Reliable Prediction Model Research Articles

P0216 PANoptosis-related genes can predict the progression of ulcerative colitis and colitis-associated colorectal cancer

Abstract Background Ulcerative colitis (UC) is a complex inflammatory bowel disease that significantly increases the risk of developing colitis-associated colorectal cancer (CAC). The etiology of UC remains unclear, and the mechanisms underlying its progression to CAC require further exploration. Recent studies have identified a novel form of cell death called pan-apoptosis, which is closely linked to inflammatory diseases. However, its specific role in UC and CAC progression is not yet fully understood. This study aims to develop a reliable predictive model based on pan-apoptosis using high-throughput RNA sequencing and validate it, along with an analysis of cellular characteristics in UC patients through single-cell sequencing data. This research offers new insights for the prognosis and treatment of UC patients. Methods This study collected single-cell and high-throughput sequencing data from UC patients and defined a pan-apoptosis gene set. The dataset was analyzed for immune infiltration, differential expression, Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Set Enrichment Analysis (GSEA). Additionally, weighted gene co-expression network analysis (WGCNA) was used to identify key genes associated with UC, which were then utilized to construct a predictive model. Finally, single-cell data analysis was conducted on UC patients, followed by in vitro validation of the expression of key genes. Results By analyzing differentially expressed genes in UC samples and using the pan-apoptosis gene set, WGCNA identified key genes associated with UC, including CASP1, LCN2, STAT3, and ZBP1. These genes were used to construct a pan-apoptosis-related predictive model, establishing a pan-apoptosis-related score (PAR-Score). The results showed that PAR-Score has strong classification ability and is positively correlated with immune cell infiltration and the inflammatory microenvironment in UC patients. Notably, CAC patients had significantly higher PAR-Scores. Single-cell sequencing data confirmed that monocyte activity in UC patients is linked to pan-apoptosis, while intercellular communication among immune cells was notably reduced. The expression levels of key genes in the predictive model were validated in vivo. Drug predictions targeting these genes suggested that Leridistim, interferon-γ, and other compounds could serve as potential therapeutic agents for UC or CAC. Conclusion The predictive model established based on pan-apoptosis exhibits good predictive performance, providing new perspectives for assessing the condition and progression of UC patients, and opening up new avenues for the treatment of UC and CAC in the future. References Nature. 2007;448(7152):427-434. doi:10.1038/nature060052. Jess T, Rungoe C, Peyrin L. Risk of Colorectal Cancer in Patients With Ulcerative Colitis: A Meta-analysis of Population-Based Cohort Studies. 2012;10(6). 3. Shin J, Baek GH, Cha B, et al. Complementary Therapeutic Effect of Fecal Microbiota Transplantation in Ulcerative Colitis after the Response to Anti-Tumor Necrosis Factor Alpha Agent Was Lost: A Case Report. Published online 2024. doi:10.3390/biomedicines12040800 Front Cell Infect Microbiol. 2019;9:406. doi:10.3389/fcimb.2019.00406 J Immunol. 2022;209(9):1625-1633. doi:10.4049/jimmunol.2200508 World J Gastroenterol. 2014;20(44):16389. doi:10.3748/wjg.v20.i44.16389

Prediction of Macroeconomic Growth Using Backpropagation Algorithms: A Review

This study aims to evaluate the effectiveness of the backpropagation algorithm in predicting macroeconomic growth through a Systematic Literature Review (SLR) approach. The review analyzes literature sourced from reputable indexes such as Scopus, DOAJ, and Google Scholar, focusing on publications from the last decade (2013–2024). It examines various studies on the application of the backpropagation algorithm, including parameter settings and model selection that influence prediction accuracy. The findings indicate that careful parameter configuration, such as the number of neurons, hidden layers, and learning rates, along with appropriate model selection, significantly enhance the performance of the backpropagation algorithm in macroeconomic prediction. This study highlights that combining optimal techniques and accurate parameter configurations substantially improves prediction accuracy and efficiency. It provides valuable insights and practical guidance for researchers and practitioners in designing more reliable and effective macroeconomic prediction models.

