Development Of Predictive Models Research Articles

MIXED DENTITION AMONG PATIENTS ATTENDING DENTAL OPDS IN TERTIARY CARE HOSPITALS IN RURAL SINDH

Objective:To evaluate the applicability of Moyers and Tanaka-Johnston mixed dentition prediction methods among patients attending dental outpatient departments in tertiary care hospitals in rural Sindh, Pakistan. Methods:A cross-sectional study was conducted over six months, involving 300 patients (150 males and 150 females) aged 8-12 years from Bhitai Dental and Medical College, Mirpurkhas. Patients selected had all sound erupted permanent teeth except second and third molars, with no history of orthodontic treatment. Mesiodistal widths of teeth were measured using Mitutoyo digimatic calipers. Data was analyzed using SPSS-16, employing Student’s t-test, paired t-tests, and simple linear regression to derive new prediction equations. Results:Significant discrepancies were found between the predicted and actual widths of unerupted canines and premolars using Moyers and Tanaka-Johnston methods, indicating these methods are unsuitable for this population. New regression equations were derived for the local demographic. Conclusion:The Moyers and Tanaka-Johnston prediction methods are ineffective for the population in rural Sindh. The study recommends the development of localized prediction models and further multicenter studies with larger, diverse samples to refine these equations. These findings are crucial for accurate orthodontic treatment planning and improving patient outcomes.

Development and validation of prediction models for nosocomial infection and prognosis in hospitalized patients with cirrhosis

BackgroundNosocomial infections (NIs) frequently occur and adversely impact prognosis for hospitalized patients with cirrhosis. This study aims to develop and validate two machine learning models for NIs and in-hospital mortality risk prediction.MethodsThe Prediction of Nosocomial Infection and Prognosis in Cirrhotic patients (PIPC) study included hospitalized patients with cirrhosis at the Qingchun Campus of the First Affiliated Hospital of Zhejiang University. We then assessed several machine learning algorithms to construct predictive models for NIs and prognosis. We validated the best-performing models with bootstrapping techniques and an external validation dataset. The accuracy of the predictions was evaluated through sensitivity, specificity, predictive values, and likelihood ratios, while predictive robustness was examined through subgroup analyses and comparisons between models.ResultsWe enrolled 1,297 patients into derivation cohort and 496 patients into external validation cohort. Among the six algorithms assessed, the Random Forest algorithm performed best. For NIs, the PIPC-NI model achieved an area under the curve (AUC) of 0.784 (95% confidence interval [CI] 0.741–0.826), a sensitivity of 0.712, and a specificity of 0.702. For in-hospital mortality, the PIPC- mortality model achieved an AUC of 0.793 (95% CI 0.749–0.836), a sensitivity of 0.769, and a specificity of 0.701. Moreover, our PIPC models demonstrated superior predictive performance compared to the existing MELD, MELD-Na, and Child-Pugh scores.ConclusionsThe PIPC models showed good predictive power and may facilitate healthcare providers in easily assessing the risk of NIs and prognosis among hospitalized patients with cirrhosis.

Evaluating the significance of ECSCR in the diagnosis of ulcerative colitis and drug efficacy assessment.

The main challenge in diagnosing and treating ulcerative colitis (UC) has prompted this study to discover useful biomarkers and understand the underlying molecular mechanisms. In this study, transcriptomic data from intestinal mucosal biopsies underwent Robust Rank Aggregation (RRA) analysis to identify differential genes. These genes intersected with UC key genes from Weighted Gene Co-expression Network Analysis (WGCNA). Machine learning identified UC signature genes, aiding predictive model development. Validation involved external data for diagnostic, progression, and drug efficacy assessment, along with ELISA testing of clinical serum samples. RRA integrative analysis identified 251 up-regulated and 211 down-regulated DEGs intersecting with key UC genes in WGCNA, yielding 212 key DEGs. Subsequently, five UC signature biomarkers were identified by machine learning based on the key DEGs-THY1, SLC6A14, ECSCR, FAP, and GPR109B. A logistic regression model incorporating these five genes was constructed. The AUC values for the model set and internal validation data were 0.995 and 0.959, respectively. Mechanistically, activation of the IL-17 signaling pathway, TNF signaling pathway, PI3K-Akt signaling pathway in UC was indicated by KEGG and GSVA analyses, which were positively correlated with the signature biomarkers. Additionally, the expression of the signature biomarkers was strongly correlated with various UC types and drug efficacy in different datasets. Notably, ECSCR was found to be upregulated in UC serum and exhibited a positive correlation with neutrophil levels in UC patients. THY1, SLC6A14, ECSCR, FAP, and GPR109B can serve as potential biomarkers of UC and are closely related to signaling pathways associated with UC progression. The discovery of these markers provides valuable information for understanding the molecular mechanisms of UC.

