscore the potential of enhancing patient care and preoperative assessment through advanced predictive modeling.

Predictive Modelling of Soaked California Bearing Ratio in Nano-Silica Stabilized Sub-grade Soil in Lesser Himalayan Region

This research employs predictive modelling techniques to comprehensively understand the soaked CBR, a key parameter indicating the soil's load-bearing capacity under wet conditions. Experimental investigations are conducted on Nano-Silica (NS) treated sub-grade soil samples collected from the Lesser Himalayan Region. These experiments encompass a range of Nano-Silica concentrations and soaking conditions to capture the diverse environmental factors prevalent in the region. Advanced statistical and machine learning tools are applied to analyze the data and develop predictive models. The objective is to establish correlations between Nano-Silica content, soaking conditions, and the resulting soaked CBR. The models are validated against independent datasets to ensure their reliability and generalizability. The findings of this study hold significant implications for sustainable infrastructure development in the Lesser Himalayan Region. A robust predictive model for soaked CBR in Nano-Silica stabilized sub-grade soil not only contributes to the understanding of soil stabilization mechanisms but also aids in optimizing construction practices for enhanced road performance in challenging terrains. This research bridges the gap between traditional geotechnical engineering and emerging nanotechnology applications in soil stabilization, paving the way for resilient and durable infrastructure in geologically complex regions.

A novel data-driven bridge monitoring data prediction framework using improved intelligent optimization-assisted deep learning

Bridge monitoring data prediction plays a crucial role in maintaining the serviceability and safety of bridges. To assess the performance of bridge structures in advance, an accurate and robust prediction model for measurement data prediction is needed. In this context, this paper presents a novel hybrid prediction model called IGWO-VMD-IGWO-LSTM, which aims to improve the accuracy and generalization ability of bridge monitoring data prediction. First, the proposed model incorporates variational mode decomposition (VMD) and an improved gray wolf optimization (IGWO) algorithm to decompose the measured data into several intrinsic mode functions (IMFs) to capture the signal’s features comprehensively. Then, long short-term memory (LSTM) models are established individually for each IMF, with hyperparameters optimization using the IGWO algorithm. Finally, the integrated reconstructed subsequences yield the prediction results of the proposed IGWO-VMD-IGWO-LSTM model. Thus, the proposed framework avoids exploring the complex internal mechanism of bridge behavior evolution. The measurement data collected from a real bridge demonstrates the feasibility of the proposed prediction method, and different deep learning models are used for comparison. The results demonstrate a significant improvement in prediction performance comparison with the other concerned models, highlighting the strong generalization ability and robustness of the proposed IGWO-VMD-IGWO-LSTM model.