Prediction Performance Of Model Research Articles

Automation of Local Regression Model Building for Spectroscopic Data

ABSTRACTThe concept of using local calibration for spectroscopic analysis has been discussed since the late 1980s. Since that time, many papers have described modifications to different aspects of the local modeling methodology. In this paper, we briefly discuss some of the modifications and describe an approach for the unattended automation of local model development. Ways to reduce calculation time are discussed. Four example spectroscopic datasets using Raman, FT‐NIR, and dispersive NIR are analyzed, and the local model prediction performance is compared to standard PLS prediction performance. Using independent prediction sets, local modeling is shown to improve prediction performance by 17% to 55%.

Machine Learning-based Dissolved Oxygen Prediction Modeling and Evaluation in the Yangtze River Estuary

Dissolved oxygen （DO） serves as a pivotal indicator, mirroring the intrinsic self-purification capacity of aquatic ecosystems and the overarching quality of the water environment. In the context of the Yangtze Estuary, a crucial hub for biodiversity and economic activities in China, understanding and forecasting levels of DO is instrumental for effective environmental stewardship and management strategies. Considering this, the introduction of sophisticated machine learning algorithms into the monitoring and predictive analytics of dissolved oxygen levels represents an important stride toward leveraging the power of data-driven insights for environmental sustainability. The Yangtze Estuary, characterized by its dynamic and complex hydrological and ecological systems, demands an insightful and nuanced approach to monitoring water quality parameters. To this end, six key monitoring stations were chosen across the estuary, including Xuliujing, Nantong Port, Qidong Port, Qinglong Port, South Port, and North Port, acting as sentinel sites for gauging the health of the water body. Leveraging three cutting-edge modeling techniques-particle swarm optimization-support vector regression （PSO-SVR）, artificial neural network （ANN）, and random forest （RF）-the research unraveled and forecasted the patterns of dissolved oxygen levels using monthly average water quality data spanning from 2004 to 2020. These models embodied the forefront of machine learning technology, each bringing distinct analytical strengths and perspectives to the table, from the nuanced, non-linear pattern recognition capabilities of ANN to the robustness and interpretability of RF. The meticulous evaluation conducted via the RF model underscored the paramount importance of three water quality variables, namely temperature, five-day biochemical oxygen demand, and ammonia nitrogen, in influencing the spatial-temporal dynamics of dissolved oxygen in the estuary. Comparative analysis of the prediction results yielded by the PSO-SVR, ANN, and RF models illuminated the superior performance of the RF model across the six monitoring stations, with an overall average error margin of 0.19, a testament to its efficacy and reliability. In comparison, the PSO-SVR and ANN models exhibited higher error rates of 0.38 and 0.47, respectively, albeit still contributing valuable insights into the complex dissolved oxygen dynamics in the Yangtze Estuary. The prediction performance of the machine learning models was evaluated, and the overall prediction performance ranking on the training set was RF （R2=0.971； RMSE=0.341 mg·L-1） > PSO-SVR （R2=0.884； RMSE=0.707 mg·L-1） > ANN （R2=0.792； RMSE=0.967 mg·L-1）. The overall prediction performance ranking on the test set was RF （R2 = 0.986； RMSE=0.165 mg·L-1） > PSO-SVR （R2=0.951； RMSE=0.332 mg·L-1） > ANN （R2=0.800； RMSE=0.633 mg·L-1）. Therefore, the RF model exhibited the best predictive ability on all monitoring sections, showing excellent performance and generalization ability both on the training and the test sets. The PSO-SVR model also performed well on most monitored profiles, with slightly lower predictive performance than that of the RF model though with better stability and generalization ability. However, the ANN model did not perform as perfectly as the other two models in some monitoring profiles and its network structure or parameters may need to be further optimized to improve the prediction accuracy and stability.

Psychological distress in biliary tract malignancy patients: influencing factors and development of a predictive nomogram model.