Systematic review of methods used in prediction models with recurrent event data

BackgroundPatients who suffer from chronic conditions or diseases are susceptible to experiencing repeated events of the same type (e.g. seizures), termed ‘recurrent events’. Prediction models can be used to predict the risk of recurrence so that intervention or management can be tailored accordingly, but statistical methodology can vary. The objective of this systematic review was to identify and describe statistical approaches that have been applied for the development and validation of multivariable prediction models with recurrent event data. A secondary objective was to informally assess the characteristics and quality of analysis approaches used in the development and validation of prediction models of recurrent event data.MethodsSearches were run in MEDLINE using a search strategy in 2019 which included index terms and phrases related to recurrent events and prediction models. For studies to be included in the review they must have developed or validated a multivariable clinical prediction model for recurrent event outcome data, specifically modelling the recurrent events and the timing between them.The statistical analysis methods used to analyse the recurrent event data in the clinical prediction model were extracted to answer the primary aim of the systematic review. In addition, items such as the event rate as well as any discrimination and calibration statistics that were used to assess the model performance were extracted for the secondary aim of the review.ResultsA total of 855 publications were identified using the developed search strategy and 301 of these are included in our systematic review. The Andersen-Gill method was identified as the most commonly applied method in the analysis of recurrent events, which was used in 152 (50.5%) studies. This was closely followed by frailty models which were used in 116 (38.5%) included studies. Of the 301 included studies, only 75 (24.9%) internally validated their model(s) and three (1.0%) validated their model(s) in an external dataset.ConclusionsThis review identified a variety of methods which are used in practice when developing or validating prediction models for recurrent events. The variability of the approaches identified is cause for concern as it indicates possible immaturity in the field and highlights the need for more methodological research to bring greater consistency in approach of recurrent event analysis. Further work is required to ensure publications report all required information and use robust statistical methods for model development and validation.PROSPERO registrationCRD42019116031.

JmBIG: enhancing dynamic risk prediction and personalized medicine through joint modeling of longitudinal and survival data in big routinely collected data

We have introduced the R package jmBIG to facilitate the analysis of large healthcare datasets and the development of predictive models. This package provides a comprehensive set of tools and functions specifically designed for the joint modelling of longitudinal and survival data in the context of big data analytics. The jmBIG package offers efficient and scalable implementations of joint modelling algorithms, allowing for integrating large-scale healthcare datasets.By utilizing the capabilities of jmBIG, researchers and analysts can effectively handle the challenges associated with big healthcare data, such as high dimensionality and complex relationships between multiple outcomes.With the support of jmBIG, analysts can seamlessly fit Bayesian joint models, generate predictions, and evaluate the performance of the models. The package incorporates cutting-edge methodologies and harnesses the computational capabilities of parallel computing to accelerate the analysis of large-scale healthcare datasets significantly. In summary, jmBIG empowers researchers to gain deeper insights into disease progression and treatment response, fostering evidence-based decision-making and paving the way for personalized healthcare interventions that can positively impact patient outcomes on a larger scale.

Development and Validation of Prediction Models for Incident Reversible Cognitive Frailty Based on Social-Ecological Predictors Using Generalized Linear Mixed Model and Machine Learning Algorithms: A Prospective Cohort Study.