A hybrid data assimilation method based on real-time Ensemble Kalman filtering and KNN for COVID-19 prediction

This study introduces a hybrid data assimilation method that significantly improves the predictive accuracy of the time-dependent Susceptible-Exposed-Asymptomatic-Infected-Quarantined-Removed (SEAIQR) model for epidemic forecasting. The approach integrates real-time Ensemble Kalman Filtering (EnKF) with the K-Nearest Neighbors (KNN) algorithm, combining dynamic real-time adjustments with pattern recognition techniques tailored to the specific dynamics of epidemics. This hybrid methodology overcomes the limitations of single-model predictions in the face of increasingly complex transmission pathways in modern society. Numerical experiments conducted using COVID-19 case data from Xi’an, Shaanxi Province, China (December 9, 2021, to January 8, 2022) demonstrate a marked improvement in forecasting accuracy relative to traditional models and other data assimilation approaches. These findings underscore the potential of the proposed method to enhance the accuracy and reliability of predictive models, providing valuable insights for future epidemic forecasting and disease control strategies.

A Comprehensive Review of Wind Power Prediction Based on Machine Learning: Models, Applications, and Challenges

Wind power prediction is essential for ensuring the stability and efficient operation of modern power systems, particularly as renewable energy integration continues to expand. This paper presents a comprehensive review of machine learning techniques applied to wind power prediction, emphasizing their advantages over traditional physical and statistical models. Machine learning methods, especially deep learning approaches such as Convolutional Neural Networks (CNNs), Long Short-Term Memory Networks (LSTMs), and ensemble learning techniques like XGBoost, excel in addressing the nonlinearity and complexity of wind power data. The review also explores critical aspects such as data preprocessing, feature selection strategies, and model optimization techniques, which significantly enhance prediction accuracy and robustness. Challenges such as data acquisition difficulties, complex terrain influences, and sensor quality issues are examined in depth, with proposed solutions discussed. Additionally, the paper highlights future research directions, including the potential of multi-model fusion, emerging deep learning technologies like Transformers, and the integration of smart sensors and IoT technologies to develop intelligent, automated, and reliable prediction systems. By addressing existing challenges and leveraging advanced machine learning techniques, this work provides valuable insights into the current state of wind power prediction research and offers strategic guidance for enhancing the applicability and reliability of prediction models in practical scenarios.

Precision Prediction of Alzheimer's Disease: Integrating Mitochondrial Energy Metabolism and Immunological Insights.

Alzheimer's disease (AD), a prevalent neurodegenerative disorder, is characterized by mitochondrial dysfunction and immune dysregulation. This study is aimed at developing a risk prediction model for AD by integrating multi-omics data and exploring the interplay between mitochondrial energy metabolism-related genes (MEMRGs) and immune cell dynamics. We integrated four GEO datasets (GSE132903, GSE29378, GSE33000, GSE5281) for differential gene expression analysis, functional enrichment, and weighted gene co-expression network analysis (WGCNA). We identified two key gene modules (turquoise and magenta) significantly correlated with AD. Subsequently, we constructed a risk prediction model incorporating five MEMRGs (MRPL15, RBP4, ABCA1, MPV17, and MRPL37) and clinical factors using LASSO regression. The model demonstrated robust predictive performance (AUC > 0.815) in both internal and external validation (GSE44770) cohorts. Downregulation of MRPL15, RBP4, MPV17, and MRPL37 in AD brain regions (validated using AlzData and qRT-PCR) suggests impaired mitochondrial function. Conversely, ABCA1 upregulation may represent a compensatory response. Furthermore, significant differences in immune cell proportions, particularly gamma delta T cells (p = 0.002) and activated CD4 memory T cells (p = 0.027), were found between AD and non-demented samples. We observed significant correlations between MEMRG expression and specific immune cell fractions, indicating a potential link between mitochondrial dysfunction and immune dysregulation in AD. Our study provides a reliable risk prediction model for AD and highlights the crucial roles of MEMRGs and immune responses in disease pathogenesis, offering potential targets for therapeutic interventions.