This study aims to investigate the psychological distress and its influencing factors in patients with biliary tract malignant tumors, alongside the development of a predictive model. A total of 219 patients diagnosed with biliary tract malignant tumors who were admitted to the Department of Liver Surgery at Fudan University Shanghai Cancer Center from July 2021 to May 2023, were selected using a convenience sampling method. Research tools involve psychological distress management screening tools, a demographic questionnaire, self-rating anxiety and depression scales, and the Chinese version of the Memorial Symptom Assessment Scale. Bootstrap method was utilized for repeated sampling to identify relevant factors influencing psychological distress in biliary tract cancer patients. The R software was employed to create a nomogram model, and the model's accuracy and predictive performance were assessed using the receiver operating characteristic curve (ROC) and the Hosmer-Lemeshow test. The average score of psychological distress among the 219 patients was (3.91 ± 2.44), with a psychological distress detection rate of 54.8%. Regression model results indicated that factors such as the presence of distant metastasis, comorbidity with other major diseases, poor sleep quality, anxiety, and severity of anxiety and depression were the primary influencers of psychological distress. The detection rate of psychological distress in patients with biliary tract malignant tumors is notably high. The predictive model constructed in this study exhibits good predictive efficacy and clinical value, providing valuable reference for healthcare professionals in developing targeted intervention strategies.

Repurposing Drugs for Infectious Diseases by Graph Convolutional Network with Sensitivity-Based Graph Reduction.

Computational systems biology employs computational algorithms and integrates diverse data sources, such as gene expression profiles, molecular interactions, and network modeling, to identify promising drug candidates through repurposing existing compounds in response to urgent healthcare needs. This study tackles the urgent need for rapid therapeutic development against emerging infectious diseases. We introduce a novel analytic expression for sensitivity analysis based on the Kronecker product and enhance model prediction performance using Graph Convolutional Networks (GCNs) with sensitivity-based graph reduction. Our algorithm refines prediction performance by leveraging sensitivity-based graph reduction. By integrating RNA-seq data, molecular interactions, and GCNs, we identify disease-related genes and pathways, construct heterogeneous graph models, and predict potential drugs. This approach involves novel analytical expressions that assess sensitivity to model loss, employing the Kronecker product approach. Subgraph analysis identifies nodes for removal, leading to a refined graph used for model retraining. This cost-effective pipeline focuses on computational methods for drug repurposing, targeting infectious diseases such as Zika virus and COVID-19 infection. Applied to these infections, our methodology integrates 659 proteins and 703 drugs for Zika virus, and 495 proteins and 468 drugs for COVID-19, along with their interactions derived from gene expression profiles. Top candidate drugs, such as Betamethasone phosphate and Bizelesin for Zika virus, and Chloroquine, Heparin Disaccharide, and Resveratrol for COVID-19, were validated through literature review or docking analysis. This scalable approach demonstrates promise in repurposing drugs for urgent healthcare challenges.

A novel nephrometry scoring system outperforms RENAL score, PADUA score, and SPARE score for feasibility prediction of nephron-sparing surgery in children with bilateral Wilms tumor.