This study aimed to develop and validate prediction models for incident reversible cognitive frailty (RCF) based on social-ecological predictors. Older adults aged ≥60 years from China Health and Retirement Longitudinal Study (CHARLS) 2011-2013 survey were included as training set (n = 1230). The generalized linear mixed model (GLMM), eXtreme Gradient Boosting, support vector machine, random forest, and Binary Mixed Model forest were used to develop prediction models. All models were evaluated internally with 5-fold cross-validation and evaluated externally via CHARLS 2013-2015 survey (n = 1631). Only GLMM showed good discrimination (AUC = 0.765, 95% CI = 0.736, 0.795) in training set, and all models showed fair discrimination (AUC = 0.578-0.667, 95% CI = 0.545, 0.725) in internal and external validation. All models showed acceptable calibration, overall prediction performance, and clinical usefulness in training and validation sets. Older adults were divided into three groups using risk score based on GLMM, which could assist healthcare providers to predict incident RCF, facilitating early identification of high-risk population.

Development and validation of novel interpretable survival prediction models based on drug exposures for severe heart failure during vulnerable period

BackgroundSevere heart failure (HF) has a higher mortality during vulnerable period while targeted predictive tools, especially based on drug exposures, to accurately assess its prognoses remain largely unexplored. Therefore, this study aimed to utilize drug information as the main predictor to develop and validate survival models for severe HF patients during this period.MethodsWe extracted severe HF patients from the MIMIC-IV database (as training and internal validation cohorts) as well as from the MIMIC-III database and local hospital (as external validation cohorts). Three algorithms, including Cox proportional hazards model (CoxPH), random survival forest (RSF), and deep learning survival prediction (DeepSurv), were applied to incorporate the parameters (partial hospitalization information and exposure durations of drugs) for constructing survival prediction models. The model performance was assessed mainly using area under the receiver operator characteristic curve (AUC), brier score (BS), and decision curve analysis (DCA). The model interpretability was determined by the permutation importance and Shapley additive explanations values.ResultsA total of 11,590 patients were included in this study. Among the 3 models, the CoxPH model ultimately included 10 variables, while RSF and DeepSurv models incorporated 24 variables, respectively. All of the 3 models achieved respectable performance metrics while the DeepSurv model exhibited the highest AUC values and relatively lower BS among these models. The DCA also verified that the DeepSurv model had the best clinical practicality.ConclusionsThe survival prediction tools established in this study can be applied to severe HF patients during vulnerable period by mainly inputting drug treatment duration, thus contributing to optimal clinical decisions prospectively.

Rapid In Situ Near-Infrared Assessment of Tetrahydrocannabinolic Acid in Cannabis Inflorescences before Harvest Using Machine Learning.

Cannabis is cultivated for therapeutic and recreational purposes where delta-9 tetrahydrocannabinol (THC) is a main target for its therapeutic effects. As the global cannabis industry and research into cannabinoids expands, more efficient and cost-effective analysis methods for determining cannabinoid concentrations will be beneficial to increase efficiencies and maximize productivity. The utilization of machine learning tools to develop near-infrared (NIR) spectroscopy-based prediction models, which have been validated from accurate and sensitive chemical analysis, such as gas chromatography (GC) or liquid chromatography mass spectroscopy (LCMS), is essential. Previous research on cannabinoid prediction models targeted decarboxylated cannabinoids, such as THC, rather than the naturally occurring precursor, tetrahydrocannabinolic acid (THCA), and utilize finely ground cannabis inflorescence. The current study focuses on building prediction models for THCA concentrations in whole cannabis inflorescences prior to harvest, by employing non-destructive screening techniques so cultivators may rapidly characterize high-performing cultivars for chemotype in real time, thus facilitating targeted optimization of crossbreeding efforts. Using NIR spectroscopy and LCMS to create prediction models we can differentiate between high-THCA and even ratio classes with 100% prediction accuracy. We have also developed prediction models for THCA concentration with a R2 = 0.78 with a prediction error average of 13%. This study demonstrates the viability of a portable handheld NIR device to predict THCA concentrations on whole cannabis samples before harvest, allowing the evaluation of cannabinoid profiles to be made earlier, therefore increasing high-throughput and rapid capabilities.

Risk factor and prediction model development for severe radiation-induced oral mucositis in head and neck tumors.

