Prediction Of Events Research Articles

Establishment of Agro-Eco Industrial Clusters: A Romanian Perspective

The Romanian economy feels the need to formalize links between the situational economic forecast and the forecast of technological events in industry and agriculture. The article studies a “dimension” of the organization for classifications of data, structures and economic / agro-eco-industrial events for clustering with the help of “dimensionality”. Meta-prognostic relationships between economics, forecasting and forecasting agents (mathematical elements) are described and the role of metricity, sub-metricity and ultra-metricity in the observation of an automatic ontology for classifications supporting agro-eco-industrial / food clustering is highlighted. The paper proposes the algorithm of the informal classes of supervised / unsupervised data about agro-eco-industrial / food companies in Romania in order to establish classifications that would motivate the formation of clusters in the field. It is concluded that the mathematical apparatus for classification must be developed because based on symbolic mathematical models can be developed computer / computer programs for calculations of ultra-metricity of transformations for linearization’s of similarities between firms, eliminating reminiscences / redundancies, aiming to reduce space / distances between data that represent, in fact, enterprises that can enter agro-eco-industrial / food clusters. Keywords: agro-eco-industrial clusters, ultra-metricity, clustering, economic forecasting, classification

Prediction of hepatocellular carcinoma and liver-related events in anti-HDV positive individuals

Chronic hepatitis D (CHD) is the most severe form of chronic viral hepatitis, with a high risk of developing hepatocellular carcinoma (HCC) and liver-related mortality. Risk stratification is needed to guide HCC surveillance strategies and to prioritize treatment with antiviral agents. We conducted a multicenter retrospective cohort of anti-HDV positive individuals managed at sites in the Netherlands and the United Kingdom. We studied the 5-year cumulative incidences of HCC and liver-related events (first of HCC, liver transplantation and liver-related mortality), in the overall cohort and among relevant subgroups. We analyzed 269 anti-HDV positive individuals with a median follow-up of 4.3 years in which 47 first events occurred. The 5-year cumulative incidences of HCC and liver-related events were 3.8% and 15.6% in the overall cohort. The 5-year cumulative incidence of HCC and liver-related events for individuals without cirrhosis was 0% and 0.9% compared to 12% and 41.3% for individuals with cirrhosis (p<0.001). The 5-year cumulative incidence of HCC and liver-related events was 0% and 2.1% among individuals with low PAGE-B scores, compared to 3.2% and 21.1% with intermediate and 25.4% and 45.5% with high risk scores (p<0.001). We found comparable results for the FIB-4 score. Findings were consistent regardless of cirrhosis or detectable HDV RNA (p<0.001). Anti-HDV positive individuals are at high risk of adverse liver-related outcomes. The incidences of HCC was negligible among individuals without cirrhosis and among individuals with low baseline PAGE-B and/or FIB-4 scores. Therefore, these score can be used to guide HCC surveillance strategies and potentially also for treatment prioritization.

Can the Charlson Comorbidity Index help to guide DOAC dosing in patients with atrial fibrillation and improve the efficacy and safety of treatment? A subanalysis of the MAS study: Charlson index and DOAC dosing in AF patients.

Frailty influences the effectiveness and safety of anticoagulant therapy in patients with atrial fibrillation (AF). The age-weighted Charlson Comorbidity Index may offer a valuable tool to assess the risk of adverse events in AF patients treated with direct oral anticoagulants (DOACs). This sub-analysis of MAS trial data aimed to assess whether using the Charlson index, instead of the standard criteria, would have led to different dosing and improved adverse event occurrence during treatment. The MAS study looked for a relationship between DOAC levels assessed at baseline and adverse events during follow-up. The study is described in detail elsewhere. Among the 1,657 patients studied, 832 (50.2%) had a relatively low Charlson index (up to 6, general median class), of whom 132 (15.9%) were treated with reduced doses. Conversely, among the 825 patients with a high Charlson index (≥7), 257 (31.1%) received standard doses. A weak but statistically significant positive correlation (r = 0.1413, p = 0.0001 by ANOVA) was observed between increasing Charlson classes and DOAC levels standardized to allow comparability among drug results. However, no significant differences were found in the incidence or number of adverse events during follow-up, or in other parameters, between patients with low and high Charlson scores. Utilizing the Charlson index would have led to notable differences in DOAC dosing compared to standard criteria. However, we found no evidence that its use would have improved the prediction of adverse events in AF patients enrolled in the MAS study.