Establishment of a Prognostic Necroptosis-Related lncRNA Signature in Ovarian Cancer.

Ovarian Cancer (OC) was known for its high mortality rate among gynecological malignancies, often resulting in a poor prognosis. This study sought to identify prognostic necroptosis-related long non-coding RNAs (lncRNAs) (NRlncRNAs) with prognostic potential and to construct a reliable risk prediction model for OC patients. The transcriptome and clinic data were sourced from TCGA and GTEx databases. Initially, NRlncRNAs were discovered by assessing gene correlations and evaluating differences in gene expression. Subsequently, Cox regression and LASSO methods were employed to develop the NRlncRNAs risk model, which was further validated through survival analysis, ROC curves, Cox regression, and nomograms across both the test and entire datasets. Multivariate Cox analysis revealed that the risk score based on 14 NRlncRNAs can independently predict the prognosis of OC. The low-risk group demonstrated significantly higher immune cell infiltration scores and lower tumor immune dysfunction, exclusion, and TIDE scores, as well as an increased number of neoantigens and higher TMB. Notably, the low-risk group also exhibited an elevated HRD score. The model's predictive accuracy was further substantiated through ROC analysis, showing superior performance compared to many existing models.Finally, the expression levels of 14 NRlncRNAs were confirmed using the qRT-PCR in two OC cell lines. These findings suggested that the NRlncRNAs risk model could serve as a more precise indicator for forecasting immune response and outcomes of targeted treatments in OC.

Predictive model of neurodevelopmental outcome in neonatal hypoxic ischemic encephalopathy.

To build an early, prognostic model for adverse outcome in infants with hypoxic ischemic encephalopathy (HIE) receiving therapeutic hypothermia (TH) based on brain magnetic resonance images (MRI), electrophysiological tests and clinical assessments were performed during the first 5days of life. Retrospective study of 182 neonates with HIE and managed with TH. The predominant pattern of HIE brain injury on MRI performed following cooling was scored by neuroradiologists. The electroencephalogram (EEG) background and evoked potential (EP) response, were analyzed. Area under the curve (AUC) of these tools for adverse outcome including death and/or moderate disabilities using the Bayley-III at 36months were calculated. A stepwise model approach was used to reach the final most efficient predictive model. Of 182 neonates, 99 were male (54.4%), with median gestational age of 39weeks (IQR 38-40) and median weight of 3.3kg (IQR 2.9-3.7). On admission, 47 (26%), 104 (57%) and 31(17%) neonates presented with mild, moderate and severe encephalopathy respectively. In multivariate analysis of 129 infants who received all prognostic modalities, the predictive value of a model of EEG plus MRI, AUC=84%) is equivalent to models of EEG plus MRI with added EP and clinical assessment at discharge (AUC=84 and 85% respectively). In the era of cooling for neonatal HIE, the combination of EEG background and MRI during the first few days of life, provide an objective and highly reliable model for prediction of death and long-term disabilities.

Leveraging Explainable AI and Multimodal Data for Stress Level Prediction in Mental Health Diagnostics

The increasing prevalence of mental health issues, particularly stress, has necessitated the development of data-driven, interpretable machine learning models for early detection and intervention. This study leverages multimodal data, including activity levels, perceived stress scores (PSS), and event counts, to predict stress levels among individuals. A series of models, including Logistic Regression, Random Forest, Gradient Boosting, and Neural Networks, were evaluated for their predictive performance. Results demonstrated that ensemble models, particularly Random Forest and Gradient Boosting, performed significantly better compared to Logistic Regression. Random Forest achieved an accuracy of 73%, while Gradient Boosting delivered a balanced precision-recall tradeoff with an accuracy of 72%. Gradient Boosting outperformed in identifying high-stress instances, achieving a recall of 70%, making it the most reliable model for stress prediction. Explainable AI (XAI), which refers to the ability of machine learning models to provide transparent and understandable explanations for their predictions, was employed in this study using SHAP (SHapley Additive exPlanations). SHAP values revealed that Perceived Stress Score (PSS) had the most significant impact on predictions, followed by event count and activity inference. Higher PSS scores strongly correlated with high-stress predictions, while increased event counts and activity levels were associated with lower stress. These findings underscore the importance of incorporating behavioral patterns in stress diagnostics and highlight the utility of explainable models in improving transparency, trust, and usability for clinical decision-making. This research establishes a robust foundation for deploying interpretable machine learning systems to support mental health diagnostics and enhance clinician decision support.