Aim: This article aims to identify risk factors for severe radiation-induced oral mucositis (RIOM) in head and neck cancer (HNC) patients. In addition, we intend to establish a predictive model in patients undergoing intensity-modulated radiotherapy. Patients & methods: In this retrospective study, several HNC patients (n=179) treated at Zhejiang Provincial People's Hospital from January 2019 to June 2023 were considered. The recruited subjects were divided into modeling and validation groups. The experimental data on clinical characteristics and treatment were collected and analyzed to identify predictive factors for severe RIOM based on the logistic regression approach. Results: The results indicated that severe RIOM occurred in 55.3% of patients. Accordingly, significant predictors included smoking history, diabetes, concurrent chemotherapy, cumulative radiation dose and weight loss of ≥5% in relative to admission weight. A nomogram based on these factors was validated, showing excellent predictive accuracy. Conclusion: In summary, the predictive model could effectively identify high-risk patients for severe RIOM, enabling the design of targeted interventions and improving patient management during radiotherapy.

Insights into gemcitabine resistance in pancreatic cancer: association with metabolic reprogramming and TP53 pathogenicity in patient derived xenografts

BackgroundWith poor prognosis and high mortality, pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal malignancies. Standard of care therapies for PDAC have included gemcitabine for the past three decades, although resistance often develops within weeks of chemotherapy initiation through an array of possible mechanisms.MethodsWe reanalyzed publicly available RNA-seq gene expression profiles of 28 PDAC patient-derived xenograft (PDX) models before and after a 21-day gemcitabine treatment using our validated analysis pipeline to identify molecular markers of intrinsic and acquired resistance.ResultsUsing normalized RNA-seq quantification measurements, we first identified oxidative phosphorylation and interferon alpha pathways as the two most enriched cancer hallmark gene sets in the baseline gene expression profile associated with intrinsic gemcitabine resistance and sensitivity, respectively. Furthermore, we discovered strong correlations between drug-induced expression changes in glycolysis and oxidative phosphorylation genes and response to gemcitabine, which suggests that these pathways may be associated with acquired gemcitabine resistance mechanisms. Thus, we developed prediction models using baseline gene expression profiles in those pathways and validated them in another dataset of 12 PDAC models from Novartis. We also developed prediction models based on drug-induced expression changes in genes from the Molecular Signatures Database (MSigDB)’s curated 50 cancer hallmark gene sets. Finally, pathogenic TP53 mutations correlated with treatment resistance.ConclusionOur results demonstrate that concurrent upregulation of both glycolysis and oxidative phosphorylation pathways occurs in vivo in PDAC PDXs following gemcitabine treatment and that pathogenic TP53 status had association with gemcitabine resistance in these models. Our findings may elucidate the molecular basis for gemcitabine resistance and provide insights for effective drug combination in PDAC chemotherapy.

Exploring potential therapeutic strategy for hepatocellular carcinoma and COVID-19 using bioinformatics analysis

BackgroundHepatocellular carcinoma (HCC) constitutes an important contributor to fatalities. Coronavirus disease 2019 (COVID-19) frequently presents with complications such as respiratory distress, systemic inflammatory responses, and damage to various organs. Several studies have investigated the relationship between COVID-19 and mortality in patients with liver cancer, but there are few research on the relationship between them. This study is to explore the correlation between the two diseases and drugs treating them.MethodsThe Gene Expression Omnibus (GEO) database provides gene datasets of COVID-19 patients and HCC patients. Through differential gene analysis and weighted gene co-expression network analysis, we determined 223 genes represented in HCC and COVID-19. We then used functional annotation, protein–protein interaction network construction, predictive model development and verification, prognostic value analysis, and miRNA–gene network construction. Besides, we created a drug–hub–gene network by predicting possible medications that interact with hub genes using the Drug–Gene Interaction Database (DGIdb). Ultimately, we applied immunohistochemistry to ascertain the hub genes expression.ResultsThis study revealed that eight core genes (RRM2, TPX2, DTL, CDT1, TYMS, CDCA5, CDC25C, and HJURP) co-existed in both HCC and COVID-19 and were differentially expressed in both HCC and normal tissues.CDC25C, RRM2, CDCA5, and HJURP had diagnostic value (AUC > 0.8) and prognostic value (adjusted P-value < 0.05). Genome enrichment analysis indicated that eight genes may function in liver cancer through engagement in the cell cycle, DNA replication, etc. In liver cancer samples, these genes were significantly and adversely associated with plasma cells while RRM2 was positively associated with neutrophil and NK cell activation and with dendritic cell resting. Using the miRNAnet database and DGIdb, 9 transcription factors, 7 miRNAs, and 51 drugs or molecular compounds were predicted to interact with the hub genes. Finally, RRM2 expression showed significant variation in clinical specimens, and analysis of the association of RRM2 with immunomodulators indicated that RRM2 was closely connected to MICB and CD276.ConclusionsOur study revealed several metabolic genes related to HCC and COVID-19. Moreover, potential drugs related to central genes were predicted. These findings may provide new ideas for treating COVID-19 and HCC.

