Prediction Of Dynamical Systems Research Articles

Artificial Intelligence Based Flood and Landslide Disaster Monitoring and Dynamic Numerical Prediction System

Monitoring and dynamic numerical prediction of flood and landslide disasters are of great significance for reducing disaster risks, ensuring urban safety and stable development. The existing systems have limitations in real-time and dynamic performance, making it difficult to provide effective support for disaster prevention and management. To improve the efficiency of disaster monitoring, this article combined artificial intelligence (AI) to conduct in-depth research on the construction of flood and landslide disaster monitoring and dynamic numerical prediction systems. This article first analyzed the system functional requirements, and then designed the system architecture based on this, dividing it into three levels: user interface layer, application layer, and data layer. Finally, the BP (Back Propagation) neural network algorithm and dynamic numerical model were utilized to analyze the monitoring and dynamic numerical prediction of flood and landslide disasters. To verify the effectiveness of the system, this article compared it with GIS (Geographic Information System) based methods, and conducted testing and analysis on the system from monitoring errors, real-time monitoring, numerical simulation, and several levels. The results showed that in terms of real-time monitoring, compared with GIS based disaster prediction systems, the average response time of the AI based disaster monitoring and prediction system in this paper was shortened by 1.36 milliseconds. The conclusion indicated that the AI-based flood and landslide disaster monitoring and dynamic numerical prediction system in this article could effectively improve the efficiency of disaster monitoring and early warning, and help promote the efficient and intelligent development of disaster prevention

Research on Dynamic Analysis and Prediction System of Energy Economic Efficiency under Fusion Algebraic Equations

Energy is an important material basis for national economic development, and it plays a vital role in social development. As China's industrialization process and urbanization construction continues to accelerate, energy demand is also increasing, and the dependence on electricity is getting higher and higher. How to improve the electric energy efficiency under the guarantee of stable quality of power generation, transmission and distribution and meeting the requirements of users' electricity consumption is an important issue to be solved in the current power grid construction. However, the serious defects of the traditional binary hierarchy and the lack of scientific understanding have led to many current new problems with more and more new and complex features. This is accompanied by the emergence of a large number of nonlinear term coefficients and an increasing research workload, and the traditional linear programming methods cannot meet the practical needs. Therefore, this paper analyzes the eigenvalues of energy economic efficiency based on algebraic equations and predicts energy economic efficiency using EMD-BP neural network algorithm, and then carries out the design and research of economic energy system based on these two points.

Enhancing water quality prediction for fluctuating missing data scenarios: A dynamic Bayesian network-based processing system to monitor cyanobacteria proliferation

Tackling the impact of missing data in water management is crucial to ensure the reliability of scientific research that informs decision-making processes in public health. The goal of this study is to ascertain the root causes associated with cyanobacteria proliferation under major missing data scenarios. For this purpose, a dynamic missing data management methodology is proposed using Bayesian Machine Learning for accurate surface water quality prediction of a river from Limia basin (Spain). The methodology used entails a sequence of analytical steps, starting with data pre-processing, followed by the selection of a reliable dynamic Bayesian missing value prediction system, leading finally to a supervised analysis of the behavioral patterns exhibited by cyanobacteria. For that, a total of 2,118,844 data points were used, with 205,316 (9.69 %) missing values identified. The machine learning testing showed the iterative structural expectation maximization (SEM) as the best performing algorithm, above the dynamic imputation (DI) and entropy-based dynamic imputation methods (EBDI), enhancing in some cases the accuracy of imputations by approximately 50 % in R2, RMSE, NRMSE, and logarithmic loss values. These findings can impact how data on water quality is being processed and studied, thus, opening the door for more reliable water management strategies that better inform public health decisions.