Comparative Analysis of Machine Learning Models for Earthquake Prediction: A Case Study of Düzce, Türkiye

This paper explores the application of machine learning models, specifically XGBoost, Stacking Regressor, and Long Short-Term Memory (LSTM), for predicting earthquake magnitudes in Düzce, Turkey. The models were trained and tested on seismic data to predict moment magnitude (Mw). The performance of each model was evaluated using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and the coefficient of determination (R²). The results indicate that the XGBoost model outperforms the other models with a higher R² value and lower error metrics, providing a more accurate prediction of seismic events.

TCT-173 Contemporary Risk Prediction and Stratification of Long-Term Major Adverse Cardiac and Cerebrovascular Events After Percutaneous Coronary Intervention: The REMEDY Score

TCT-173 Contemporary Risk Prediction and Stratification of Long-Term Major Adverse Cardiac and Cerebrovascular Events After Percutaneous Coronary Intervention: The REMEDY Score

Prediction of Low-Visibility Events by Integrating the Potential of Persistence and Machine Learning for Aviation Services

Fog typically results in reduced atmospheric visibility. Severely limited visibility has a significant impact on transportation, particularly the operations of aircraft. Precise forecasts of low visibility are essential for aviation services, primarily for the efficient planning of airport activities. Despite the utilization of sophisticated numerical weather prediction (NWP) models, the prediction of fog and limited visibility remains challenging. The intricacy of fog prediction is due to limitations in understanding the micro-scale factors that lead to fog genesis, intensification, persistence, and dissipation. This study investigates the occurrence of fog (surface visibility &lt;1000 m) and dense fog (surface visibility &lt; 200 m) throughout the climatological low-visibility months (November to February) to analyze the persistence of low-visibility events and predict them in the specific conditions of the frog prone Indo-Gangetic Plain (IGP) regions. A representative station, Jay Prakash Narayan International (JPNI) Airport in Patna, India, has been considered given the availability of instrumental quality datasets. The analysis investigates the long-term and short-term persistence and prediction of the series using a diverse variety of machine learning (ML) algorithms. To conduct a comprehensive analysis over an extended period, detrended fluctuation analysis (DFA) is employed to determine the similarities between the time series of large-scale fog and dense fog. A Markov chain model is used to look at the binary time series and figure out how long low-visibility events (like fog and dense fog) last in the short term ( 1-5 hours). Ultimately, we analyze a short-term forecast (Nowcast) with a lead time of one to five hours for instances of low visibility (fog or dense fog). This nowcasting is generated utilizing diverse methodologies, including Markov chain models, persistence analysis, and machine learning (ML) methods. Finally, establish that the most favorable and reliable results in this prediction problem are attained by employing a Mixture of Experts model that integrates persistence-based methods and ML algorithms.

Strategic data augmentation with CTGAN for smart manufacturing: Enhancing ML predictions of paper breaks in pulp-and-paper production

This article addresses the significant challenge posed by the rare occurrence of paper breaks in the pulp-and-paper industry, which, despite their infrequency during production, have considerable economic consequences. We analyze operational data from a paper manufacturing machine, focusing on the rare instances of paper breaks (124 out of 18,398 cases) identified through a quality assurance protocol. This scarcity presents a challenge for machine learning-based predictive models. To overcome this, we introduce a novel data augmentation strategy using Conditional Generative Adversarial Networks (CTGAN) and Synthetic Minority Oversampling Technique (SMOTE), aimed at generating synthetic data that not only reflects the real operational data’s distribution but also improves the predictive models’ performance metrics. We assess the impact of this data augmentation on three machine learning algorithms—Decision Trees (DT), Random Forest (RF), and Logistic Regression (LR)—both before and after augmentation. The use of a CTGAN-enhanced dataset led to notable enhancements in predictive maintenance metrics. Specifically, the ability of the models to detect machine breaks (Class 1) increased significantly: over 30% for Decision Trees, 20% for Random Forest, and nearly 90% for Logistic Regression. This methodological progress offers valuable contributions to the fields of industrial quality control and maintenance scheduling by improving the prediction of rare events in manufacturing processes.