Enhancing stroke disease classification through machine learning models via a novel voting system by feature selection techniques.

Heart disease remains a leading cause of mortality and morbidity worldwide, necessitating the development of accurate and reliable predictive models to facilitate early detection and intervention. While state of the art work has focused on various machine learning approaches for predicting heart disease, but they could not able to achieve remarkable accuracy. In response to this need, we applied nine machine learning algorithms XGBoost, logistic regression, decision tree, random forest, k-nearest neighbors (KNN), support vector machine (SVM), gaussian naïve bayes (NB gaussian), adaptive boosting, and linear regression to predict heart disease based on a range of physiological indicators. Our approach involved feature selection techniques to identify the most relevant predictors, aimed at refining the models to enhance both performance and interpretability. The models were trained, incorporating processes such as grid search hyperparameter tuning, and cross-validation to minimize overfitting. Additionally, we have developed a novel voting system with feature selection techniques to advance heart disease classification. Furthermore, we have evaluated the models using key performance metrics including accuracy, precision, recall, F1-score, and the area under the receiver operating characteristic curve (ROC AUC). Among the models, XGBoost demonstrated exceptional performance, achieving 99% accuracy, precision, F1-Score, 98% recall, and 100% ROC AUC. This study offers a promising approach to early heart disease diagnosis and preventive healthcare.

Predicting Transient Anomalous Transport in Two‐Dimensional Discrete Fracture Networks With Dead‐End Fractures

AbstractPollutant transport in discrete fracture networks (DFNs) exhibits complex dynamics that challenge reliable model predictions, even with detailed fracture data. To address this issue, this study derives an upscaled integral‐differential equation to predict transient anomalous diffusion in two‐dimensional (2D) DFNs. The model includes both transmissive and dead‐end fractures (DEFs), where stagnant water zones in DEFs cause non‐uniform flow and transient sub‐diffusive transport, as shown by both literature and DFN flow and transport simulations using COMSOL. The upscaled model's main parameters are quantitatively linked to fracture properties, especially the probability density function of DEF lengths. Numerical experiments show the model's accuracy in predicting the full‐term evolution of conservative tracers in 2D DFNs with power‐law distributed fracture lengths and two orientation sets. Field applications indicate that while model parameters for transient sub‐diffusion can be predicted from observed DFN distributions, predicting parameters controlling solute displacement in transmissive fractures requires additional field work, such as tracer tests. Parameter sensitivity analysis further correlates late‐time solute transport dynamics with fracture properties, such as fracture density and average length. Potential extensions of the upscaled model are also discussed. This study, therefore, proves that transient anomalous transport in 2D DFNs with DEFs can be at least partially predicted, offering an initial step toward improving model predictions for pollutant transport in real‐world fractured aquifer systems.

