Dynamic Prediction Model Research Articles

Dynamic prediction model of segmented combined electromagnetic launch weft insertion for super-wide width industrial fabrics

Stable, reliable and high-speed weft insertion is an urgent challenge for high-speed industrial textile weaving devices. More stable and large electromagnetic forces can be provided by the segmented combined electromagnetic launch weft insertion, compared to the conventional electromagnetic launch weft insertion, which facilitates the realization of high-speed weft insertion. For digital electromagnetic launch weft insertion textile devices, it is necessary to respond quickly to the signals fed back by the sensors. Therefore, a fast and high-precision dynamical model capable of characterizing the kinematic properties of the weft inserter is extremely essential for control, which is conducive to the stable and reliable operation of the device. In this paper, we consider the electromagnetic force, weft yarn tension, air resistance, pipe wall friction, and gravity during the movement of the weft inserter, and establish a dynamic prediction model of segmented combined electromagnetic launch weft insertion (DPMSCEI) for control. Combining simulations and experiments to evaluate the performance of DPMSCEI validates the effectiveness and professionalism of DPMSCEI in solving electromagnetic weft insertion dynamics.

Two web-based dynamic prediction models for the diagnosis and prognosis of gastric cancer with bone metastases: evidence from the SEER database.

Our aim was to identify the clinical characteristics and develop and validate diagnostic and prognostic web-based dynamic prediction models for gastric cancer (GC) with bone metastasis (BM) using the SEER database. Our study retrospectively analyzed and extracted the clinical data of patients aged 18-85 years who were diagnosed with gastric cancer between 2010 and 2015 in the SEER database. We randomly divided all patients into a training set and a validation set according to the ratio of 7 to 3. Independent factors were identified using logistic regression and Cox regression analyses. Furthermore, we developed and validated two web-based clinical prediction models. We evaluated the prediction models using the C-index, ROC, calibration curve, and DCA. A total of 23,156 patients with gastric cancer were included in this study, of whom 975 developed bone metastases. Age, site, grade, T stage, N stage, brain metastasis, liver metastasis, and lung metastasis were identified as independent risk factors for the development of BM in GC patients. T stage, surgery, and chemotherapy were identified as independent prognostic factors for GC with BM. The AUCs of the diagnostic nomogram were 0.79 and 0.81 in the training and test sets, respectively. The AUCs of the prognostic nomogram at 6, 9, and 12 months were 0.93, 0.86, 0.78, and 0.65, 0.69, 0.70 in the training and test sets, respectively. The calibration curve and DCA showed good performance of the nomogram. We established two web-based dynamic prediction models in our study. It could be used to predict the risk score and overall survival time of developing bone metastasis in patients with gastric cancer. In addition, we also hope that these two web-based applications will help physicians comprehensively manage gastric cancer patients with bone metastases.

Analysis of prevailing pertussis strains from different regions in China after the application of acellular vaccines.

Analysis of prevailing pertussis strains from different regions in China after the application of acellular vaccines.

Physics-guided LSTM model for heat load prediction of buildings

The thermal process of heating buildings is characterized by the multivariable, nonlinear, and large time lag. Accurately predicting building heat loads is crucial for regulating building heating systems. However, the current black-box method of heat load prediction has challenges such as extensive data usage and weak integration with the building's heat transfer process. To address these challenges, a physics-guided long short-term memory (LSTM) model for dynamic heat load prediction in the next seven days based on data-driven and mechanism analysis was proposed. The data for the model were collected from in-situ measurement and numerical simulation. Specifically, the thermal mechanism was introduced to the training loss function of LSTM and the selection of input variables. The influence of the adaptive data time step on prediction accuracy was investigated. The main conclusions obtained from the current data in this paper are drawn as follows. 1) Refining solar radiation intensity calculation for each orientation reduced mean absolute percentage error (MAPE) of heat load prediction results by 0.02%–1.8% compared to horizontal solar radiation intensity. 2) Improving the training loss function decreased MAPE of heat load prediction results by 0.04%–0.43% compared with the original loss function. 3) Adaptive data time step lowered the MAPE of heat load prediction results by 0.08%–1.8% compared with fixed time step. 4) Mean absolute errors corresponding to those three aspects decreased by 25 kJ–558 kJ, 6 kJ–157 kJ, and 41 kJ–1881 kJ. In the future, wider promotion and application of the prediction model will lead to better performance and be more beneficial to energy saving and carbon reduction of the heating system.

