Development Of Predictive Models Research Articles

Development of Prediction Model for Phosphorus Removal During LD/BOF Steelmaking Using Machine Learning Algorithms

Development of Prediction Model for Phosphorus Removal During LD/BOF Steelmaking Using Machine Learning Algorithms

Development of an Ensemble Model for Power Demand Prediction in High-Rise Hotels with Environmental Variables

Development of an Ensemble Model for Power Demand Prediction in High-Rise Hotels with Environmental Variables

Artificial intelligence, medications, pharmacogenomics, and ethics.

Artificial Intelligence (AI) and Machine Learning (ML) are revolutionizing various scientific and clinical disciplines including pharmacogenomics (PGx) by enabling the analysis of complex datasets and the development of predictive models. The integration of AI and ML with PGx has the potential to provide more precise, data-driven insights into new drug targets, drug efficacy, drug selection, and risk of adverse events. While significant effort to develop and validate these tools remain, ongoing advancements in AI technologies, coupled with improvements in data quality and depth is anticipated to drive the transition of these tools into clinical practice and delivery of individualized treatments and improved patient outcomes. The successful development and integration of AI-assisted PGx tools will require careful consideration of ethical, legal, and social issues (ELSI) in research and clinical practice. This paper explores the intersection of PGx with AI, highlighting current research and potential clinical applications, and ELSI including privacy, oversight, patient and provider knowledge and acceptance, and the impact on patient-provider relationship and new roles.

Predicting neurodevelopmental disorders using machine learning models and electronic health records - status of the field.

Machine learning (ML) is increasingly used to identify patterns that could predict neurodevelopmental disorders (NDDs), such as autism spectrum disorder (ASD) and attention-deficit hyperactivity disorder (ADHD). One key source of multilevel data for ML prediction models includes population-based registers and electronic health records. These can contain rich information on individual and familial medical histories and socio-demographics. This review summarizes studies published between 2010-2022 that used ML algorithms to develop predictive models for NDDs using population-based registers and electronic health records. A literature search identified 1191 articles, of which 32 were retained. Of these, 47% developed ASD prediction models and 25% ADHD models. Classical ML methods were used in 82% of studies and in particular tree-based prediction models performed well. The sensitivity of the models was lower than 75% for most studies, while the area under the curve (AUC) was greater than 75%. The most important predictors were patient and familial medical history and sociodemographic factors. Using private in-house datasets makes comparing and validating model generalizability across studies difficult. The ML model development and reporting guidelines were adopted only in a few recently reported studies. More work is needed to harness the power of data for detecting NDDs early.

Development and validation of a risk prediction model for cognitive impairment in breast cancer patients.

Breast cancer patients often experience cognitive impairment as a complication during treatment, which seriously affects their quality of life. This study aimed to assess the risk factors associated with cognitive impairment in breast cancer patients and to construct and validate a nomogram model to predict cognitive impairment in this population. In this study, we used a convenience sampling method to select 423 breast cancer patients who attended the Department of Breast Surgery at the First Hospital of Jinzhou Medical University from September 2023 to March 2024. We analyzed these patients' cognitive impairment risk factors through LASSO regression and logistic regression analysis to develop a predictive model. The model was evaluated using the area under the curve (AUC) from the receiver operating characteristic (ROC) curve and the calibration curve and decision curve analysis. This study found a prevalence of cognitive impairment of 19.62% among breast cancer patients. A nomogram model was developed based on six influencing factors: age, educational level, pathological type, treatment program, emotional state, and fatigue. The area under the curve (AUC) for the model's training and validation groups was 0.944 and 0.931, respectively. The model calibration curves showed a high degree of consistency, and the decision curve analysis (DCA) indicated good clinical applicability of the model. This nomogram demonstrates good discrimination, calibration, and clinical applicability, making it a more intuitive predictor of the risk of cognitive impairment in breast cancer patients.