Several nephrometry scores have been proposed and applied in the adult population. We previously established a novel model to predict the feasibility of nephron-sparing surgery (NSS) in pediatric bilateral Wilms tumor (WT) patients. This study aimed to evaluate whether our model had better predictive performance compared to other scores. Data from 58 patients during the period 2008-2019 were retrospectively reviewed. Nephrometry scores, including RENAL [radius (R), exophytic/endophytic (E), nearness to collecting system/sinus (N), anterior/posterior (A), and location relative to polar lines (L)], PADUA (preoperative aspects and dimensions used for an anatomical classification), and SPARE (Simplified PADUA REnal) scores, were calculated. A senior radiologist and a senior urologist independently reviewed computed tomography (CT) or magnetic resonance scans while blinded to clinical outcomes. A urology resident re-evaluated the imaging scans using our RSR [relation with collecting system (R), size (S), remaining renal parenchyma (R)] model. The areas under the curve (AUCs) were compared. Inter-rater reliability of the RSR model was assessed using kappa statistics and intraclass correlation coefficient (ICC). The AUC was 0.982 for the RSR model, which outperformed the other three scores. Decision curve analysis (DCA) showed the same results. Compared to the RENAL score, the RSR model showed significant reclassification ability [0.573 (95% confidence interval: 0.161, 1.204), P=0.03]. Significant overall discrimination ability was found with RSR compared to the other three scores (P<0.001). The ICC value was 0.871 for S, and the kappa values of the two R indicators were 0.806 and 0.777, respectively. A web-based calculator was constructed. The RSR model's predictive performance outperforms the established adult nephrometry scores while offering overall fair interobserver agreement. A web-based calculator could help to make personalized treatment decisions.

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.

Impact analysis of uncertainty in thermal resistor-capacitor models on model predictive control performance

Model Predictive Control (MPC) is extensively utilized for optimal control in building systems. Despite substantial research being dedicated to exploring the impact of uncertainties in external and internal disturbances on the performance of MPC, the existing studies neglect the potential impact of uncertainties in model parameter identification on control performance. To address this gap, this study quantifies the impact of model uncertainty on MPC performance through a test case in a virtual environment. Various levels of uncertainties for parameters R and C are artificially introduced to assess the MPC performance. The causes of the impact of model uncertainty on control performance are further explored through analysis. We select a first-order RC model to modelling building thermal dynamics. MPC is employed to optimize the heat pump signal with the goal of minimizing the energy cost while maintaining thermal comfort. The simulation results demonstrate that a negative deviation in model parameter identification has a more pronounced impact on MPC performance than a positive deviation, which has a negligible effect on MPC control performance. Deviations in parameters from their true values affect both heat losses from the zone and thermal capacity, thereby influencing the estimated temperature by the RC model. Consequently, these factors, in turn, affect the system’s control decisions, leading to variations in the objective function values. This study can provide an insight into the relationship of model parameters uncertainties and MPC performance and facilitate the practical application of MPC in buildings.

Neural Network Approaches for Early Breast Cancer Detection

Breast cancer research remains essential due to its high mortality rates and the critical need for improved diagnostic accuracy. This study investigates the application of neural network techniques to enhance early breast cancer detection, utilizing Artificial Neural Networks (ANN) implemented in MATLAB. By analyzing datasets from the UCI Machine Learning Repository, specifically the Coimbra and Wisconsin Breast Cancer datasets, this research demonstrates the accuracy and efficiency of neural networks in classifying malignant and benign cases. Methodologically, the study involves data preprocessing, ANN pattern recognition modeling, and testing across multiple metrics including confusion matrices and ROC plots to validate the model's predictive performance. The findings underscore the potential of neural networks as a diagnostic support tool, facilitating faster and more accurate cancer detection, thereby contributing to improved patient outcomes and supporting global health initiatives for early diagnosis.

Construction and validation of a novel prognostic model with palmitoylation-related genes for glioblastoma.