SIPN South: six years of coordinated seasonal Antarctic sea ice predictions

Antarctic sea ice prediction has garnered increasing attention in recent years, particularly in the context of the recent record lows of February 2022 and 2023. As Antarctica becomes a climate change hotspot, as polar tourism booms, and as scientific expeditions continue to explore this remote continent, the capacity to anticipate sea ice conditions weeks to months in advance is in increasing demand. Spurred by recent studies that uncovered physical mechanisms of Antarctic sea ice predictability and by the intriguing large variations of the observed sea ice extent in recent years, the Sea Ice Prediction Network South (SIPN South) project was initiated in 2017, building upon the Arctic Sea Ice Prediction Network. The SIPN South project annually coordinates spring-to-summer predictions of Antarctic sea ice conditions, to allow robust evaluation and intercomparison, and to guide future development in polar prediction systems. In this paper, we present and discuss the initial SIPN South results collected over six summer seasons (December-February 2017-2018 to 2022-2023). We use data from 22 unique contributors spanning five continents that have together delivered more than 3000 individual forecasts of sea ice area and concentration. The SIPN South median forecast of the circumpolar sea ice area captures the sign of the recent negative anomalies, and the verifying observations are systematically included in the 10-90% range of the forecast distribution. These statements also hold at the regional level except in the Ross Sea where the systematic biases and the ensemble spread are the largest. A notable finding is that the group forecast, constructed by aggregating the data provided by each contributor, outperforms most of the individual forecasts, both at the circumpolar and regional levels. This indicates the value of combining predictions to average out model-specific errors. Finally, we find that dynamical model predictions (i.e., based on process-based general circulation models) generally perform worse than statistical model predictions (i.e., data-driven empirical models including machine learning) in representing the regional variability of sea ice concentration in summer. SIPN South is a collaborative community project that is hosted on a shared public repository. The forecast and verification data used in SIPN South are publicly available in near-real time for further use by the polar research community, and eventually, policymakers.

Signal-control refined dynamic traffic graph model for movement-based arterial network traffic volume prediction

Forecasting the forthcoming intersection movement-based traffic volume enables adaptive traffic control systems to dynamically respond to the fluctuation of traffic demands. In this paper, a deep-learning based Signal-control Refined Dynamic Traffic Graph (ScR-DTG) Model is proposed for advancing the network-level movement-based traffic volume prediction task. The proposed model attempts to further improve the-state-of-art and practice algorithms in traffic prediction for arterial network adaptive signal control utilizing tradition traffic flow theory boosted deep-learning methodology. For precisely inferencing the movement-based demand at cycle-to-cycle level, the proposed model incorporates spatial graph convolution inferencing layer and temporal inferencing layer to explore both the intricate spatial temporal dependencies, respectively. A signal control refining module is contrived to deduce the controlled movement saturation flow and introduce the essential control inferences, which is of great significance but frequently neglected in the previous researches. Additionally, according to the real-time movement specified travel time, this paper creatively constructs an adjacent graph with dynamic order for more accurately capturing the ever-changing spatial relevancies. Field experiments with multiple signal schemes were conducted in the downtown area of Zhangzhou (China). The promising results demonstrated the state-of-the-art accuracy than other high-performance volume prediction algorithms. Implementing the proposed model enables to obtain accurate movement-based volume predictions, which would assist the traffic management agencies in adjusting signal timing adaptively and further improve the efficiency of signal intersection.

A dynamic life cycle assessment model for long-term carbon emissions prediction of buildings: A passive building as case study

Carbon emissions of buildings in the life cycle are huge due to the high resource-intensive and energy consumption of the buildings industry and its upstream industries. And for carbon emissions prediction of the buildings, the life cycle assessment (LCA) with the static parameters input was used to obtain long-term carbon emissions for the buildings in the existed studies. The study presented a dynamic LCA model to predict life cycle carbon emissions of the buildings and optimize the insulation parameter with the improved temporal validation of upstream parameters. A case passive building was selected to perform the dynamic method based on the basic and operational information of this building. The key temporal parameters were collected and predicted via grey prediction. And the electric mix structure was predicted via scenario analysis method. The dynamic life cycle carbon emissions of energy systems were evaluated in different scenarios by the combination of predicting results and historical data. The life cycle carbon emissions of the case building in temporal scenarios reduced approximately 23.4% of that in static scenario. Also, optimal insulation layer thickness in temporal scenarios of and that in static scenario were 130 mm, 120 mm and 250 mm, respectively.