Prediction models for ischemic stroke and bleeding in dialysis patients: a systematic review and meta-analysis

Abstract Background Patients with kidney failure on maintenance dialysis have a high stroke and bleeding risk. Multivariable prediction models can be used to estimate the risk of ischemic stroke and bleeding. A systematic review and meta-analysis was performed to determine the performance of the existing models in patients on dialysis. Methods MEDLINE and Embase databases were searched, from inception through January 12, 2024, for studies of prediction models for stroke or bleeding, derived or validated in dialysis cohorts. Discrimination measures for models with c-statistic data from ≥ 3 cohorts were pooled by random effects meta-analysis and a 95% prediction interval (PI) was calculated. Risk of bias was assessed using PROBAST. The review was conducted according to the PRISMA statement and the CHARMS checklist. Results Eight studies were included in this systematic review. All the included studies validated pre-existing models that were derived in cohorts from the general population. None of the identified studies reported the development of a new dialysis specific prediction model for stroke, while dialysis specific risk scores for bleeding were proposed by two studies. In meta-analysis of c-statistics, the CHA2DS2-VASc, CHADS2, ATRIA, HEMORR(2)HAGES and HAS-BLED scores showed very poor discriminative ability in the dialysis population. Six of the eight included studies were at low or unclear risk of bias and certainty of evidence was moderate. Conclusions The existing prediction models for stroke and bleeding have very poor performance in the dialysis population. New dialysis-specific risk scores should be developed to guide clinical decision making in these patients.

Prediction of spirometry outcome in Croatian patients with chronic obstructive pulmonary disease

The current study offers an extensive examination of the influence of 29 diverse parameters on spirometry measurement variables in a cohort of 534 patients with chronic obstructive pulmonary disease (COPD) from five different centers in Croatia. The study elucidates both the magnitude and direction of the effect exerted by the 29 predictors on forced vital capacity (FVC), forced expiratory volume in one second (FEV1), the ratio FEV1/FVC, and predicted forced expiratory flow at 50% of FVC. Additionally, the development of prediction models for these parameters has been undertaken using several statistical methods. The study identifies fat-free mass index, 6-minute walk distance, predicted diffusing capacity of the lung for carbon monoxide, arterial partial pressure of oxygen, and both arterial and tissue hemoglobin oxygen saturation percentage as robust positive predictors for all four spirometry parameters. Body mass index is recognized as a weak positive predictor for FEV1 and FEV1/FVC, commonly observed in COPD patients. As expected, smoking years is identified as a strong negative predictor for all four spirometry parameters, while age and illness duration exhibit strong predictive negative associations. Furthermore, modified medical research council, arterial partial pressure carbon dioxide, St George's respiratory questionnaire, COPD assessment test, depression anxiety stress scales, and nutritional risk screening are identified as weak negative predictors. Charlson comorbidity index, phase angle, and number of comorbidities do not exhibit a significant impact on spirometry variables. Ultimately, the performed factorial analysis categorized the 29 parameters into five groups, which were identified as relating to lung function, health status, nutritional status, age, and smoking. Multiple regression analysis, including four newly derived parameters based on the results of factorial analysis, identified nutritional status as a positive predictor for spirometry readings, while smoking, poor health status, and age were identified as negative predictors in successive order.

Filtration Behaviour of Ostrea edulis: Diurnal Rhythmicity Influenced by Light Cycles, Body Size and Water Temperature

AbstractDespite the existing studies on oyster physiology, it is a concern that, by ignoring the diurnal rhythmicity of feeding activities, the current knowledge on oyster filtration might be unrepresentative of their real filtration capacity. In the present study, the diurnal and nocturnal fluctuations of Ostrea edulis clearance rate were investigated for the first time under laboratory conditions, also considering the combined effects of water temperature and oyster size. A positive correlation was found between the mean clearance rate (CRm) and body size for both length and wet weight, with an average of 0.5 L h−1 and 0.1 L h−1 of water filtered by large and small adult oysters respectively during the 24-h experiment. A positive significant correlation between mean clearance rate and water temperature was found only for large oysters, suggesting further studies utilising temperature ranges closer to O. edulis thermal optimum are required to clarify the influence of temperature on oyster filtration activities across different size classes. The results of this study also showed a discontinuous ingestion of food by O. edulis throughout 24-h intervals, with higher values of clearance rate observed during hours of darkness at most temperatures for both size classes. This suggests photoperiod may have a potential influence on O. edulis filtration. Providing a better understanding of O. edulis filtration behaviour, these findings have implications for ecological restoration, aquaculture, and comparative studies on oyster bioenergetics and physiology, with direct applications to the development of predictive models and the assessment of ecosystem services provided by O. edulis habitat.