Prediction of Tropical Cyclogenesis Based on Machine Learning Methods and Its SHAP Interpretation

AbstractThis study trains three machine learning models with varying complexity—Random Forest, Support Vector Machine, and Neural Network—to predict cyclogenesis at a forecast lead time of 24 hr for given tropical disturbances identified by an optimized Kalman Filter algorithm. The overall performance is competent in terms of f1‐scores (∼0.8) compared to previous research of the same kind. An assessment by SHapley Additive exPlanations (SHAP) values reveals that mid‐level (500 hPa) vorticity is the most influential factor in deciding if a tropical disturbance is developing or non‐developing for all three models. Wind shear and tilting are found to hold a certain level of importance as well. These results encourage further experiments that use physical models to explore the dynamical, mid‐level pathway to tropical cyclogenesis. Another usage of SHAP values in this work is to explain how a machine learning model decides if an individual tropical disturbance case will develop, by listing the contribution of each feature to the output genesis probability, illustrated by a case study of Typhoon Halong. This increases the reliability of the machine learning models, and forecasters can take advantage of such information to issue tropical cyclone formation warnings more accurately. Several caveats of the current machine learning application in the studies of tropical cyclogenesis are discussed and can be considered for future research. These can benefit the interpretation and emphasis of certain output fields in the operational dynamical prediction system, which can contribute to more timely cyclogenesis forecasts.

Subseasonal-to-seasonal prediction of arctic sea ice Using a Fully Coupled dynamical ensemble forecast system

With the opening of Arctic passage for maritime transportation under global warming, accurate prediction of Arctic sea ice (ASI) on subseasonal-to-seasonal (S2S) timescales becomes increasingly crucial for the economy and society but challenging. This study examined the S2S hindcast skills of monthly ASI during 1992–2019 using Flexible Global Ocean-Atmosphere-Land System, Finite-Volume version 2 (FGOALS-f2), a global coupled model that includes an interactive dynamical sea ice component. Firstly, the prediction system can accurately capture the seasonal cycle of Arctic sea ice extent (SIE). Still, it slightly overestimates SIE in March–June and significantly underestimates SIE in August–December. Meanwhile, the declining trend of SIE is underestimated in predictions for all target seasons, particularly for spring. Secondly, relatively high skills with an average ACC of 0.44 are found in predicting monthly SIE anomalies (SIEA) after removing the long-term linear trend at a one-to-six-month lead. However, the one-to-six-month lead prediction skills in April drop significantly by 36% when the interdecadal variation is removed. We find that the variation of April SIEA after removing the linear trend is dominated by the Pacific Decadal Oscillation (PDO), which can be better predicted than interannual variability ahead of subseasonal timescales. Unlike the SIE error-based metrics considered in previous studies, the one-to-three-month lead prediction in FGOALS-f2 shows significantly better performance in predicting the spatial distribution of September ASI for extreme years compared to normal years. Finally, the skill for predicting minimum SIEA in FGOALS-f2 is considerably higher at a three-to-six-month lead than anomaly persistence, indicating the advantage of this dynamical prediction system at a longer lead time.

FORECAST VERIFICATION OF AN ENSEMBLE TROPICAL CYCLONE STORM SURGE SYSTEM

Tropical Cyclone (TC) induced storm surges can be especially damaging and require emergency preparations. A unique challenge of providing TC storm surge advice to emergency services relates to the sensitivity of storm surge forecasts to uncertainty in the TC forecast. Probabilistic forecast products are a good fit for such advice, as long as forecast skill characteristics are known and understood. The Australian Bureau of Meteorology operates a dynamical ensemble prediction system to provide forecasts of storm surge driven by TCs. This is a step change from the legacy parametric and/or scenario-based techniques previously used at the Bureau. There are a number of benefits to be gained from dynamic storm surge ensemble forecasts over more traditional systems. Most importantly, the surge forecast is very sensitive to errors in the TC location, velocity, intensity, and size. Ensemble storm surge forecasting is required in order to take account of this uncertainty in the TC forecast.