Yaw Optimisation for Wind Farm Production Maximisation Based on a Dynamic Wake Model

In recent years, a major focus on wind farm wake control is to maximise the production of wind farms. To improve the power generation efficiency of wind farms through wake regulation, this study investigates yaw optimisation for wind farm production maximisation from the perspective of time-varying wakes. To this end, we first deduce a simplified dynamic wake model according to the momentum conservation theory and backward difference method. The accuracy of the proposed model is verified by simulation comparisons. Then, the time lag of wake propagation and its impact on wind farm production maximisation through wake meandering is analysed. On this basis, a yaw optimisation method for increasing wind farm energy capture is presented. This optimisation method uses the proposed dynamic wake model for wind farm prediction. The results indicate that the optimisation period is critical to the effect of the optimisation method on wind farm energy capture.

A Methodology to Evaluate the Vulnerability of the Natural Gas Supply Chain Based on Set Pair Analysis and Markov Chain

In this study, a methodology to evaluate and predict the vulnerability of the natural gas supply chain is proposed, and both the dynamics and stochasticity of the natural gas supply chain are considered. In the methodology, the evaluation system of the vulnerability is established firstly based on the definition of the vulnerability of the natural gas supply chain, and the evaluation system is composed of 12 evaluation indicators that cover the upstream, midstream, and downstream of the natural gas supply chain. Then, a combination weighting method based on the analytic hierarchy process method and entropy weighting method is employed to determine the weight of each indicator. Finally, the dynamic evaluation and prediction models of the vulnerability of the natural gas supply chain are developed based on the set pair analysis and Markov chain. Furthermore, the methodology is applied to evaluate and predict the vulnerability of the natural gas supply chain in China. According to the evaluation results, the vulnerability level in the year of 2021 is relatively vulnerable, and the natural gas supply chain in China was relatively stable in the last 10 years, but it is developing toward a more vulnerable trend. Moreover, some suggestions to reduce the vulnerability of China’s natural gas supply chain are proposed as well.

A visualized dynamic prediction model for overall survival in patients diagnosed with brain metastases from lung squamous cell carcinoma.

Patients presenting with brain metastases (BMs) from lung squamous cell carcinoma (LUSC) often encounter an extremely poor prognosis. A well-developed prognostic model would assist physicians in patient counseling and therapeutic decision-making. Patients with LUSC who were diagnosed with BMs between 2000 and 2018 were reviewed in the Surveillance, Epidemiology, and End Results (SEER) database. Using the multivariate Cox regression approach, significant prognostic factors were identified and integrated. Bootstrap resampling was used to internally validate the model. An evaluation of the performance of the model was conducted by analyzing the area under the curve (AUC) and calibration curve. A total of 1812 eligible patients' clinical data was retrieved from the database. Patients' overall survival (OS) was significantly prognosticated by five clinical parameters. The nomogram achieved satisfactory discrimination capacity, with 3-, 6-, and 9-month AUC values of 0.803, 0.779, and 0.760 in the training cohort and 0.796, 0.769, and 0.743 in the validation cohort. As measured by survival rate probabilities, the calibration curve agreed well with actual observations. There was also a substantial difference in survival curves between the different prognostic groups stratified by prognostic scores. For ease of access, the model was deployed on a web-based server. In this study, a nomogram and a web-based predictor were developed to assist physicians with personalized clinical decisions and treatment of patients who presented with BMs from LUSC.

Dynamic nomogram prediction model for diabetic retinopathy in patients with type 2 diabetes mellitus

BackgroundTo develop a dynamic prediction model for diabetic retinopathy (DR) using systemic risk factors.MethodsThis retrospective study included type 2 diabetes mellitus (T2DM) patients discharged from the Second Affiliated Hospital of Kunming Medical University between May 2020 and February 2022. The early patients (80%) were used for the training set and the late ones (20%) for the validation set.ResultsFinally, 1257 patients (1049 [80%] in the training set and 208 [20%] in the validation set) were included; 360 (28.6%) of them had DR. The areas under the curves (AUCs) for the multivariate regression (MR), least absolute shrinkage and selection operator regression (LASSO), and backward elimination stepwise regression (BESR) models were 0.719, 0.727, and 0.728, respectively. The Delong test showed that the BESR model had a better predictive value than the MR (p = 0.04899) and LASSO (P = 0.04999) models. The DR nomogram risk model was established according to the BESR model, and it included disease duration, age at onset, treatment method, total cholesterol, urinary albumin to creatinine ratio (UACR), and urine sugar. The AUC, kappa coefficient, sensitivity, specificity, and compliance of the nomogram risk model in the validation set were 0.79, 0.48, 71.2%, 78.9%, and 76.4%, respectively.ConclusionsA relatively reliable DR nomogram risk model was established based on the BESR model.