Development and initial evaluation of a clinical prediction model for risk of treatment resistance in first-episode psychosis: Schizophrenia Prediction of Resistance to Treatment (SPIRIT).

A clinical tool to estimate the risk of treatment-resistant schizophrenia (TRS) in people with first-episode psychosis (FEP) would inform early detection of TRS and overcome the delay of up to 5 years in starting TRS medication. To develop and evaluate a model that could predict the risk of TRS in routine clinical practice. We used data from two UK-based FEP cohorts (GAP and AESOP-10) to develop and internally validate a prognostic model that supports identification of patients at high-risk of TRS soon after FEP diagnosis. Using sociodemographic and clinical predictors, a model for predicting risk of TRS was developed based on penalised logistic regression, with missing data handled using multiple imputation. Internal validation was undertaken via bootstrapping, obtaining optimism-adjusted estimates of the model's performance. Interviews and focus groups with clinicians were conducted to establish clinically relevant risk thresholds and understand the acceptability and perceived utility of the model. We included seven factors in the prediction model that are predominantly assessed in clinical practice in patients with FEP. The model predicted treatment resistance among the 1081 patients with reasonable accuracy; the model's C-statistic was 0.727 (95% CI 0.723-0.732) prior to shrinkage and 0.687 after adjustment for optimism. Calibration was good (expected/observed ratio: 0.999; calibration-in-the-large: 0.000584) after adjustment for optimism. We developed and internally validated a prediction model with reasonably good predictive metrics. Clinicians, patients and carers were involved in the development process. External validation of the tool is needed followed by co-design methodology to support implementation in early intervention services.

Investigation of Machine Learning Regression Techniques to Predict Critical Heat Flux Over a Large Parameter Space

A unifying and accurate model to predict critical heat flux (CHF) over a wide range of conditions has been elusive since wall boiling research emerged. With the release of the experimental data utilized in the development of the 2006 Groeneveld CHF lookup table (LUT), by far the most extensive public CHF database available to date (nearly 25 000 data points), the development of data-driven prediction models over a large parameter space in simple geometry (vertical, uniformly heated round tubes) can be achieved. Furthermore, the popularization of machine learning techniques to solve regression problems has led to more advanced tools for analyzing large and complex databases. This work compares three machine learning algorithms to predict the CHF database. For each selected regression algorithm (ν-Support vector, Gaussian process, and neural network), a set of optimized hyperparameters is applied. Among the investigated algorithms, the neural network emerges as the most effective, achieving a CHF predicted/measured factor standard deviation of 12.3%, three times less than that of the LUT. In comparison, the Gaussian process regression and the ν-support vector regression achieve a standard deviation of 17.7%, about two times lower than the LUT. Hence, all considered algorithms significantly outperform the LUT prediction performance. The neural network model and training methodology are designed to prevent overfitting, which is confirmed by data analyses of the CHF predictions. Finally, future development directions (including data coverage, transfer learning, and uncertainty quantification) are discussed.

Development of laminar burning velocity prediction model and correlation of iso-octane air mixtures using artificial neural network