AcFibroMASH Index for the Diagnosis of Fibrotic MASH and Prediction of Liver-related Events: An International Multicenter Study

Background & AimsMetabolic dysfunction-associated steatohepatitis (MASH) and fibrotic MASH are significant health challenges. This multi-national study aimed to validate the acMASH index (including serum creatinine and aspartate aminotransferase concentrations) for MASH diagnosis and develop a new index (acFibroMASH) for non-invasively identifying fibrotic MASH and exploring its predictive value for liver-related events (LREs). MethodsWe analyzed data from 3,004 individuals with biopsy-proven metabolic dysfunction-associated fatty liver disease (MAFLD) across 29 Chinese and nine international cohorts to validate the acMASH index and develop the acFibroMASH index. Additionally, we utilized the independent external data from a multi-national cohort of 9,034 patients with MAFLD to examine associations between the acFibroMASH index and the risk of LREs. ResultsIn the pooled global cohort, the acMASH index identified MASH with an AUROC of 0.802 (95%CI 0.786-0.818). The acFibroMASH index (including the acMASH index plus liver stiffness measurement) accurately identified fibrotic MASH with an AUROC of 0.808 in the derivation cohort and 0.800 in the validation cohort. Notably, the AUROC for the acFibroMASH index was 0.835 (95% CI 0.786-0.882), superior to that of the FAST score at 0.750 (95% CI 0.693-0.800, P<0.01) in predicting the 5-year risk of LREs. Patients with acFibroMASH >0.39 had a higher risk of LREs than those with acFibroMASH <0.15 (adjusted-hazard ratio: 11.23 95%CI 3.98-31.66). ConclusionsThis multi-ethnic study validates the acMASH index as a reliable, non-invasive test for identifying MASH. The newly proposed acFibroMASH index is a reliable test for identifying fibrotic MASH and predicting the risk of LREs.

Lake responses and mechanisms to El Niño on the Tibetan Plateau using deep learning-based semantic segmentation

Numerous lakes across the Tibetan Plateau (TP) serve as crucial indicators of climate change and are significantly influenced by El Niño events. Previous studies of lake response to El Niño events have focused on a limited number of lakes. Despite advances in remote sensing technology, there have been few comprehensive and large-scale studies using deep learning, and there are still gaps in understanding the response mechanisms on a larger scale. This study leverages advanced deep-learning techniques to map lake responses, offering unprecedented insights into the large-scale hydrological impacts of El Niño. Our results show that lakes shrink significantly during El Niño events on the TP. Lakes located in the central and southern parts of the TP and small lakes with areas ranging from 1 to 50 km2 (over 60 % of them) exhibited strong responses. The range of lake response to El Niño events varies with their intensity, with stronger El Niño events causing an expansion of the response range along the latitudinal direction. We propose four possible mechanisms for lake response patterns to El Niño from the perspective of lake water sources. Strong shrinkage is primarily caused by decreased precipitation and increased evaporation, with a possible contribution from reduced meltwater. Strong expansion is due to increased precipitation, more glacier and frozen soil meltwater, and reduced evaporation. For slight shrinkage and expansion patterns, the balance of meltwater may offset or even counteract the El Niño signal. The study’s results could improve predictions of extreme weather events like droughts and floods in the Third Pole region, enhance water resource management and responsiveness, and offer valuable insights for ecological monitoring and early warning systems development.

Machine-Learning Approaches for Major Adverse Cardiovascular Events (MACE) Prediction in Patients on Hemodialysis

Machine-Learning Approaches for Major Adverse Cardiovascular Events (MACE) Prediction in Patients on Hemodialysis

TCT-809 Examining Prognostic Scores for Short-Term Major Adverse Cardiovascular Events Prediction After Primary PCI: A Comparative Analysis

TCT-809 Examining Prognostic Scores for Short-Term Major Adverse Cardiovascular Events Prediction After Primary PCI: A Comparative Analysis

Machine learning for adverse event prediction in outpatient parenteral antimicrobial therapy: a scoping review.

This study aimed to conduct a scoping review of machine learning (ML) techniques in outpatient parenteral antimicrobial therapy (OPAT) for predicting adverse outcomes and to evaluate their validation, implementation and potential barriers to adoption. This scoping review included studies applying ML algorithms to adult OPAT patients, covering techniques from logistic regression to neural networks. Outcomes considered were medication intolerance, toxicity, catheter complications, hospital readmission and patient deterioration. A comprehensive search was conducted across databases including Cochrane Central, Cochrane Reviews, Embase, Ovid MEDLINE and Scopus, from 1 January 2000 to 1 January 2024. Thirty-two studies met the inclusion criteria, with the majority being single-centre experiences primarily from North America. Most studies focused on developing new ML models to predict outcomes such as hospital readmissions and medication-related complications. However, there was very little reporting on the performance characteristics of these models, such as specificity, sensitivity and C-statistics. There was a lack of multi-centre or cross-centre validation, limiting generalizability. Few studies advanced beyond traditional logistic regression models, and integration into clinical practice remains limited. ML shows promise for enhancing OPAT outcomes by predicting adverse events and enabling pre-emptive interventions. Despite this potential, significant gaps exist in development, validation and practical implementation. Barriers include the need for representative data sets and broadly applicable, validated models. Future research should address these barriers to fully leverage ML's potential in optimizing OPAT care and patient safety. Models must deliver timely, accurate and actionable insights to improve adverse event prediction and prevention in OPAT settings.