Early prediction of acute-on-chronic liver failure development in patients with diverse chronic liver diseases.

Acute-on-chronic liver failure (ACLF) is a syndrome characterized by the acute decompensation of chronic liver disease, resulting in organ failure and high short-term mortality. The progression of ACLF is dynamic and reversible in a considerable proportion of patients during hospitalization. Early detection and accurate assessment of ACLF are essential; however, ideal methods for this purpose are still lacking. Therefore, this study aimed to develop a new score for predicting the onset of ACLF in patients with various chronic liver diseases.A total of 6,188 patients with various chronic liver diseases were included in the study. Clinical and laboratory data were collected, and the occurrence of ACLF within 28 days was recorded. The Lasso-Cox regression method was employed to develop prediction models for the onset of ACLF at 7, 14, and 28 days.Among 5,221 patients without ACLF, 477 progressed to ACLF within 28 days. Seven predictors were identified as significantly associated with the occurrence of ACLF at 7, 14, and 28 days. A new scoring system was developed as follows: [NEUT ≥ 7, 109/L; 1 or 0] × 0.49 + [PLT < 100, 109/L; 1 or 0] × 0.44 + [TBIL ≥ 35, µmol/L; 1 or 0] × 0.05 + [HDL-C < 0.5, mmol/L; 1 or 0] × 1.04 - Ln[Hb, g/L] × 0.89 + [BUN > 7, mmol/L; 1 or 0] × 0.51 + Ln[INR] × 0.87 + 3.40. This new score demonstrated superior discrimination, with the C-indexes of 0.958, 0.944, and 0.938 at 7, 14, and 28 days, respectively, outperforming those of four other scores (CLIF-C-ACLF-Ds, MELD, MELD-Na, and CLIF-C-ADs score; all P < 0.001). Additionally, the new score improved in predictive accuracy, time-dependent receiver operating characteristics, probability density function evaluations, and calibration curves, making it highly predictive for the onset of ACLF at all time points. The optimal cut-off value of 9.6 effectively distinguished between high- and low-risk patients for ACLF onset. These findings were further validated in a separate cohort of patients. A new progressive score, based on seven predictors, has been developed to accurately forecast the occurrence of ACLF within 7, 14, and 28 days in patients with various chronic liver diseases. This tool may be utilized to identify high-risk patients, tailor follow-up management, and guide the escalation of care, prognostication, and transplant evaluation.

Evaluating the use of in-season measures of pest abundance to predict end-of-season damage: a study in commercial almond (Prunus dulcis).

Pre-harvest pest management tools are essential to minimizing crop loss. The development of predictive models using early warning signs of pest abundance to predict imminent crop loss can guide management decisions and enable targeted, well-calibrated intervention. With sufficient data, in-season measures of pest abundance can be an important factor in generating accurate predictions of damage. However, sampling plans for tracking insect phenology and those designed to guide informed pest-management may not be equivalent. Using data from a commercial almond orchard setting, we evaluated five different lure-trap combinations used for monitoring navel orangeworm (Amyelois transitella; a primary pest of almond) under different management conditions. Using these data, we developed a predictive model of almond damage and evaluated the contribution of in-season measures of pest abundance towards accurate predictions of end-of-season damage. Mating disruption in orchard management influenced the efficacy of multiple lure-trap combinations. Despite this effect, there was as strong correlation between measured pest abundance and damage across multiple trap types regardless of management method. While statistically significant, measures of pest abundance did not enhance the accuracy of predictive models and characteristics of almond variety were stronger predictors of damage at harvest. Repeated significant correlations between navel orangeworm abundance and almond damage across multiple trap types justifies continued exploration of predictive modeling as a management tool. Lure-trap combinations exhibiting resistance to external factors are clear candidates for further development. However, generating accurate predictions of damage using these data will likely require calibration or modification of current sampling protocols. © 2024 Society of Chemical Industry. This article has been contributed to by U.S. Government employees and their work is in the public domain in the USA.