Predictability of the East Africa long rains through Congo zonal winds

AbstractEast Africa is highly vulnerable to extreme weather events, such as droughts and floods. Skillful seasonal forecasts exist for the October–November–December short rains, enabling informed decisions, whereas seasonal forecasts for the March–April–May (MAM) long rains have historically had low skill, limiting preparation capacity. Therefore, improved long rains prediction is a high priority and would contribute to climate change resilience in the region. Recent work has highlighted how lower‐troposphere Congo zonal winds in MAM strongly impact regional moisture fluxes and the long rains total precipitation. We therefore approach long rains predictability through the predictability of the Congo winds. We analyze a set of hindcasts from a dynamical prediction system that is able to reproduce the long rains—Congo winds relationship in its individual ensemble members. Encouragingly, in observations, the strength of MAM Congo zonal winds and East Africa rainfall show substantial correlation with the MAM Atlantic (including North Atlantic Oscillation, NAO) and Indo‐Pacific variability, suggestive of ocean influence and potential predictability. However, these features are replaced by different teleconnections in the hindcast ensemble mean fields. This is also true for NAO linkage to Congo winds, despite correct representation in individual members, and good skill in hindcasting the NAO itself. The net effect is strongly negative skill for the Congo winds. We explore statistical correction methods, including using the Congo zonal wind as an anchor index in a signal‐to‐noise calibration for the long rains. This is considered a demonstration of concept, for subsequent implementation using models with better Congo zonal wind skill. Indeed, the clear signals found in the Atlantic (including Mediterranean) and Indo‐Pacific, studied here both in observations and a dynamical prediction system, motivate evaluation of these features across other prediction systems, and offer the prospect of improved physically‐informed long rains dynamical predictions.

GEFSv12 High- and Low-Skill Day-10 Tornado Forecasts

Abstract On average, modern numerical weather prediction forecasts for daily tornado frequency exhibit no skill beyond day 10. However, in this extended-range lead window, there are particular model cycles that have exceptionally high forecast skill for tornadoes because of their ability to correctly simulate the future synoptic pattern. Here, model initial conditions that produced a more skillful forecast for tornadoes over the United States were exploited while also highlighting potential causes for low-skill cycles within the Global Ensemble Forecasting System, version 12 (GEFSv12). There were 88 high-skill and 91 low-skill forecasts in which the verifying day-10 synoptic pattern for tornado conditions revealed a western U.S. thermal trough and an eastern U.S. thermal ridge, a favorable configuration for tornadic storm occurrence. Initial conditions for high skill forecasts tended to exhibit warmer sea surface temperatures throughout the tropical Pacific Ocean and Gulf of Mexico, an active Madden–Julian oscillation, and significant modulation of Earth-relative atmospheric angular momentum. Low-skill forecasts were often initialized during La Niña and negative Pacific decadal oscillation conditions. Significant atmospheric blocking over eastern Russia—in which the GEFSv12 overforecast the duration and characteristics of the downstream flow—was a common physical process associated with low-skill forecasts. This work helps to increase our understanding of the common causes of high- or low-skill extended-range tornado forecasts and could serve as a helpful tool for operational forecasters. Significance Statement This research provides a framework for the anticipation of a more (or less) skillful 10-day tornado forecast in an operational numerical weather prediction system. High-skill forecasts were associated with substantial tropical convection and warm sea surface temperature throughout the Pacific Ocean and Gulf of Mexico, whereas the underlying cause of low-skill forecasts were typically associated with a blocking anticyclone over eastern Russia. These findings are important because they permit increased or decreased confidence in a long-range forecast of tornado occurrence based on a dynamical prediction system.

Dynamic prediction of displacement and deformation of any point on mining surface based on B-normal model.