THERMAL COMFORT PREDICTION USING SVM AND RANDOM FOREST MODEL

Predicting thermal comfort is crucial for optimizing built environments for human habitation, as it impacts health, productivity, and overall well-being. To address this imperative, interdisciplinary collaboration among architects, engineers, psychologists, and data scientists is needed to develop reliable predictive models that anticipate occupants’ thermal comfort preferences across diverse environmental conditions and architectural designs. Traditional methods rely on human comfort models, which can be subjective and time-consuming. Machine learning algorithms, such as Support Vector Machines (SVM) and Random Forest (RF), have been utilized to predict thermal comfort with high accuracy and efficiency. The Internet of Things (IoT) is revolutionizing the building management systems industry, with adaptive control algorithms and modular architectures exploring the IoT paradigm. This paper discusses the use of SVM and Random Forest algorithms for predicting thermal comfort in buildings, exploring their strengths and weaknesses and comparing their performance in different scenarios. The study analyzed a dataset of thermal comfort data, filtering by quantity and removing outliers. The data was split into 80% for training and 20% for testing. The study used SVM and Random Forest models to capture complex relationships between environmental parameters and thermal comfort responses. The results showed that the IQR method provided 3-4% accuracy, while the reducing label values method offered 20-23% accuracy. The study also tested the parameters of the models, resulting in a 2-4% difference between the two models. The study concluded that Random Forest appears more stable than SVM and plans to add new features to improve accuracy.

Leaflet: Operative Steps for Interventional Studies in Neuroscience.

Drug development involves multiple stages, spanning from initial discovery to clinical trials. This intricate process entails understanding disease mechanisms, identifying potential drug targets, and evaluating the efficacy and safety of candidate drugs. Clinical trials are designed to assess the effects of drugs on humans, focusing on determining safety profiles, appropriate modes of administration, and comparative efficacy against placebos. Notably, neuroscience drug development encounters distinct challenges, including the complex nature of diseases, limitations imposed by the blood-brain barrier, the absence of reliable predictive preclinical models, and regulatory hurdles. Ethical and safety considerations are pivotal due to the potential cognitive and motor effects of CNS-active drugs. Our manuscript outlines the procedures for CNS clinical trials and highlights the key elements of study design, methodological considerations, and ethical frameworks. To achieve our objectives, we considered the official websites of regulatory authorities, the EQUATOR network, and recent publications in the field. The paper includes key elements such as criteria for subject selection, methods of evaluation, variable analysis, and statistical methodology approaches. We want to furnish a concise and comprehensive guide tailored to individuals new to CNS clinical trials, providing foundational elements necessary for the design and execution of such trials. The manuscript seeks to outline sources of relevant materials and elucidate adaptability, particularly in instances where sponsors may be absent. By meeting the needs of less-experienced researchers or those with limited resources, the intention is to facilitate an understanding of the intricate nature of the process and offer guidance on appropriately navigating its complexities. It is essential to note that this manuscript does not aim to be exhaustive but endeavors to serve as a structured checklist. Through its approach, the manuscript aspires to offer guidance and support to individuals navigating the challenges inherent in this intricate domain.

Prediction of miRNA-disease associations based on PCA and cascade forest

BackgroundAs a key non-coding RNA molecule, miRNA profoundly affects gene expression regulation and connects to the pathological processes of several kinds of human diseases. However, conventional experimental methods for validating miRNA-disease associations are laborious. Consequently, the development of efficient and reliable computational prediction models is crucial for the identification and validation of these associations.ResultsIn this research, we developed the PCACFMDA method to predict the potential associations between miRNAs and diseases. To construct a multidimensional feature matrix, we consider the fusion similarities of miRNA and disease and miRNA-disease pairs. We then use principal component analysis(PCA) to reduce data complexity and extract low-dimensional features. Subsequently, a tuned cascade forest is used to mine the features and output prediction scores deeply. The results of the 5-fold cross-validation using the HMDD v2.0 database indicate that the PCACFMDA algorithm achieved an AUC of 98.56%. Additionally, we perform case studies on breast, esophageal and lung neoplasms. The findings revealed that the top 50 miRNAs most strongly linked to each disease have been validated.ConclusionsBased on PCA and optimized cascade forests, we propose the PCACFMDA model for predicting undiscovered miRNA-disease associations. The experimental results demonstrate superior prediction performance and commendable stability. Consequently, the PCACFMDA is a potent instrument for in-depth exploration of miRNA-disease associations.