Surrogate fuels provide an economical alternative for forecasting the combustion properties of transportation fuels like diesel, gasoline, and kerosene. Iso-octane (2,2,4-trimethylpentane), a primary reference fuel for gasoline, is extensively used as a surrogate component. This study utilizes a Feed-Forward Artificial Neural Network (FFANN) with back-propagation (BP) to predict the laminar burning velocity (LBV) of iso-octane/air mixtures. The ANN model was developed using a dataset of 6339 data points, including 3788 experimental data points from literature since 2004 and 2551 simulation data points from reduced kinetic reaction mechanism (RKM) CHEMKIN simulations. The grid search cross-validation (CV) method was employed to optimize the ANN hyperparameters. Implemented in Python using Keras, the ANN model addresses research gaps by forecasting LBV for hydrocarbons beyond n-heptane and modelling LBV under conditions reflective of actual engine operations. Unlike prior studies, we optimized various ML models, their complexity, size, and hyperparameters to achieve high prediction accuracy. We also integrated a hybrid model combining Particle Swarm Optimization (PSO) with Genetic Algorithms (GA) for hyperparameter optimization, ensuring efficient configuration of the ANN. The constructed ANN model was compared to other ML models, including generalized linear regression (GLR), support vector machine (SVM), random forest (RF), and XGBoost regression. For the testing set (20 % of the total dataset), the ANN model outperformed all other ML models and empirical correlations, achieving a coefficient of determination (R2) of 0.9903, root mean square error (RMSE) of 1.911, and mean absolute error (MAE) of 1.206. Optimizing the ANN model with PSO further enhanced prediction accuracy, achieving a correlation coefficient of 0.9973 and a reduced mean absolute error of 0.753. To evaluate the ANN model's predictive ability, a reduced mechanism from the Lawrence Livermore National Laboratory (LLNL) was used. Compared to the LLNL RKM, the ANN predicted values showed lower percentage deviation from experimental data. Additionally, computation cost comparisons revealed that while RKM simulations in CHEMKIN took 244.2 s per case, the PSO-optimized ANN model required only 900 s for 150 cases. A 16-term correlation for laminar burning velocity (LBV) was derived from ANN predictions, serving as a practical tool to predict LBV based on pressure, temperature, and equivalence ratio. Unlike existing literature correlations, this newly developed correlation is valid across a wide range of operating conditions, offering a fundamental database for real-time combustion applications.

Prediction of segmental motor outcomes in traumatic spinal cord injury: Advances beyond sum scores

Background and objectivesNeurological and functional recovery after traumatic spinal cord injury (SCI) is highly challenged by the level of the lesion and the high heterogeneity in severity (different degrees of in/complete SCI) and spinal cord syndromes (hemi-, ant-, central-, and posterior cord). So far outcome predictions in clinical trials are limited in targeting sum motor scores of the upper (UEMS) and lower limb (LEMS) while neglecting that the distribution of motor function is essential for functional outcomes. The development of data-driven prediction models of detailed segmental motor recovery for all spinal segments from the level of lesion towards the lowest motor segments will improve the design of rehabilitation programs and the sensitivity of clinical trials. MethodsThis study used acute-phase International Standards for Neurological Classification of SCI exams to forecast 6-month recovery of segmental motor scores as the primary evaluation endpoint. Secondary endpoints included severity grade improvement, independent walking, and self-care ability. Different similarity metrics were explored for k-nearest neighbor (kNN) matching within 1267 patients from the European Multicenter Study about Spinal Cord Injury before validation in 411 patients from the Sygen trial. The kNN performance was compared to linear and logistic regression models. ResultsWe obtained a population-wide root-mean-squared error (RMSE) in motor score sequence of 0.76(0.14, 2.77) and competitive functional score predictions (AUCwalker = 0.92, AUCself-carer = 0.83) for the kNN algorithm, improving beyond the linear regression task (RMSElinear = 0.98(0.22, 2.57)). The validation cohort showed comparable results (RMSE = 0.75(0.13, 2.57), AUCwalker = 0.92). We deploy the final historic control model as a web tool for easy user interaction (https://hicsci.ethz.ch/). DiscussionOur approach is the first to provide predictions across all motor segments independent of the level and severity of SCI. We provide a machine learning concept that is highly interpretable, i.e. the prediction formation process is transparent, that has been validated across European and American data sets, and provides reliable and validated algorithms to incorporate external control data to increase sensitivity and feasibility of multinational clinical trials.