TANGO: A temporal spatial dynamic graph model for event prediction

The event prediction task learns the dynamic changes of past events and infers the upcoming ones. This task helps people to understand how events evolve in our real world and make rapid, accurate, and efficient reactions in emergencies. The complex relation of events and entities (events participants) changes dynamically over time. The current solutions of event prediction have not fully exploited such event temporal dependency and entity relation dependency. In addition, an event may relate to a considerable number of entities. Among such related entities, only a few of them are the key participants in driving the evolution of the event. Previous works fail to emphasize such key entities and limit the non-related entities.In this paper, we introduce a novel gating and attention mechanism and propose a novel Temporal spAtial dyNamic Graph mOdel (TANGO) that is composed of a graph model (based on Graph Convolutional Network with gated and attention mechanisms) and a sequential model (based on Temporal Convolutional Network). TANGO is able to model the event temporal dependency and entity relation dependency simultaneously; learn the representation of events relations; control the aggregation of relational information among entities with varying degrees of mutual influence; support long effective historical sizes. We demonstrate the validity and effectiveness of our approach on three different datasets (i.e., ICEWS18, GDELT, and ICEWS14). The result shows almost 6.2% MRR improvement and 3.9% Hits@10 improvement over the previous state-of-the-art.

Benefits of Dynamic Nomogram Models for Elderly Diffuse Large B-Cell Lymphoma Patients' Early Death Prediction and Clinical Application.

Diffuse large B-cell lymphoma (DLBCL) is an aggressive malignancy, and about 60% of the patients are diagnosed in their elderly age (≥65 years old). However, little is known about the early mortality and risk factors related to elderly patients with DLBCL. Methodology. From 2000 to 2019, elderly patients diagnosed with DLBCL in the Surveillance, Epidemiology, and End Result (SEER) database were involved in this research and served as test cohort. Moreover, elderly DLBCL patients from Peking University Third Hospital were used for external validation cohort. Risk factors were identified by univariate and multivariate logistic regression analyses. Nomogram models were constructed based on significance risk factors to predict the overall and cancer-specific early death. Besides that, the predictive value of the models was validated by receiver operating characteristic (ROC) analysis. Calibration plots were used to evaluate the calibrating ability. Clinical benefits of nomogram were evaluated by decision curve analysis (DCA). 15242 elderly DLBCL patients obtained from the SEER database and 152 patients from Peking University Third Hospital were enrolled in this research. In the SEER database, 36.6% (5584/15242) of the patients had early death and 30.7% (4680/15242) of them were cancer-specific early death. Marital status, Ann Arbor stage, surgical treatment, radiotherapy, and chemotherapy were significant risk factors for overall and cancer-specific early death of elderly DLBCL patients. Nomograms were constructed according to these risk factors. Then, ROC analysis showed that the AUC of OS was 0.764 (0.756~0.772), and CSS was 0.742 (0.733~0.751). In the validation group, the AUC of OS was 0.767 (0.689~0.846) and CSS was 0.742 (0.743~0.83). The calibration plots and DCA analysis revealed that the nomograms were good at early death prediction and clinical application. Predictive dynamic nomogram models for elderly DLBCL patients were established and validated, which might play an essential role in helping physicians enact better treatment strategies.

Study on Dynamic Modulus and Damping Characteristics of Modified Expanded Polystyrene Lightweight Soil under Cyclic Load.