Integration of Thermo-, Hydrodynamic, and Kinetic Factors in the Mathematical Modeling of the Catalytic Reforming Process

Background: The integration of various factors affecting processes in oil refining is crucial for enhancing both the efficiency and sustainability of the industry. In a changing market and increasingly stringent environmental regulations, it is essential to continuously update approaches, develop innovative solutions, and optimize production processes to achieve the best possible outcomes. Aim: The study aims to integrate thermodynamic, kinetic and hydrodynamic aspects into a unified model, and to validate the outcome based on experimental data and real-world operating conditions to ensure the accuracy and reliability of model predictions. Materials and methods: The primary research methods include statistical data analysis, process modeling, and experimental studies at various stages of the production cycle. Results: The study identified the key parameters that significantly impact the quality of the final product and production efficiency. Furthermore, it offers recommendations for optimizing production processes based on the data obtained. Conclusion: The study concludes that integrating various factors can significantly enhance production performance and reduce refining costs. The study emphasizes the importance of an integrated approach to the management of production processes in the oil refining industry, which can facilitate the further development of the industry. The model created can be utilized for training personnel in process simulation. With its user-friendly interface, it requires no extensive programming knowledge, making it well-suited for the initial training of specialists.

Rainfall From Brazilian Flying Rivers: Evaluating the Effectiveness of Precipitation Gridded Databases

ABSTRACTThe uneven global distribution of rainfall significantly impacts water resources and environmental sustainability, emphasising the need for reliable climate prediction models. Accurate predictions are vital for sectors such as food security, urban planning and disaster management. Data from ground stations, radars and satellites are essential, despite challenges like instrumental errors. Satellites, with their comprehensive sensors, are crucial for atmospheric observations, aiding in the prediction of large‐scale climatic events. Climate models such as CHIRPS, GLDAS, TerraClimate, and PERSIANN use different approaches to analyse precipitation data, which is key to understanding its spatial and temporal variability. This study evaluated (rainfall data) from these four climate models over 20 years (within the Brazilian territory), focusing on the spatiotemporal behaviour of rainfall using statistical metrics such as R2, RMSE, and MAPE. The findings showed that CHIRPS had the best performance (R2 = 0.843; RMSE = 42.83; MAPE = 0.09%), excelling in both overall database and extreme event analyses. TerraClimate, initially the lowest‐performing model (R2 = 0.413; RMSE = 91.56; MAPE = 0.23%), improved significantly when combined with elevation through multiple linear regression (MLR), achieving R2 of 0.718, RMSE of 31.14, and MAPE of 9.56%. This made TerraClimate a viable model for studying the Flying Rivers. The study highlights that model selection should align with the specific characteristics of the area under consideration, with CHIRPS being particularly suitable for the studied region. This research enhances the understanding of the effectiveness of these models in estimating rainfall compared to in situ measurements, which is crucial for various applications. The authors advocate for further studies to advance research on the Flying Rivers, their significance, and the impacts of climate change on them.

ECMHA-PP: A Breast Cancer Prognosis Prediction Model Based on Energy-Constrained Multi-Head Self-Attention.

Breast cancer is a significant threat to women's health. Precise prognosis prediction for breast cancer can help doctors implement more rational treatment strategies. Artificial intelligence can assist doctors in decision-making and enhance predictionaccuracy. In this paper, a deep learning model ECMHA-PP (Energy Constrained Multi-Head Self-Attention based Prognosis Prediction) is proposed to predict the prognosis of breast cancer. ECMHA-PP utilizes patients' clinical data and extracts features through a cross-position mix and a channel mix multi-layer perceptron. Then, it incorporates an energy-constrained multi-head self-attention layer to improve feature extraction capability. The source code of ECMHA-PP has been hosted on GitHub and is available at https://github.com/xiaoliu166370/ECMHA-PP. To evaluate our proposed method, prognostic prediction experiments were performed on the METABRIC dataset, yielding outstanding results with an average accuracy of 93.0% and an average area under the curve of 0.974. To further validate the model's performance, we conducted tests on another independent dataset, BRCA, achieving an accuracy of 87.6%. In comparison with other widely used advanced methods, ECMHA-PP demonstrated higher comprehensive performance, making it a reliable prognostic prediction model for breast cancer. Given its robust feature extraction and predictioncapabilities.