A novel multitask learning algorithm for tasks with distinct chemical space: zebrafish toxicity prediction as an example

Data scarcity is one of the most critical issues impeding the development of prediction models for chemical effects. Multitask learning algorithms leveraging knowledge from relevant tasks showed potential for dealing with tasks with limited data. However, current multitask methods mainly focus on learning from datasets whose task labels are available for most of the training samples. Since datasets were generated for different purposes with distinct chemical spaces, the conventional multitask learning methods may not be suitable. This study presents a novel multitask learning method MTForestNet that can deal with data scarcity problems and learn from tasks with distinct chemical space. The MTForestNet consists of nodes of random forest classifiers organized in the form of a progressive network, where each node represents a random forest model learned from a specific task. To demonstrate the effectiveness of the MTForestNet, 48 zebrafish toxicity datasets were collected and utilized as an example. Among them, two tasks are very different from other tasks with only 1.3% common chemicals shared with other tasks. In an independent test, MTForestNet with a high area under the receiver operating characteristic curve (AUC) value of 0.911 provided superior performance over compared single-task and multitask methods. The overall toxicity derived from the developed models of zebrafish toxicity is well correlated with the experimentally determined overall toxicity. In addition, the outputs from the developed models of zebrafish toxicity can be utilized as features to boost the prediction of developmental toxicity. The developed models are effective for predicting zebrafish toxicity and the proposed MTForestNet is expected to be useful for tasks with distinct chemical space that can be applied in other tasks.Scieific contributionA novel multitask learning algorithm MTForestNet was proposed to address the challenges of developing models using datasets with distinct chemical space that is a common issue of cheminformatics tasks. As an example, zebrafish toxicity prediction models were developed using the proposed MTForestNet which provide superior performance over conventional single-task and multitask learning methods. In addition, the developed zebrafish toxicity prediction models can reduce animal testing.

Current limitations and future research needs for predicting soil precompression stress: A synthesis of available data

Precompression stress, compression index, and swelling index are used for characterizing the compressive behavior of soils, and are essential soil properties for establishing decision support tools to reduce the risk of soil compaction. Because measurements are time-consuming, soil compressive properties are often derived through pedotransfer functions. This study aimed to develop a comprehensive database of soil compressive properties with additional information on basic soil properties, site characteristics, and methodological aspects sourced from peer-reviewed literature, and to develop random forest models for predicting precompression stress using various subsets of the database. Our analysis illustrates that soil compressive properties data primarily originate from a limited number of countries. There is a predominance of precompression stress data, while little data on compression index or recompression index are available. Most precompression stress data were derived from the topsoils of conventionally tilled arable fields, which is not compatible with knowledge that subsoil compaction is a serious problem. The data compilation unveiled considerable variations in soil compression test procedures and methods for calculating precompression stress across different studies, and a concentration of data at soil moisture conditions at or above field capacity. The random forest models exhibited unsatisfactory predictive performance although they performed better than previously developed models. Models showed slight improvement in predictive power when the underlying data were restricted to a specific precompression stress calculation method. Although our database offers broader coverage of precompression stress data than previous studies, the lack of standardization in methodological procedures complicates the development of predictive models based on combined datasets. Methodological standardization and/or functions to translate results between methodologies are needed to ensure consistency and enable data comparison, to develop robust models for precompression stress predictions. Moreover, data across a wider range of soil moisture conditions are needed to characterize soil mechanical properties as a function of soil moisture, similar to soil hydraulic functions, and to develop models to predict the parameters of such soil mechanical functions.

Unlocking treatment success: predicting atypical antipsychotic continuation in youth with mania

PurposeThis study aimed to create and validate robust machine-learning-based prediction models for antipsychotic drug (risperidone) continuation in children and teenagers suffering from mania over one year and to discover potential variables for clinical treatment.MethodThe study population was collected from the national claims database in China. A total of 4,532 patients aged 4–18 who began risperidone therapy for mania between September 2013 and October 2019 were identified. The data were randomly divided into two datasets: training (80%) and testing (20%). Five regularly used machine learning methods were employed, in addition to the SuperLearner (SL) algorithm, to develop prediction models for the continuation of atypical antipsychotic therapy. The area under the receiver operating characteristic curve (AUC) with a 95% confidence interval (CI) was utilized.ResultsIn terms of discrimination and robustness in predicting risperidone treatment continuation, the generalized linear model (GLM) performed the best (AUC: 0.823, 95% CI: 0.792–0.854, intercept near 0, slope close to 1.0). The SL model (AUC: 0.823, 95% CI: 0.791–0.853, intercept near 0, slope close to 1.0) also exhibited significant performance. Furthermore, the present findings emphasize the significance of several unique clinical and socioeconomic variables, such as the frequency of emergency room visits for nonmental health disorders.ConclusionsThe GLM and SL models provided accurate predictions regarding risperidone treatment continuation in children and adolescents with episodes of mania and hypomania. Consequently, applying prediction models in atypical antipsychotic medicine may aid in evidence-based decision-making.