Development and validation of a risk prediction model for severe postoperative complications in elderly patients with hip fracture.

This study aimed to investigate risk factors associated with severe postoperative complications following hip fracture surgery in elderly patients and to develop a nomogram-based risk prediction model for these complications. A total of 627 elderly patients with hip fractures treated at Yongchuan Hospital of Chongqing Medical University from January 2015 to April 2024 were collected. 439 patients were assigned to the training cohort for model development, and 188 to the validation cohort for model assessment. The training cohort was stratified based on the presence or absence of severe complications. We employed LASSO regression, as well as univariate and multivariate logistic regression analyses, to identify significant factors. A nomogram was constructed based on the outcomes of the multivariate regression. The model's discriminative ability was assessed using the area under the receiver operating characteristic curve (AUC), while calibration plots and decision curve analysis (DCA) evaluated its calibration and stability. Internal validation was performed using the validation cohort. Out of the 627 patients, 118 (18.82%) experienced severe postoperative complications. Both LASSO regression and multivariate logistic analysis identified the modified 5-item frailty index (mFI-5) and the preoperative C-reactive protein to albumin ratio (CAR) as significant predictors of severe complications. The nomogram model, derived from the multivariate analysis, exhibited strong discriminative ability, with an AUC of 0.963 (95% CI: 0.946-0.980) for the training cohort and 0.963 (95% CI: 0.938-0.988) for the validation cohort. Calibration plots demonstrated excellent agreement between the nomogram's predictions and actual outcomes. Decision curve analysis (DCA) indicated that the model provided clinical utility across all patient scenarios. These findings were consistent in the validation cohort. Both the mFI-5 and CAR are predictive factors for severe postoperative complications in elderly patients undergoing hip fracture surgery.

Scaling Correlation Analysis of Particulate Matter Concentrations of Three South Indian Cities

Analyzing the fluctuations of particulate matter (PM) concentrations and their scaling correlation structures are useful for air quality management. Multifractal characterization of PM2.5 and PM10 of three cities in India wase considered using the detrended fluctuation procedure from 2018 to 2021. The cross-correlation of PM concentration in a multifractal viewpoint using the multifractal cross-correlation analysis (MFCCA) framework is proposed in this study. It was observed that PM2.5 was more multifractal and complex than PM10 at all the locations. The PM–gaseous pollutant (GP) and PM–meteorological variable (MV) correlations across the scales were found to be weak to moderate in different cities. There was no definite pattern in the correlation of PM with different meteorological and gaseous pollutants variables. The nature of correlation in the pairwise associations was found to be of diverse and mixed nature across the time scales and locations. All the time series exhibited multifractality when analyzed pairwise using multifractal cross-correlation analysis. However, there was a reduction in multifractality in individual cases during PM–GP and PM–MV paired analyses. The insights gained into the scaling behavior and cross-correlation structure from this study are valuable for developing prediction models for PMs by integrating them with machine learning techniques.

Weather-aware energy management for unmannedaerial vehicles: a machine learning application with global data integration

This study introduces a machine learning (ML) framework to predict unmanned aerial vehicle (UAV) energy requirements under diverse environmental conditions. The framework correlates UAV flight patterns with publicly accessible weather data, to yield an energy management tool applicable to a wide range of UAV configurations. The model employs the Cross-industry standard process for data mining and advanced feature engineering, offering an in-depth analysis of meteorological factors and UAV energy demands. The study assesses several multi-regression linear and ML models, whereby ensemble models gradient boosting (GB) and eXtreme gradient boosting demonstrate superior performance and accuracy. Specifically, the GB model achieved a test mean absolute error (MAE) of 0.0395 V (V) for voltage, 0.808 A (A) for current, and 9.758 mA-hours (mAh) for discharge, with prediction accuracy of over 99.9% for voltage and discharge, and 97% for current, derived from the coefficient of determination (R2). A novel integration of real-world UAV logs and weather data underpins the development of a weather-aware ML prediction model for UAV energy consumption. Our framework is capable of concurrently predicting three components of energy and power with almost uniform accuracy, a feature not found in contemporary models. Empirical test flights show a discrepancy of only 0.005 W-hour (Wh) between total predicted and actual energy consumption. This work enhances both efficiency and safety in UAV operations. The resulting energy-predictive flight planning tool sets a new benchmark for artificial intelligence (AI) applications in intelligent automation for UAVs.