Development and validation of web-based risk score predicting prognostic nomograms for elderly patients with primary colorectal lymphoma: A population-based study.

Primary colorectal lymphoma (PCL) is an infrequently occurring form of cancer, with the elderly population exhibiting an increasing prevalence of the disease. Furthermore, advanced age is associated with a poorer prognosis. Accurate prognostication is essential for the treatment of individuals diagnosed with PCL. However, no reliable predictive survival model exists for elderly patients with PCL. Therefore, this study aimed to develop an individualized survival prediction model for elderly patients with PCL and stratify its risk to aid in the treatment and monitoring of patients. Patients aged 60 or older with PCL from 1975 to 2013 in the Surveillance, Epidemiology, and End Results database were selected and randomly divided into a training cohort (n = 1305) and a validation cohort (n = 588). The patients from 2014-2015 (n = 207) were used for external validation. The research team utilized both Cox regression and the least absolute shrinkage and selection operator (LASSO) regression to analyze potential predictors, in order to identify the most suitable model for constructing an OS-nomogram and an associated network version. The risk stratification is constructed on the basis of this model. The performance of the model was evaluated based on the consistency index (C-index), calibration curve, and decision curve analysis (DCA) to determine its resolving power and calibration capability. Age, gender, marital status, Ann Arbor staging, primary site, surgery, histological type, and chemotherapy were independent predictors of Overall Survival (OS) and were therefore included in our nomogram. The Area Under the Curve (AUC) of the 1, 3, and 5-year OS in the training, validation, and external validation sets ranged from 0.732 to 0.829. The Receiver Operating Characteristic (ROC) curves showed that the nomogram model outperformed the Ann Arbor stage system when predicting elderly patients with PCL prognosis at 1, 3, and 5 years in the training set, validation dataset, and external validation cohort. The Concordance Index (C-index) also demonstrated that the nomogram had excellent predictive accuracy and robustness. The calibration curves demonstrated a strong agreement between observed and predicted values. In the external validation cohort, the C-index (0.769, 95%CI: 0.712-0.826) and calibration curves of 1000 bootstrap samples also indicated a high level of concordance between observed and predicted values. The nomogram-related DCA curves exhibited superior clinical utility when compared to Ann Arbor stage. Furthermore, an online prediction tool for overall survival has been developed: https://medkuiwang.shinyapps.io/DynNomapp/. This was the first study to construct and validate predictive survival nomograms for elderly patients with PCL, which is better than the Ann Arbor stage. It will help clinicians manage elderly patients with PCL more accurately.

TPE-xgboost for explainable predictions of concrete compressive strength considering compositions, and mechanical and microstructure properties of testing samples

The use of machine learning (ML) for predicting concrete compressive strength (CCS) has shown promising and accurate results, making it a valuable tool in the field. However, efficient prediction requires not only a robust ML approach but also a comprehensive and well-curated dataset. This study addresses this challenge by investigating an extensive and integrated dataset of (1) concrete compositions (mixture proportions) and (2) testing conditions (mechanical and microstructure properties of concrete testing samples), containing 1525 observations relating to 39 parameters. On algorithms side, this research proposes novel tree-structured parzen estimator based extreme gradient boosting (TPE-xgboost) for accurate and confident CCS predictions. Moreover, SHapley Additive exPlanations (SHAP) analysis was performed to provide a comprehensive understanding of feature importance, dependencies, and interactions. Compared to prior research, this study demonstrates significant improvements in CCS prediction accuracy for separate investigations on concrete compositions (3.77%) and mechanical and microstructure properties (28.57%), while maintaining reasonable accuracy for the integrated dataset despite its complexity and sparseness. SHAP analysis reveals key influential factors, including age and water-to-binder ratio in concrete composition, and peak load and height of testing samples, as well as microstructural characteristics such as mean values of global autocorrelation length and its integral range. This research provides valuable insights into model optimization, predictive performance, and feature dependencies and interactions, contributing to the development of more accurate and reliable CCS prediction models.