In recent years, expanded polystyrene (EPS) lightweight soil has been widely used as subgrade in soft soil areas because of its light weight and environmental protection. This study aimed to investigate the dynamic characteristics of sodium silicate modified lime and fly ash treated EPS lightweight soil (SLS) under cyclic loading. The effects of EPS particles on the dynamic elastic modulus (Ed) and damping ratio (λ) of SLS were determined through dynamic triaxial tests at various confining pressures (σ3), amplitudes, and cycle times. Mathematical models of the Ed of the SLS, cycle times, and σ3 were established. The results revealed that the EPS particle content played a decisive role in the Ed and λ of the SLS. The Ed of the SLS decreased with an increase in the EPS particle content (EC). The Ed decreased by 60% in the 1-1.5% range of the EC. The existing forms of lime fly ash soil and EPS particles in the SLS changed from parallel to series. With an increase in σ3 and amplitude, the Ed of the SLS gradually decreased, the λ generally decreased, and the λ variation range was within 0.5%. With an increase in the number of cycles, the Ed of the SLS decreased. The Ed value and the number of cycles satisfied the power function relationship. Additionally, it can be found from the test results that 0.5% to 1% was the best EPS content for SLS in this work. In addition, the dynamic elastic modulus prediction model established in this study can better describe the varying trend of the dynamic elastic modulus of SLS under different σ3 values and load cycles, thereby providing a theoretical reference for the application of SLS in practical road engineering.

Monthly streamflow forecasting by machine learning methods using dynamic weather prediction model outputs over Iran

Seasonal hydrological forecasts play a critical role in water resources management. The Copernicus Climate Change Service (C3S) data store provides open access to monthly hydrological forecasts for up to six-months. This study aims to evaluate, for the first time, 1- to 3-month runoff forecasts using the European Centre for Medium-Range Weather Forecasts (ECMWF) ensembles of precipitation, runoff, and temperature in 1981–2015 period over a total of 30 s-level basins in Iran. We adopted the 5th, 50th and 95th ECMWF ensemble quantiles for each variable that represent low, medium and high probability of occurrence, respectively. Pearson correlation analysis (Pca), Recursive Feature Elimination (RFE) via random forest (RF) model, and Bayesian Networks (BN) feature selection algorithms were used in order to reduce input variable dimension and select potential predictors to be fed to the machine learning models. Multiple Linear Regression (MLR), Artificial Neural Networks (ANN), Support Vector Regression (SVR), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost) machine learning models were used with Repeated K-Fold cross validation (rK-Fold CV) while model efficiency was evaluated using modified Kling-Gupta efficiency coefficient (KGE’), Nash-Sutcliffe Efficiency coefficient (NSE), and Normalized Root Mean Square Error (NRMSE). Results of this study revealed that C3S runoff ensembles have the highest impact on forecast accuracy of streamflow, followed by precipitation and temperature. Overall, model performance yield a best-to-worst ranking of ANN, XGBoost, RF, MLR, and SVR with KGE’ values of 0.70, 0.68, 0.66, 0.57, and 0.41, respectively. The predictive performance of all models decreased with lead times beyond 1-month, where ANN and XGBoost outperformed other models with KGE’ of 0.65 for 2-month lead time and 0.60 for 3-month lead time. The three superior models of XGBoost, ANN, and RF, were employed with RFE and BN FSAs most frequently across Iran’s 30 s level basins in all lead times. Almost all models in the arid central region of Iran showed the lowest performance while highest skills were achieved in the western regions of Iran. Finally, for all models and over all regions, the model performance reduced by increase in lead-time.

Dynamic compact thermal models for skin temperature prediction of portable electronic devices based on convolution and fitting methods

In most portable electronic devices, besides the temperature of multiple heat sources, i.e. junction temperature, the temperature of the enclosure, i.e. skin temperature, should also be controlled to protect the user experience. Thus, generating the device-level compact thermal model for predicting the skin temperature will not only improve the efficiency of early-stage thermal design but also help to develop the model-based temperature control strategy. However, most of the existing modeling methods mainly focus on predicting the junction temperature. In this paper, the dynamic compact thermal models of two portable electronic devices, including a smartphone and laptop, are first generated based on the convolution method. Under the assumption of linear time-invariant (LTI) systems, the skin temperature of the two test devices could be fast calculated once the step response of each heat source is obtained. The results show that compared to the computational fluid dynamics (CFD) model generated in ANSYS Icepak, the transient deviation of the convolution-based model is within 5%. Then, a linear fitting model is proposed to real-time predict the skin temperature of laptops. Both the experimental and simulation results demonstrate that once the scale factors are trained, the linear fitting model could accurately real-time predict the skin temperature after 60s. The maximum transient deviation in the experiment and simulation are 4% and 7% respectively. The results indicate that the proposed modeling method has tremendous potential for both thermal design and control optimization.