Beyond Ensemble Averages: Leveraging Climate Model Ensembles for Subseasonal Forecasting

Abstract Producing high-quality forecasts of key climate variables, such as temperature and precipitation, on subseasonal time scales has long been a gap in operational forecasting. This study explores an application of machine learning (ML) models as postprocessing tools for subseasonal forecasting. Lagged numerical ensemble forecasts (i.e., an ensemble where the members have different initialization dates) and observational data, including relative humidity, pressure at sea level, and geopotential height, are incorporated into various ML methods to predict monthly average precipitation and 2-m temperature 2 weeks in advance for the continental United States. For regression, quantile regression, and tercile classification tasks, we consider using linear models, random forests, convolutional neural networks, and stacked models (a multimodel approach based on the prediction of the individual ML models). Unlike previous ML approaches that often use ensemble mean alone, we leverage information embedded in the ensemble forecasts to enhance prediction accuracy. Additionally, we investigate extreme event predictions that are crucial for planning and mitigation efforts. Considering ensemble members as a collection of spatial forecasts, we explore different approaches to using spatial information. Trade-offs between different approaches may be mitigated with model stacking. Our proposed models outperform standard baselines such as climatological forecasts and ensemble means. In addition, we investigate feature importance, trade-offs between using the full ensemble or only the ensemble mean, and different modes of accounting for spatial variability. Significance Statement Accurately forecasting temperature and precipitation on subseasonal time scales—2 weeks–2 months in advance—is extremely challenging. These forecasts would have immense value in agriculture, insurance, and economics. Our paper describes an application of machine learning techniques to improve forecasts of monthly average precipitation and 2-m temperature using lagged physics-based predictions and observational data 2 weeks in advance for the entire continental United States. For lagged ensembles, the proposed models outperform standard benchmarks such as historical averages and averages of physics-based predictions. Our findings suggest that utilizing the full set of physics-based predictions instead of the average enhances the accuracy of the final forecast.

Home Blood Pressure Variability Risk Prediction Score for Cardiovascular Disease Using Data From the J-HOP Study.

Home blood pressure (BP) is more closely associated with cardiovascular event risk than office BP, but cardiovascular risk prediction based on home BP variability is lacking. This study developed a simple cardiovascular event prediction score, including home BP variability data, from the J-HOP study (Japan Morning Surge-Home Blood Pressure). The J-HOP study extended follow-up from December 2017 to May 2018 generated the study data set (4231 patients). Cardiovascular events included fatal/nonfatal stroke (n=94), coronary heart disease (n=124), heart failure (n=42), and aortic dissection (n=8). Cox proportional hazards models were used to predict overall cardiovascular risk. Potential covariates included age, sex, body mass index, smoking, history of diabetes, statin use, history of cardiovascular disease, total cholesterol:high-density lipoprotein cholesterol ratio, office systolic BP (SBP), mean of morning-evening average (MEave), home SBP, and average real variability of MEave home SBP. A risk score and models were constructed, and model performance was assessed. Model performance was best when average real variability of MEave SBP was included (C statistic, 0.760). The risk score assigns points for age (5-year bands), sex, cardiovascular disease history, high-density lipoprotein cholesterol, mean MEave home SBP, and average real variability of MEave home SBP. Estimated 10-year cardiovascular risk ranged from ≤0.6% (score ≤0) to >32% (score ≥26). Calibration 2 statistics values for the model (2.66) and risk score (5.29) indicated excellent goodness of fit. This simple cardiovascular disease prediction algorithm, including day-by-day home BP variability, could be used as part of a home BP-centered approach to hypertension management in clinical practice.

On the influence of discourse connectives on the predictions of humans and language models.