Glioblastoma multiforme (GBM), the most prevalent and aggressive primary brain tumor, poses substantial challenges in both treatment and prognosis. Post-translational modifications, like palmitoylation, are known to have critical roles in the development and progression of glioma. Yet, the molecular mechanisms involved in palmitoylation and its prognostic significance in GBM are still not fully understood. This study aimed to explore prognostic biomarkers for GBM based on palmitoylation-related genes and to construct a prognostic risk model. The messenger ribonucleic acid (mRNA) expressions data and the clinical information were downloaded from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) to explore palmitoylation-related mechanisms in GBM. The Cox regression analysis was performed to identify prognostic palmitoylation-related genes and the consensus clustering was used for molecular classification. The package "limma" was used for differential gene expression analysis and the least absolute shrinkage and selection operator (LASSO) regression was applied to construct a risk signature. A nomogram model was established using the risk score and clinical variables. Receiver operating characteristic (ROC), calibration curve, and decision curve analysis (DCA) were used to assess the predicted accuracy and clinical benefit of the model. The difference in immune cell infiltration was compared between different risk groups. The drug susceptibility analysis and immunotherapy response prediction were conducted to access the ability of the risk signature in predicting the therapeutic effect. Based on datasets from TCGA, five palmitoylation-related genes were identified as prognostic markers, allowing for the categorization of GBM patients into two subtypes with differing survival rates. Through differential expression analysis, 570 specific genes linked to GBM advancement were uncovered. A total of seven signature genes (COL22A1, IGFBP6, SOD3, UPP1, CA14, TIMP4 and FERMT1) were applied to establish a prognostic risk model, which was demonstrated to be an independent prognostic indicator for patients with GBM. Kaplan-Meier analysis indicted that the GBM patients in low-risk group exhibited a better survival outcome compared the patients in high-risk group. The ROC curve analyses demonstrated that the risk score model was reliable. The nomograms showed excellent predictive ability. Two external cohort of patients from the GSE74187 and GSE83300 in the GEO database confirmed the model's strong predictive performance. The immune infiltration, drug sensitivity and immunotherapy responses were significantly different between the low- and high-risk groups. Our study offers insights into the molecular classification and prognostic assessment of GBM, focusing on palmitoylation-related mechanisms. The prognostic model we constructed provides valuable guidance for tailoring personalized treatment strategies for GBM patients.

Comparing experimental designs for parameterizing semi-empirical and deep learning-based lithium-ion battery aging models

Design of Experiment (DOE) methods can be applied to optimize test plans of cycle life aging studies with the aim to efficiently parameterize lithium-ion battery aging models. Since different DOEs exist and their effect on the prediction performance of battery aging models has not yet been investigated, we conducted a cycle life aging study with six commonly used DOEs (One-factor-at-a-time, Taguchi, Box–Behnken, Central Composite, Full Factorial, and D-optimal) and compare their influence on the prediction performance of a semi-empirical and a deep learning-based battery aging model. The results show that the semi-empirical model benefits the most from statistically optimized test plans. Compared to randomly selecting test plans, applying DOE methods helps to consistently achieve one of the lowest possible prediction errors for a given number of test points. Furthermore, it is shown that a D-optimal test plan and the test plans obtained from response surface methods (Box–Behnken and Central Composite) require only half as many test points as a Full Factorial test design, but still result in semi-empirical models with a high prediction accuracy that is similar to the Full Factorial test design. In contrast, deep learning-based battery aging models benefit significantly less from statistically optimized test plans. The highest prediction accuracy is achieved by the Full Factorial test plan and all other DOEs result in higher prediction errors and are even outperformed by several randomly defined test plans. Instead of using static designs, deep-learning-based models profit from a dynamic test optimization, which reduces the number of tested batteries during cycle life testing based on their information gain. We demonstrate that with our proposed dynamic test reduction algorithm, which analyzes the information gain based on aging features extracted after 100 EFC of cycling, up to 50% of all tested batteries of a Full Factorial test plan can be excluded from the cycle life study without deteriorating the prediction accuracy of the resulting deep learning-based battery aging model.

Novel deep learning based soft sensor feature extraction for part weight prediction in injection molding processes