Deep Learning-Based Automatic Segmentation Combined with Radiomics to Predict Post-TACE Liver Failure in HCC Patients.

To develop and validate a deep learning-based automatic segmentation model and combine with radiomics to predict post-TACE liver failure (PTLF) in hepatocellular carcinoma (HCC) patients. This was a retrospective study enrolled 210 TACE-trated HCC patients. Automatic segmentation model based on nnU-Net neural network was developed to segment medical images and assessed by the Dice similarity coefficient (DSC). The screened clinical and radiomics variables were separately used to developed clinical and radiomics predictive model, and were combined through multivariate logistic regression analysis to develop a combined predictive model. The area under the curve (AUC), calibration curve, and decision curve analysis (DCA) were applied to compare the performance of the three predictive models. The automatic segmentation model showed satisfactory segmentation performance with an average DSC of 83.05% for tumor segmentation and 92.72% for non-tumoral liver parenchyma segmentation. The international normalized ratio (INR) and albumin (ALB) was identified as clinically independent predictors for PTLF and used to develop clinical predictive model. Ten most valuable radiomics features, including 8 from non-tumoral liver parenchyma and 2 from tumor, were selected to develop radiomics predictive model and to calculate Radscore. The combined predictive model achieved the best and significantly improved predictive performance (AUC: 0.878) compared to the clinical predictive model (AUC: 0.785) and the radiomics predictive model (AUC: 0.815). This reliable combined predictive model can accurately predict PTLF in HCC patients, which can be a valuable reference for doctors in making suitable treatment plan.

Development and Validation of a New Model Including Inflammation Indexes for the Long-Term Prognosis of Hepatitis B-Related Acute-On-Chronic Liver Failure.

Acute-on-chronic liver failure (ACLF) is a severe condition characterized by a systemic inflammatory response and associated with high mortality. Currently, there is no reliable prediction model for long-term prognosis in ACLF. This study aimed to develop and validate a prognostic model incorporating inflammation indexes to predict the long-term outcome of patients with hepatitis B virus-related ACLF (HBV-ACLF). A retrospective analysis of clinical data from HBV-ACLF patients (n = 986) treated at the Third Affiliated Hospital of Sun Yat-sen University between January 2014 and December 2018 was conducted. Patients were randomly divided into training (n = 690) and validation (n = 296) cohorts. The Least Absolute Shrinkage and Selection Operator (LASSO) and Cox regression analyses were used to identify independent risk factors for long-term mortality. The following variables were identified as independent predictors of long-term mortality: age, cirrhosis, hepatic encephalopathy, total bilirubin (TBIL), international normalized ratio (INR), monocyte-to-lymphocyte ratio (MLR), and neutrophil-to-platelet ratio (NPR). A novel nomogram was established by assigning weights to each variable. The C-index of the nomogram was 0.777 (95% confidence interval [CI]: 0.752-0.802). In the training set, the area under the curve (AUC) for predicting mortality at 1, 3, and 12 months was 0.841 (95% CI: 0.807-0.875), 0.827 (95% CI: 0.796-0.859), and 0.829 (95% CI: 0.798-0.859), respectively. The nomogram demonstrated superior predictive performance for 12-month survival compared to the model for end-stage liver disease (MELD) score (0.767, 95% CI: 0.730-0.804, p < 0.001) and the clinical overt sepsis in acute liver failure clinical practice Guidelines-ACLF II score (0.807, 95% CI: 0.774-0.840, p = 0.028). Finally, calibration curves and decision curve analysis (DCA) confirmed the clinical utility of the nomogram. The novel inflammation-based scoring system, incorporating MLR and NPR, effectively predicts long-term mortality in HBV-ACLF patients.