Digital twin–based dynamic prediction and simulation model of carbon efficiency in gear hobbing process

Digital twin–based dynamic prediction and simulation model of carbon efficiency in gear hobbing process

Path Planning and Obstacle Avoidance Control for Autonomous Multi-Axis Distributed Vehicle Based on Dynamic Constraints

Obstacle avoidance problem for the autonomous vehicle has received more and more attention these years, especially for the autonomous heavy vehicle, which has the high gravity center, large carrying capacity and promising commercial prospect. However, the vehicle dynamic properties have been rarely considered for path planning, which is a crucial factor for driving safety. An integrated collision avoidance framework that consists of a new path planner and a tracking controller considering multiple dynamic constraints is proposed for an autonomous multi-axis distributed vehicle. In the upper level, a new optimal path generator based on the B-splines is designed for safety and stability purposes, and the collision avoidance path is optimized considering multiple dynamic constraints, including the tire nonlinear dynamic, lateral stability limits and the anti-slip constraints. Meanwhile, a safety assessment evaluator according to the time to collision calculation is designed for risk level recognition and path planning weight determination. In the lower level, a path tracking controller based on the nonlinear model predictive control (NMPC) is designed to improve the control accuracy, and a high-fidelity nonlinear dynamic predictive vehicle model with 16 degrees of freedom is established to depict the vehicle dynamic properties comprehensively. To solve the nonlinear optimization problem and get the optimal control steering angle output, an adaptive weight adjustment strategy and a varying predictive duration method are formulated. The optimization constraints related to steering actuator ability, plane stability limits, rollover prevention thresholds and tire adhesion requirements are constructed and converted as the bounds of the vehicle dynamic states. To verify the proposed framework, the simulation and HIL test platform are built and different driving scenarios are validated. The test results show the satisfactory performance of path planning and path tracking considering the dynamic constraints for safety and stability

Research on the method of searching and tracking of the time-sensitive target through the mega-constellation

The time-sensitive target has the characteristics of high-speed motion and an uncertain model, which present challenges to the observation process. Aiming at the problem of searching and tracking the time-sensitive target in a region, the scheduling method of the mega-constellation based on grids coding is proposed in this paper. Above all, the observation problem is assessed, after which the grids coding model of observation resources and tasks is constructed. Afterwards, the dynamic updating process of probability is established as a Markov state transition, taking into consideration the movement of the time-sensitive target during the searching phase, and the dynamic prediction model of target position for target tracking is founded, which takes into account the unpredictability of the time-sensitive target model. Ultimately, the mega-constellation searching technique is developed with the enhancement of information entropy of covering grids as the objective function. Based on the optimal acquisition probability of grids, a method for determining the field-of-view center grid code is established to realize cooperative tracking. It is proved by experiments that, the continuous tracking of time-sensitive targets can be realized and the percentage of searches achieved within half an hour is greater than 92%, which is 10% to 35% higher than the conventional methods.

A machine learning model for visualization and dynamic clinical prediction of stroke recurrence in acute ischemic stroke patients: A real-world retrospective study.

Recurrent stroke accounts for 25-30% of all preventable strokes, and this study was conducted to establish a machine learning-based clinical predictive rice idol for predicting stroke recurrence within 1 year in patients with acute ischemic stroke (AIS). A total of 645 AIS patients at The Second Affiliated Hospital of Xuzhou Medical University were screened, included and followed up for 1 year for comprehensive clinical data. Univariate and multivariate logistic regression (LR) were used to screen the risk factors of stroke recurrence. The data set was randomly divided into training set and test set according to the ratio of 7:3, and the following six prediction models were established by machine algorithm: random forest (RF), Naive Bayes model (NBC), decision tree (DT), extreme gradient boosting (XGB), gradient boosting machine (GBM) and LR. The model with the strongest prediction performance was selected by 10-fold cross-validation and receiver operating characteristic (ROC) curves, and the models were investigated for interpretability by SHAP. Finally, the models were constructed to be visualized using a web calculator. Logistic regression analysis showed that right hemisphere, homocysteine (HCY), C-reactive protein (CRP), and stroke severity (SS) were independent risk factors for the development of stroke recurrence in AIS patients. In 10-fold cross-validation, area under curve (AUC) ranked from 0.777 to 0.959. In ROC curve analysis, AUC ranged from 0.887 to 0.946. RF model has the best ability to predict stroke recurrence, and HCY has the largest contribution to the model. A web-based calculator https://mlmedicine-re-stroke2-re-stroke2-baylee.streamlitapp.com/ has been developed accordingly. This study identified four independent risk factors affecting recurrence within 1 year in stroke patients, and the constructed RF-based prediction model had good performance.