Exploring Factors Influencing Patient Delay Behavior in Oral Cancer: The Development of a Risk Prediction Model in Western China

Background and Aims: To study the unknown influencing factors of delayed medical treatment behavior in oral cancer patients in western China and to develop a prediction model on the risk of delayed medical treatment in oral cancer patients. Method: We investigated oral cancer patients attending a tertiary Grade A dental hospital in western China from June 2022 to July 2023. The logistic regression and four machine learning models (nearest neighbors, the RBF SVM, random forest, and QDA) were used to identify risk factors and establish a risk prediction model. We used the established model to predict the data before and after the COVID-19 pandemic and test whether the prediction effect can still remain stable and accurate under the interference of COVID-19. Result: Out of the 495 patients included in the study, 122 patients (58.65%) delayed seeking medical treatment before the lifting of the restrictions of the pandemic, while 153 patients (53.13%) did so after the lifting of restrictions. The logistic regression model revealed that living with adult children was a protective factor for patients in delaying seeking medical attention, regardless of the implementation of pandemic control measures. After comparing each model, it was found that the statistical indicators of the random forest algorithm such as the AUC score (0.8380) and specificity (0.8077) ranked first, with the best prediction performance and stable performance. Conclusions: This study systematically elucidates the critical factors influencing patient delay behavior in oral cancer diagnosis and treatment, employing a comprehensive risk prediction model that accurately identifies individuals at an elevated risk of delay. It represents a pioneering large-scale investigation conducted in western China, focusing explicitly on the multifaceted factors affecting the delayed medical treatment behavior of oral cancer patients. The findings underscore the imperative of implementing early intervention strategies tailored to mitigate these delays. Furthermore, this study emphasizes the pivotal role of robust social support systems and positive family dynamics in facilitating timely access to healthcare services for oral cancer patients, thereby potentially improving outcomes and survival rates.

Quantitative Analysis of Predictors of Acoustic Materials for Noise Reduction as Sustainable Strategies for Materials in the Automotive Industry

This study proposes a qualitative analysis for identifying the best predictors for ensuring passive noise control, aiming to achieve superior acoustic comfort in transportation systems. The study is based on real experimental data, collected through acoustic measurements performed by the authors on materials from six different classes and employs a multidisciplinary approach, including Mann–Whitney U tests, Kruskal–Wallis analysis with Dunn’s post hoc multiple comparisons and multilinear regression. This research presents an analysis and evaluation of how the physical properties of various materials influence acoustic comfort, acoustic absorption class and absorption class performance and proposes quantitative models for material selection to address sustainable strategies in the automotive industry. The results highlight significant differences between material categories in terms of acoustic absorption properties and demonstrate the importance of rigorous material selection in vehicle design to enhance acoustic comfort. Additionally, the research contributes to the development of predictive models that estimate acoustic performance based on the physical properties of materials, providing a basis for optimizing material selection in the design phase.

Research progress on the development of myopia prediction models and their predictive performance

The incidence of myopia is continuously increasing worldwide, especially in China, where the prevention and control of myopia is of great significance. Accurate prediction of myopia and early intervention are of great importance for slowing down the development of myopia and reducing the social burden. This article reviews the research progress of myopia prediction models and their predictive performance in recent years. It includes models based on refractive power and ocular biological characteristics, such as those using parameters such as refractive power, axial length, and percentile curves for prediction; models combined with environmental factors, which explore the relationship between myopia in school-age children and the age of onset, parents' myopia, education, and various living environmental factors; models based on genetic factors, which predict by evaluating parents' myopia status and single nucleotide polymorphisms discovered in genome-wide association studies. In addition, the article also introduces the application of artificial intelligence in myopia prediction, such as machine learning and deep learning algorithms that can predict changes in axial length growth and refractive power. These research advances provide a reference for establishing more accurate myopia prediction models and help achieve precise and personalized myopia prevention and control.