The surface displacement and deformation of goaf caused by coal mining destroy the underground rock structure and surface ecological environment in the mining area and endanger the safety of human life and property. An accurate and efficient dynamic prediction system of mining subsidence is indispensable. Given the limited scope of the application of the probability integral model on the edge of the mobile basin, its poor prediction effect, and its low accuracy, a new mining subsidence prediction model based on the Boltzmann function is proposed. Combined with the transformed normal distribution time function, a B-normal prediction model that can predict the dynamic displacement and deformation of any point on the surface was constructed. The global optimal solution of the parameters of the dynamic prediction model was inversed by introducing particle swarm optimization shuffled frog leaping intelligent algorithm (PSO-SFLA), and then, the model was applied to the 8102 working face of the Guobei coal mine to dynamically predict the subsidence, inclination, curvature, horizontal displacement, and horizontal deformation of the goaf surface. The prediction results showed that on the strike and dip observation lines, the prediction accuracy of the dynamic subsidence and horizontal displacement of the surface could reach the centimeter level, the predicted root mean square error (RMSE) of dynamic tilt and horizontal deformation was less than 0.51 mm/m, and the predicted RMSE of dynamic curvature was within 0.020 mm/m2. The prediction results reflected the dynamic evolution law of surface displacement and deformation and verified the reliability of the B-normal dynamic prediction model, which can fully meet the needs of practical engineering applications.

Deep learning for skillful long-lead ENSO forecasts

El Niño-Southern Oscillation (ENSO) is one of the fundamental drivers of the Earth's climate variability. Thus, its skillful prediction at least a few months to years ahead is of utmost importance to society. Using both dynamical and statistical methods, several studies reported skillful ENSO predictions at various lead times. Predictions with long lead times, on the other hand, remain difficult. In this study, we propose a convolutional neural network (CNN)-based statistical ENSO prediction system with heterogeneous CNN parameters for each season with a modified loss function to predict ENSO at least 18–24 months ahead. The developed prediction system indicates that the CNN model is highly skillful in predicting ENSO at long lead times of 18–24 months with high skills in predicting extreme ENSO events compared with the Scale Interaction Experiment-Frontier ver. 2 (SINTEX-F2) dynamical system and several other statistical prediction systems. The analysis indicates that the CNN model can overcome the spring barrier, a major hindrance to dynamical prediction systems, in predicting ENSO at long lead times. The improvement in the prediction skill can partly be attributed to the heterogeneous parameters of seasonal CNN models used in this study and also to the use of a modified loss function in the CNN model. In this study, we also attempted to identify various precursors to ENSO events using CNN heatmap analysis.

Toward Industry 4.0 in Surface Mount Technology: Smart Manufacturing in Stencil Printing Operations

As the foremost point of contact in surface mount technology (SMT) operations, the stencil/solder paste printer (SPP) is a major contributor to assembly defects in the printed circuit board (PCB) fabrication process. Initial printing defects lead to reductions in throughput, quality and reliability of the entire downstream system. Because of the high variation in SMT processes, a predictive, real-time machine learning (ML) algorithm to optimize printer settings based on the estimation of printing volumes is a promising solution to improving production performance and process capability. Building upon existing literature, this research discusses the implementation of a lab-trained ML model to predict optimal printing parameters and reduce changeover time within SPP processes through the training and testing of large solder paste inspection (SPI) datasets. To sustain growing market demand for smart electronics manufacturing (SEM), successful development and validation of a dynamic prediction system for optimizing printing parameters provides significant potential for improvements in first pass yield rates and system throughput.

Development and validation of a dynamic risk prediction system for constipation in patients with amyotrophic lateral sclerosis.