Psycholinguistic literature has consistently shown that humans rely on a rich and organized understanding of event knowledge to predict the forthcoming linguistic input during online sentence comprehension. We, the authors, expect sentences to maintain coherence with the preceding context, making congruent sentence sequences easier to process than incongruent ones. It is widely known that discourse relations between sentences (e.g., temporal, contingency, comparison) are generally made explicit through specific particles, known as discourse connectives, (e.g., and, but, because, after). However, some relations that are easily accessible to the speakers, given their event knowledge, can also be left implicit. The goal of this paper is to investigate the importance of discourse connectives in the prediction of events in human language processing and pretrained language models, with a specific focus on concessives and contrastives, which signal to comprehenders that their event-related predictions have to be reversed. Inspired by previous work, we built a comprehensive set of story stimuli in Italian and Mandarin Chinese that differ in the plausibility and coherence of the situation being described and the presence or absence of a discourse connective. We collected plausibility judgments and reading times from native speakers for the stimuli. Moreover, we correlated the results of the experiments with the predictions given by computational modeling, using Surprisal scores obtained via Transformer-based language models. The human judgements were collected using a seven-point Likert scale and analyzed using cumulative link mixed modeling (CLMM), while the human reading times and language model surprisal scores were analyzed using linear mixed effects regression (LMER). We found that Chinese NLMs are sensitive to plausibility and connectives, although they struggle to reproduce expectation reversal effects due to a connective changing the plausibility of a given scenario; Italian results are even less aligned with human data, with no effects of either plausibility and connectives on Surprisal.

Comprehensive Assessment of Lipid Markers in Cardiovascular Events Prediction.

Many studies have reported a relationship between various lipids, such as cholesterol, fatty acids, and lipoproteins, and cardiovascular events. Low-density lipoprotein cholesterol (LDL-C) is often cited as a representative marker. However, there is still room for discussion regarding which markers, among other lipids, should take clinical precedence.This observational study focused on patients without residual stenosis on post-coronary angiography. It was based on blood tests, including lipid profiles at that time, and assessed the association with the subsequent occurrence of major adverse cardiovascular events (MACE, a composite of all-cause mortality, hospitalization due to heart failure, myocardial infarction, stroke, and all revascularizations).Of the 375 patients analyzed, 134 experienced MACE (median follow-up duration: 1031 days). When comparing the MACE and non-MACE groups, significant differences were observed in lipid markers such as non-high-density lipoprotein cholesterol (non-HDL-C) and remnant-like particle cholesterol (RLP-C) (non-HDL-C; P = 0.003, RLP-C; P < 0.001). Furthermore, the area under the curve for RLP-C was 0.656 (95% CI: 0.598-0.714). Improvement in MACE risk discrimination was observed when LDL-C was replaced with non-HDL-C or RLP-C, in addition to atherosclerosis risk factors (non-HDL-C; net reclassification improvement (NRI) = 0.366, 95% CI: 0.159-0.572, RLP-C; NRI = 0.224, 95% CI: 0.016-0.433).It is highly likely that non-HDL-C and RLP-C can serve as significant lipid markers for predicting the occurrence of MACE.

Validation of machine learning-based risk stratification scores for patients with acute coronary syndrome treated with percutaneous coronary intervention

Abstract Aims This study aimed to validate the machine learning-based Global Registry of Acute Coronary Events (GRACE) 3.0 score and PRAISE (Prediction of Adverse Events following an Acute Coronary Syndrome) in patients with acute coronary syndrome (ACS) treated with percutaneous coronary intervention (PCI) for predicting mortality. Methods and results Data of consecutive patients with ACS treated with PCI in a tertiary centre in the Netherlands between 2014 and 2021 were used for external validation. The GRACE 3.0 score for predicting in-hospital mortality was evaluated in 2759 patients with non-ST-elevation acute coronary syndrome (NSTE-ACS) treated with PCI. The PRAISE score for predicting one-year mortality was evaluated in 4347 patients with ACS treated with PCI. Both risk scores were compared with the GRACE 2.0 score. The GRACE 3.0 score showed excellent discrimination [c-statistic 0.90 (95% CI 0.84, 0.94)] for predicting in-hospital mortality, with well-calibrated predictions (calibration-in-the large [CIL] −0.19 [95% CI −0.45, 0.07]). The PRAISE score demonstrated moderate discrimination [c-statistic 0.75 (95% CI 0.70, 0.80)] and overestimated the one-year risk of mortality [CIL −0.56 (95% CI −0.73, −0.39)]. Decision curve analysis demonstrated that the GRACE 3.0 score offered improved risk prediction compared with the GRACE 2.0 score, while the PRAISE score did not. Conclusion This study in ACS patients treated with PCI provides suggestive evidence that the GRACE 3.0 score effectively predicts in-hospital mortality beyond the GRACE 2.0 score. The PRAISE score demonstrated limited potential for predicting one-year mortality risk. Further external validation studies in larger cohorts including patients without PCI are warranted.