Contributions to the development of prediction models for the toxicity of ionic liquids

Contributions to the development of prediction models for the toxicity of ionic liquids

Risk factors and predictive model development for high blood loss in minimally invasive distal pancreatectomy: a retrospective cohort study.

High blood loss is an adverse event related to increased morbidity and poorer outcomes in pancreatic surgery patients. The aim of this study was to identify risk factors and establish a predictive model for high perioperative blood loss (HPBL) in minimally invasive distal pancreatectomy (MIDP). We collected data from 353 patients who underwent MIDP at a university affiliated tertiary hospital between January 2016 and October 2023. Perioperative blood loss was calculated based on pre- and postoperative hemoglobin concentrations according to a combination of the formulas provided by Nadler and Gross. Multivariate logistic regression analyses were performed for the training cohort to identify the clinical factors independently associated with perioperative blood loss (PBL). A predictive nomogram based on these factors was established and validated. Weight, imaging findings, serum albumin concentration, MIDP experience, spleen treatment, and operation time were independent predictors for HPBL. The established model for predicting HPBL showed that the area under the curve (AUC) was 0.799 (95% CI = 0.746-0.853) and 0.852 (95% CI = 0.760-0.943) for the training cohort and validation cohort, respectively. When utilized to predict blood transfusion, the AUC was 0.778 (95% CI = 0.691-0.865) in the training cohort and 0.818 (95% CI = 0.681-0.955) in the validation cohort. Patients with a high predicted risk had significantly higher incidences of postoperative pancreatic fistula, intra-abdominal infection, and longer hospital stays than patients with a low risk. We established and validated a model for predicting HPBL in MIDP patients. This novel model may have future utility when generating surgical strategies.

Broaching Digital Twin to Predict Forces, Local Overloads, and Surface Topography Irregularities

Broaching is a key manufacturing process that directly influences the surface integrity of critical components, impacting their functional performance in sectors such as aeronautics, automotive, and energy. Such components are subjected to severe conditions, including high thermomechanical loads, fatigue, and corrosion. For this reason, the development of predictive models is essential for determining the optimal tool design and machining conditions to ensure proper in-service performance. This study, therefore, presents a broaching digital twin based on hybrid modelling, which combines analytical, numerical, and empirical approaches to provide rapid and accurate predictions of the forces per tooth, local overloads, and surface topography irregularities. The digital twin was validated with a critical industrial case study involving fir-tree broaching of turbine discs made of forged and age-hardened Inconel 718. The accuracy of the digital twin was demonstrated by the results: the average error in force predictions was below 10%, and the model effectively identified the most critical teeth and zones prone to failure. It also predicted surface topography irregularities with an error of less than 15%. Interestingly, the relationship between surface topography irregularities and surface residual stress variations across the machined surface was observed experimentally for the first time.

Determinant analysis of oil agglomeration of coal fines using multiple regression and neural network models

ABSTRACT Oil agglomeration has gained importance owing to its simplicity and efficiency in beneficiating coal fines. In this study, high ash coal samples taken from six mines in India were put through an oil agglomeration process to investigate the usage of castor oil or blend of castor and turpentine for recovery of coal fines from coal-water slurry. The performance of the process has been evaluated based on ash rejection [AR (%)] and yield (%). Various Statistical analyses were carried out to investigate the role of various process parameters, such as PD, OD, AT, and oil-type on AR (%) and yield (%). Step-wise regression was performed for development of prediction model for AR (%) and yield (%). Determinant analysis was performed using general linear model (GLM). Our findings indicate that pulp density was the strongest determinant for AR (%), followed by oil dosage and agitation time. Similarly, oil dosage was the primary determinant for yield (%), followed by pulp density and agitation time. Sensitivity analysis was also carried out using artificial neural network (ANN) and the results in respect AR (%) revealed that agitation time was the most important predictor followed by pulp density and oil dosage.