In the current injection molding (IM) industry, it remains challenging to monitor and estimate production quality promptly. It is costly and time-consuming to measure part quality manually after each production cycle ends, which results in quality defects difficult to be captured in time. In this case, a soft sensor is essential to model the IM process and predict the final quality in real time with multi-source industrial production data. However, traditional data-driven modeling methods fail to take advantage of the information in complex high-frequency data from in-mold sensors, resulting in an inaccurate IM model and unsatisfactory quality prediction performance. To solve this problem, this paper proposes a novel soft sensor framework based on a teacher-student structure. After specialized preprocessing of multiple sensor time series data, a GRU-based autoencoder with an attention mechanism (GRU-A-AE) is trained as a teacher model, extracting deep implicit features involving valuable time sequential information. Then, a cascaded relationship among shallow feature points from sensor signals, deep features, and final part weights is established using back propagation neural networks (BPNNs). To demonstrate its effectiveness and superiority, the proposed soft sensor is trained and tested with practical IM data under normal and fluctuating production conditions, respectively. Compared with conventional methods, our method has higher prediction accuracy with testing RMSE of 0.1049 and R2 of 0.9950 under normal conditions, which proves more valuable information in high-frequency sensor signals are explored from the teacher model and IM production dynamics are captured precisely. In addition, its better prediction performance in the case of production condition fluctuation verifies its strong robustness.

Feature Variable Selection for Near‐Infrared Spectroscopy Based on Simulated Annealing Bee Colony Algorithm

ABSTRACTVariable selection is an effective method to enhance the modeling performance of near‐infrared spectroscopy. Given the promising application prospects of intelligent optimization algorithms in spectral feature variable selection, this article combines the artificial bee colony algorithm with the simulated annealing algorithm to construct a simulated annealing bee colony algorithm (SABC). To explore the feasibility of SABC for spectral variable selection, SABC was applied to construct a partial least squares spectral quantitative detection model for corn stover cellulose and soil organic matter contents. The modeling performance was compared with that of the full spectrum, genetic algorithm, simulated annealing algorithm, and artificial bee colony algorithm; it was found that the model regression precision established by SABC was the best. For the cellulose and organic matter content detection models, the coefficients of determination of the validation set were 0.9433 and 0.9853, with the relative root mean squared error of 1.7901% and 0.8011%, and the residual prediction deviation of 4.1741 and 8.3931, respectively, which could meet the corresponding actual detection needs. SABC adopted the strategy of multiple runs to select the repeated wavelength variables, effectively reduced variable dimensions and model complexity, improved the prediction performance of the regression model, and provided a new approach for building a high‐performance near‐infrared spectroscopy (NIRS) quantitative calibration model.

Differentiation between invasive ductal carcinoma and ductal carcinoma in situ by combining intratumoral and peritumoral ultrasound radiomics

BackgroundThis study aimed to develop and validate an ultrasound radiomics model for distinguishing invasive ductal carcinoma (IDC) from ductal carcinoma in situ (DCIS) by combining intratumoral and peritumoral features.MethodsRetrospective analysis was performed on 454 patients from Chengzhong Hospital. The patients were randomly divided in accordance with a ratio of 8:2 into a training group (363 cases) and validation group (91 cases). In addition, 175 patients from Yanghu Hospital were used as the external test group. The peritumoral ranges were set to 2, 4, 6, 8, and 10 mm. Mann–Whitney U-test, recursive feature elimination, and a least absolute shrinkage and selection operator were used to in the dimension reduction of the radiomics features and clinical knowledge, and machine learning logistic regression classifiers were utilized to construct the diagnostic model. The area under the curve (AUC) of the receiver operating characteristics, accuracy, sensitivity, and specificity were used to evaluate the model performance.ResultsBy combining peritumoral features of different ranges, the AUC of the radiomics model was improved in the validation and test groups. In the validation group, the maximum increase in AUC was 9.7% (P = 0.031, AUC = 0.803) when the peritumoral range was 8 mm. Similarly, when the peritumoral range was only 8 mm in the test group, the maximum increase in AUC was 4.9% (P = 0.005, AUC = 0.770). In this study, the best prediction performance was achieved when the peritumoral range was only 8 mm.ConclusionsThe ultrasound-based radiomics model that combined intratumoral and peritumoral features exhibits good ability to distinguish between IDC and DCIS. The selection of peritumoral range size exerts an important effect on the prediction performance of the radiomics model.