Although constipation is a common non-motor symptom in patients with amyotrophic lateral sclerosis (ALS), it is poorly valued. Moreover, there is a bidirectional effect between constipation and neuropsychiatric and sleep disturbances. Thus, these symptoms are better treated simultaneously. Therefore, this study aimed to develop and validate a model for predicting the risk of constipation in ALS patients, to help clinicians identify and treat constipation early. Data of 118 ALS admissions from an observational prospective cohort, registered between March 2017 and December 2021, were analyzed. Demographic data were obtained. Constipation was assessed using the Knowles-Eccersley-Scott Symptom Questionnaire. The severity of ALS was assessed using the Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised (ALSFRS-R). Anxiety and depressive symptoms were measured using the Hospital Anxiety and Depression Scale (HADS). The Pittsburgh Sleep Quality Index (PSQI) was used to assess patients' sleep status. The least absolute shrinkage and selection operator (LASSO) regression model was used to select factors and construct a nomogram. Nomogram model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curve, decision curve analysis (DCA), and clinical impact curve (CIC). The model was internally validated using bootstrap validation in the current cohort. Age, family history of constipation, total ALSFRS-R score, site of onset, total PSQI score, and depressed, were identified as significant predictors of the risk of constipation in ALS patients. The prediction model was validated to have good accuracy (Hosmer-Lemeshow test: χ2 = 11.11, P > 0.05) and discrimination (AUC = 0.856, 95% confidence interval: 0.784-0.928). DCA and CIC showed that the nomogram model had excellent clinical performance. A web-based ALS constipation risk calculator with good predictive performance was constructed to identify patients at high risk of constipation and to allow early intervention in a clinical context.

What potential for improving sub‐seasonal predictions of the winter NAO?

AbstractThe North Atlantic Oscillation (NAO) is the leading mode of variability across the Atlantic sector and is a key metric of extratropical forecast performance. Skilful predictions of the NAO are possible at medium‐range (1–2 weeks) and seasonal time scales. However, in a leading dynamical prediction system, we find that sub‐seasonal predictions (1 month NAO with a lead time of 20–30 days) are not statistically significant and represent a gap in forecast skill. In this study, we have investigated the potential for improving predictions using a large ensemble of dynamical hindcasts. First, we find that monthly predictions of the NAO are only weakly related to forecast errors at the medium‐range. This implies that improving medium‐range forecast performance is unlikely to drive significant improvements at longer lead times. Second, the Madden‐Julian Oscillation (MJO) is the leading mode of sub‐seasonal variability in the Tropics and projects onto the NAO with a lag of 10–15 days, but its teleconnection is only partially represented in current forecast systems. We, therefore, assess whether improved MJO‐NAO teleconnections are likely to lead to improved monthly NAO predictions. We find that even perfect MJO forecasts and teleconnections lead to only small improvements in NAO prediction skills. This work indicates that monthly timescales may represent a predictability gap for the NAO and hence the Euro‐Atlantic winter climate in which genuine skill improvements are difficult to achieve. Potential progress in this area could stem from currently unknown sources of skill and large initialised climate ensembles will be a vital tool for investigating these.

Designing Collaborative Edge Computing for Electricity Heterogeneous Data Based on Social IoT Systems

Power generation, transmission, maintenance costs, and electricity prices are heavily influenced by accurate load forecasts at energy suppliers' operation centers. Every aspect of our life has been transformed by the social internet of things (SIoT). Collaborative edge computing (CEC) has emerged as a new paradigm for meeting the demands of the internet of things by alleviating resource congestion (IoT). Remote devices can connect to CEC's processing, storage, and network resources. About short-term electrical load forecasting, this study explores the application of feed-forward deep neurological networking (FF-DNN) and recurrent deep neuronal networking (R-DNN) methods and analyzes their accuracy and computing performance. A dynamic prediction system using a deep neural network (DPS-DNN) is proposed in this research. The recently unveiled smartgrid with the results shows the higher performance of the proposed DPS-DNN model than the existing models with an enhancement of 93.15% based on collaborative edge networks based on SIoT.