Research Trends in Postpartum Post-Taumatic Stress Disorder Among High-Risk Pregnancies: A CiteSpace Bibliometric Analysis

Background: High-risk pregnancies are associated with increased complications that can jeopardize maternal and fetal health, leading to adverse outcomes such as intrauterine growth restriction and preterm labor. Postpartum psychiatric challenges, including postpartum depression and post-traumatic stress disorder (PTSD), often emerge due to hormonal shifts and physical trauma during childbirth. Previous research indicates that postpartum PTSD is more prevalent among high-risk pregnancy populations, yet the field lacks centralization. Methods: This study utilized CiteSpace software to analyze 175 relevant publications on postpartum PTSD in high-risk pregnancies retrieved from the Web of Science Core Collection (2004–2024). A systematic literature review, employing a Boolean search strategy, ensured comprehensive data collection. The analysis included citation patterns, keyword co-occurrence, and clustering to identify key trends and emerging research areas. Results: Findings revealed a marked increase in research output, particularly from 2019 onward, with the United States leading in publication volume and centrality. High-frequency keywords included "posttraumatic stress disorder," "postpartum depression," and "risk factors," indicating primary focus areas in this domain. Cluster analysis identified eight thematic areas, with significant attention on factors influencing PP-PTSD, clinical symptoms, and prevention strategies. Keyword emergence maps highlighted evolving interests, particularly the impact of adverse childhood experiences and traumatic childbirth. Conclusion: The study underscores the urgent need for targeted mental health interventions for women with high-risk pregnancies, as these populations are particularly susceptible to postpartum PTSD. Continued research into risk factors and the development of predictive models is essential for improving maternal mental health care, ensuring that interventions are tailored to the specific needs of this vulnerable group. Enhanced understanding of the interplay between maternal health and postpartum outcomes is critical for optimizing care strategies.

An AI-Based Method for Estimating the Potential Runout Distance of Post-Seismic Debris Flows

The widely distributed sediments following an earthquake presents a continuous threat to local residential areas and infrastructure. These materials become more easily mobilized due to reduced rainfall thresholds. Before establishing an effective management plan for debris flow hazards, it is crucial to determine the potential reach of these sediments. In this study, a deep learning-based method—Dual Attention Network (DAN)—was developed to predict the runout distance of potential debris flows after the 2022 Luding Earthquake, taking into account the topography and precipitation conditions. Given that the availability of reliable precipitation data remains a challenge, attributable to the scarcity of rain gauge stations and the relatively coarse resolution of satellite-based observations, our approach involved three key steps. First, we employed the DAN model to refine the Global Precipitation Measurement (GPM) data, enhancing its spatial and temporal resolution. This refinement was achieved by leveraging the correlation between precipitation and regional environment factors (REVs) at a seasonal scale. Second, the downscaled GPM underwent calibration using observations from rain gauge stations. Third, mean absolute error (MAE), mean square error (MSE), and root mean square error (RMSE) were employed to evaluate the performance of both the downscaling and calibration processes. Then the calibrated precipitation, catchment area, channel length, average channel gradient, and sediment volume were selected to develop a prediction model based on debris flows following the Wenchuan Earthquake. This model was applied to estimate the runout distance of potential debris flows after the Luding Earthquake. The results show that: (1) The calibrated GPM achieves an average MAE of 1.56 mm, surpassing the MAEs of original GPM (4.25 mm) and downscaled GPM (3.83 mm); (2) The developed prediction model reduces the prediction error by 40 m in comparison to an empirical equation; (3) The potential runout distance of debris flows after the Luding Earthquake reaches 0.77 km when intraday rainfall is 100 mm, while the minimum distance value is only 0.06 km. Overall, the developed model offers a scientific support for decision makers in taking reasonable measurements for loss reduction caused by post-seismic debris flows.