Exploring the Potential of Adaptive, Local Machine Learning in Comparison to the Prediction Performance of Global Models: A Case Study from Bayer's Caco-2 Permeability Database.

Machine learning (ML) techniques are being widely implemented to fill the gap in simple molecular design guidelines for newer therapeutic modalities in the extended and beyond rule of five chemical space (eRo5, bRo5). These ML techniques predict molecular properties directly from the structure, allowing for the prioritization of promising compounds. However, the performance of models varies greatly among ML use cases. A molecular property for which achieving sufficient performance in generalizing global models still remains difficult is Caco-2 permeability. Especially within the lower permeability ranges, which are specific for larger molecules belonging to the e/bRo5 space, accurate regression predictions have proven to be challenging. The present study, therefore, identifies a suitable combination of ML algorithm and descriptors, consisting of the LightGBM algorithm and RDKit molecular property descriptors, to predict Caco-2 permeability very efficiently by a simple global model. An additionally introduced local model uses the same algorithm and descriptors but selects its training data based on Tanimoto fingerprint similarity to match the individual test compound's structure. Evaluation of this adaptive model, by systematically varying the number of most similar structures for training, shows that, in comparison to the global model, there was only marginally improved performance with specific training data constellations. These random improvements indicate that deriving general rules for local model parametrization is not possible a priori for the chosen algorithm and descriptor combination, and preselecting training data does not seem advantageous over global ML based on all available data, while creation of more data-efficient models was generally proven to be possible.

Predicting Monthly Runoff: A BO-CNN-LSTM Approach for Enhanced Water Resource Management

Monthly runoff prediction is critical for sustainable water resource management, flood control, and drought preparedness. Accurate forecasts assist decision-making and management of water resources, which are critical for human and environmental demands. Previous prediction algorithms frequently lack accuracy due to human-defined hyperparameters. Autonomous hyperparameter tuning is required for improved predictions. Objectives: To improve the accuracy of monthly runoff predictions, this research proposes a BO-CNN-LSTM model that integrates Bayesian optimization (BO) with a convolutional neural network-long short-term memory neural network (CNN-LSTM). Methods/Analysis: The BO algorithm is utilized to enhance the CNN-LSTM model's hyperparameters, which helps to overcome the limits of human-set parameters. Model training and validation were carried out using monthly runoff data from 1956 to 2019 from Lanxi Station in the Hulan River Basin. The model's prediction performance was assessed using root mean square error (RMSE), mean absolute error (MAE), fitting coefficient (R²), and mean relative error (MAPE), and compared with other previous models. Findings: The results show that the BO-CNN-LSTM model outperforms conventional models in prediction accuracy and reduction of errors. The BO method efficiently enhances hyperparameters, resulting in higher performance metrics. This model gives more precise and dependable monthly runoff forecasts, which are critical for handling the Hulan River Basin's water resources. Novelty/Improvement: By combining BO and CNN-LSTM, the proposed model solves the constraints of previous models, improving prediction accuracy and reducing errors. This innovative technique presents a promising novel approach for monthly runoff prediction, with possible applications in larger hydrological prediction.

Solar Irradiance Prediction Method for PV Power Supply System of Mobile Sprinkler Machine Using WOA-XGBoost Model