Medical dilemma: Programmed death 1 blockade (sintilimab) therapy in patients suffering from tumours combined with psoriasis.

Tumour immunotherapy represented by immune checkpoint inhibitors (ICIs) has greatly improved the overall prognosis of patients with malignant tumours, and is regarded as an important breakthrough in the field of medicine in recent years. ICIs have gradually become the core of tumour therapy and are increasingly used in the clinic. In order to achieve early clinical prediction and management of immune-related adverse events (irAEs), it is still necessary to perform further research on the mechanisms, risk factors, and predictors of irAE occurrence in the future. Zhou et al describe the consultation of a patient with advanced gastric cancer combined with chronic plaque psoriasis. This case provides an important reference for the use of programmed cell death protein-1 (PD-1) inhibitors in patients of tumours combined with chronic plaque psoriasis. This case also highlights that screening of high-risk groups for irAEs is critical before applying PD-1 inhibitors to patients with chronic psoriasis combined with tumours. PD-1 inhibitors are new and potent antineoplastic agents that can cause serious immune-related adverse events such as toxic epidermal necrolysis release and psoriasis. Glucocorticosteroids are the first-line agents for irAEs. The incidence of rheumatic irAEs may be higher in reality, which will inevitably become a new challenge for rheumatologists and dermatologists.

Urban Spatiotemporal Event Prediction Using Convolutional Neural Network and Road Feature Fusion Network

The security challenges faced by smart cities are attracting more attention from more people. Criminal activities and disasters can have a significant impact on the stability of a city, resulting in a loss of safety and property for its residents. Therefore, predicting the occurrence of urban events in advance is of utmost importance. However, current methods fail to consider the impact of road information on the distribution of cases and the fusion of information at different scales. In order to solve the above problems, an urban spatiotemporal event prediction method based on a convolutional neural network (CNN) and road feature fusion network (FFN) named CNN-rFFN is proposed in this paper. The method is divided into two stages: The first stage constructs feature map and structure of CNN then selects the optimal feature map and number of CNN layers. The second stage extracts urban road network information using multiscale convolution and incorporates the extracted road network feature information into the CNN. Some comparison experiments are conducted on the 2018–2019 urban patrol events dataset in Zhengzhou City, China. The CNN-rFFN method has an R2 value of 0.9430, which is higher than the CNN, CNN-LSTM, Dilated-CNN, ResNet, and ST-ResNet algorithms. The experimental results demonstrate that the CNN-rFFN method has better performance than other methods.

Coupling WRF with HEC-HMS and WRF-Hydro for flood forecasting in typical mountainous catchments of northern China

Abstract. The atmospheric–hydrological coupling systems are essential to flood forecasting because they allow for more improved and comprehensive prediction of flood events with an extended forecast lead time. Achieving this goal requires a reliable hydrological model system that enhances both rainfall predictions and hydrological forecasts. This study evaluates the potential of coupling the mesoscale numerical weather prediction model, i.e., the weather research and forecasting (WRF) model, with different hydrological modeling systems to improve the accuracy of flood simulation. The fully distributed WRF-Hydro modeling system and the semi-distributed Hydrological Engineering Center Hydrological Modeling System (HEC-HMS) were coupled with the WRF model, and the lumped HEC-HMS model was also adopted using the observed gauge precipitation as a benchmark to test the model uncertainty. Four distinct storm events from two mountainous catchments in northern China characterized by varying spatial and temporal rainfall patterns were selected as case studies. Comparative analyses of the simulated flooding processes were carried out to evaluate and compare the performance of the coupled systems with different complexities. The coupled WRF–HEC-HMS system performed better for long-duration storm events and obtained optimal performance for storm events uniformly distributed both temporally and spatially, as it adapted to more rapid recession processes of floods. However, the coupled WRF–HEC-HMS system did not adequately capture the magnitude of the storm events as it had a larger flow peak error. On the other hand, the fully distributed WRF–WRF-Hydro system performed better for shorter-duration floods with higher flow peaks as it can adapt to the simulation of flash floods. However, the performance of the system became poor as uniformity decreased. The performance of the lumped HEC-HMS indicates some source of uncertainty in the hydrological model when compared with the coupled WRF–HEC-HMS system, but a larger magnitude error was found in the WRF output rainfall. The results of this study can help establish an adaptive atmospheric–hydrologic coupling system to improve flood forecasting for different watersheds and climatic characteristics.