Solar energy can mitigate the power supply shortage in remote regions for portable irrigation systems. The accurate prediction of solar irradiance is crucial for determining the power capacity of photovoltaic power generation (PVPG) systems for mobile sprinkler machines. In this study, a prediction method is proposed to estimate the solar irradiance of typical irrigation areas. The relation between meteorological parameters and solar irradiance is studied, and four different parameter combinations are formed and considered as inputs to the prediction model. Based on meteorological data provided by ten typical radiation stations uniformly distributed nationwide, an Extreme Gradient Boosting (XGBoost) model optimized using the Whale Optimization Algorithm (WOA) is developed to predict solar radiation. The prediction accuracy and stability of the proposed method are then evaluated for different input parameters through training and testing. The differences between the prediction performances of models trained based on single-station data and mixed data from multiple stations are also compared. The obtained results show that the proposed model achieves the highest prediction accuracy when the maximum temperature, minimum temperature, sunshine hours ratio, relative humidity, wind speed, and extraterrestrial radiation are used as input parameters. In the model testing, the RMSE and MAE of WOA-XGBoost are 2.142 MJ·m−2·d−1 and 1.531 MJ·m−2·d−1, respectively, while those of XGBoost are 2.298 MJ·m−2·d−1 and 1.598 MJ·m−2·d−1. The prediction effectiveness is also verified based on measured data. The WOA-XGBoost model has higher prediction accuracy than the XGBoost model. The model developed in this study can be applied to forecast solar irradiance in different regions. By inputting the meteorological parameter data specific to a given area, this model can effectively produce accurate solar irradiance predictions for that region. This study provides a foundation for the optimization of the configuration of PVPG systems for mobile sprinkler machines.

Analysis of factors influencing bronchiectasis patients with active pulmonary tuberculosis and development of a nomogram prediction model.

To identify the risk factors for bronchiectasis patients with active pulmonary tuberculosis (APTB) and to develop a predictive nomogram model for estimating the risk of APTB in bronchiectasis patients. A retrospective cohort study was conducted on 16,750 bronchiectasis patients hospitalized at the Affiliated Hospital of Guangdong Medical University and the Second Affiliated Hospital of Guangdong Medical University between January 2019 and December 2023. The 390 patients with APTB were classified as the case group, while 818 patients were randomly sampled by computer at a 1:20 ratio from the 16,360 patients with other infections to serve as the control group. Relevant indicators potentially leading to APTB in bronchiectasis patients were collected. Patients were categorized into APTB and inactive pulmonary tuberculosis (IPTB) groups based on the presence of tuberculosis. The general characteristics of both groups were compared. Variables were screened using the least absolute shrinkage and selection operator (LASSO) analysis, followed by multivariate logistic regression analysis. A nomogram model was established based on the analysis results. The model's predictive performance was evaluated using calibration curves, C-index, and ROC curves, and internal validation was performed using the bootstrap method. LASSO analysis identified 28 potential risk factors. Multivariate analysis showed that age, gender, TC, ALB, MCV, FIB, PDW, LYM, hemoptysis, and hypertension are independent risk factors for bronchiectasis patients with APTB (p < 0.05). The nomogram demonstrated strong calibration and discrimination, with a C-index of 0.745 (95% CI: 0.715-0.775) and an AUC of 0.744 for the ROC curve. Internal validation using the bootstrap method produced a C-index of 0.738, further confirming the model's robustness. The nomogram model, developed using common clinical serological characteristics, holds significant clinical value for assessing the risk of APTB in bronchiectasis patients.

A Hybrid Quantum-Classical Model for Stock Price Prediction Using Quantum-Enhanced Long Short-Term Memory.

The stock markets have become a popular topic within machine learning (ML) communities, with one particular application being stock price prediction. However, accurately predicting the stock market is a challenging task due to the various factors within financial markets. With the introduction of ML, prediction techniques have become more efficient but computationally demanding for classical computers. Given the rise of quantum computing (QC), which holds great promise for being exponentially faster than current classical computers, it is natural to explore ML within the QC domain. In this study, we leverage a hybrid quantum-classical ML approach to predict a company's stock price. We integrate classical long short-term memory (LSTM) with QC, resulting in a new variant called QLSTM. We initially validate the proposed QLSTM model by leveraging an IBM quantum simulator running on a classical computer, after which we conduct predictions using an IBM real quantum computer. Thereafter, we evaluate the performance of our model using the root mean square error (RMSE) and prediction accuracy. Additionally, we perform a comparative analysis, evaluating the prediction performance of the QLSTM model against several other classical models. Further, we explore the impacts of hyperparameters on the QLSTM model to determine the best configuration. Our experimental results demonstrate that while the classical LSTM model achieved an RMSE of 0.0693 and a prediction accuracy of 0.8815, the QLSTM model exhibited superior performance, achieving values of 0.0602 and 0.9736, respectively. Furthermore, the QLSTM outperformed other classical models in both metrics.

Fake News Detection Revisited: An Extensive Review of Theoretical Frameworks, Dataset Assessments, Model Constraints, and Forward-Looking Research Agendas

The emergence and acceptance of digital technology have caused information pollution and an infodemic on Online Social Networks (OSNs), blogs, and online websites. The malicious broadcast of illegal, objectionable and misleading content causes behavioural changes and social unrest, impacts economic growth and national security, and threatens users’ safety. The proliferation of AI-generated misleading content has further intensified the current situation. In the previous literature, state-of-the-art (SOTA) methods have been implemented for Fake News Detection (FND). However, the existing research lacks multidisciplinary considerations for FND based on theories on FN and OSN users. Theories’ analysis provides insights into effective and automated detection mechanisms for FN, and the intentions and causes behind wide-scale FN propagation. This review evaluates the available datasets, FND techniques, and approaches and their limitations. The novel contribution of this review is the analysis of the FND in linguistics, healthcare, communication, and other related fields. It also summarises the explicable methods for FN dissemination, identification and mitigation. The research identifies that the prediction performance of pre-trained transformer models provides fresh impetus for multilingual (even for resource-constrained languages), multidomain, and multimodal FND. Their limits and prediction capabilities must be harnessed further to combat FN. It is possible by large-sized, multidomain, multimodal, cross-lingual, multilingual, labelled and unlabelled dataset curation and implementation. SOTA Large Language Models (LLMs) are the innovation, and their strengths should be focused on and researched to combat FN, deepfakes, and AI-generated content on OSNs and online sources. The study highlights the significance of human cognitive abilities and the potential of AI in the domain of FND. Finally, we suggest promising future research directions for FND and mitigation.

Study on the predictive model of delirium risk after surgery for elderly hip fractures based on meta-analysis.

To develop and validate a risk prediction model for postoperative delirium in elderly patients with hip fractures, aiming to identify high-risk patients and implement preventive measures. A systematic search of five authoritative medical databases was conducted, retrieving a total of 1368 relevant articles. After screening, 44 high-quality studies were included in the meta-analysis, analyzing 13 potential risk factors, such as age, gender, diabetes, and history of stroke. A risk prediction model was constructed and validated in a cohort of 189 elderly hip fracture patients. The model's predictive performance was evaluated using ROC curves, with calibration assessed through the Hosmer-Lemeshow test, and clinical utility examined via Decision Curve Analysis (DCA) and Clinical Impact Curves (CIC). The meta-analysis identified the following as independent risk factors for postoperative delirium: age (≥ 70years), male gender, diabetes, history of stroke, preoperative comorbidities (≥ 2), previous delirium, preoperative cognitive impairment, low preoperative albumin levels (≤ 40g/L), prolonged preoperative waiting time (≥ 48h), anemia (≤ 100g/L), ASA classification (≥ 3), use of general anesthesia, and prolonged surgery duration (≥ 2h). The prediction model demonstrated strong efficiency in the validation cohort, with an AUC of 0.956, sensitivity of 87.3%, specificity of 94.8%, and a Brier score of 0.144, indicating high predictive accuracy and calibration. DCA and CIC analyses showed the model to have strong clinical decision-making value and impact across most thresholds. The risk prediction model developed in this study shows high predictive accuracy and clinical utility, making it valuable for identifying high-risk patients and implementing preventive measures in clinical practice. However, the study has limitations, such as potential retrospective bias, and further validation in larger, multicenter prospective studies is needed to confirm the model's broader applicability and stability.