Deep Learning for Predicting Winter Temperature in North China

It is difficult to improve the seasonal prediction skill of winter temperature over North China, owing to the complex dynamics of East Asian winter and the relatively low prediction skill level of current climate models. Deep learning (DL) may be an informative and promising tool to enhance seasonal prediction, particularly in regions where the underlying mechanisms are not clear. Here, using a DL model based on the Convolutional Neural Network (CNN), we have found that the prediction skill for North China winter temperature (NCWT) can be extended up to five months by considering the remote impact of the Northeast Pacific sea-surface temperature (SST) on North China. Based on historical simulations of winter temperatures in North China, we selected six CMIP5 models with relatively small deviations for training the CNN, and the period chosen for training was 1852–1991. The ERA5 data during 1995–2017 were utilized to evaluate the performance of the CNN. Our CNN shows the best performance in a recent 10-year period (2008–2017), showing a significantly improved level of NCWT prediction skill with a correlation skill of 0.65 at a 5-month lead time, which is much better than the forecast skill of the state-of-the-art dynamic seasonal prediction system. Heat map analysis was used to explore the possible physical mechanisms associated with the NCWT anomaly from the perspective of the CNN; the results showed that the SST over the Northeast Pacific is highly relevant to NCWT prediction. The Northeast Pacific warming in the boreal summer is related to the development of the El Niño event in the coming winter, which may induce NCWT anomalies by atmospheric teleconnection. Climate model experiments support the role of Northeast Pacific warming in the boreal summer on NCWT. The improved capability for prediction from using the CNN may help to establish the energy policy for the coming winter and reduce the economic losses from extremely cold in North China.

Impacts of Interannual Variations of Chlorophyll on Seasonal Predictions of the Tropical Pacific

In this study, we explored impacts of interannual variations of chlorophyll on seasonal predictions of the tropical Pacific by the SINTEX-F2 dynamical climate prediction system, which is highly skillful at predicting El Niño/Southern Oscillation (ENSO) and other tropical climate phenomena. We conducted twin re-forecast experiments; one system used the observed climatology of chlorophyll to compute the shortwave absorption in the upper ocean, while the other used the observed chlorophyll with year-to-year variations. Although the chlorophyll impacts on predictions of the Niño 3.4 index were limited, improvements are noticed in the predictions of sea surface temperature over the eastern edge of the Western Pacific Warm Pool. This region corresponds to the separation between warm, low-salinity waters of the warm Pool and cold, high-salinity upwelled waters of the Pacific cold tongue in the central-eastern equatorial Pacific. The improvement was very striking in the 2015 case, when a super El Nino occurred.

Propagation of Thermohaline Anomalies and Their Predictive Potential along the Atlantic Water Pathway

Abstract We assess to what extent seven state-of-the-art dynamical prediction systems can retrospectively predict winter sea surface temperature (SST) in the subpolar North Atlantic and the Nordic seas in the period 1970–2005. We focus on the region where warm water flows poleward (i.e., the Atlantic water pathway to the Arctic) and on interannual-to-decadal time scales. Observational studies demonstrate predictability several years in advance in this region, but we find that SST skill is low with significant skill only at a lead time of 1–2 years. To better understand why the prediction systems have predictive skill or lack thereof, we assess the skill of the systems to reproduce a spatiotemporal SST pattern based on observations. The physical mechanism underlying this pattern is a propagation of oceanic anomalies from low to high latitudes along the major currents, the North Atlantic Current and the Norwegian Atlantic Current. We find that the prediction systems have difficulties in reproducing this pattern. To identify whether the misrepresentation is due to incorrect model physics, we assess the respective uninitialized historical simulations. These simulations also tend to misrepresent the spatiotemporal SST pattern, indicating that the physical mechanism is not properly simulated. However, the representation of the pattern is slightly degraded in the predictions compared to historical runs, which could be a result of initialization shocks and forecast drift effects. Ways to enhance predictions could include improved initialization and better simulation of poleward circulation of anomalies. This might require model resolutions in which flow over complex bathymetry and the physics of mesoscale ocean eddies and their interactions with the atmosphere are resolved. Significance Statement In this study, we find that dynamical prediction systems and their respective climate models struggle to realistically represent ocean surface temperature variability in the eastern subpolar North Atlantic and Nordic seas on interannual-to-decadal time scales. In previous studies, ocean advection is proposed as a key mechanism in propagating temperature anomalies along the Atlantic water pathway toward the Arctic Ocean. Our analysis suggests that the predicted temperature anomalies are not properly circulated to the north; this is a result of model errors that seems to be exacerbated by the effect of initialization shocks and forecast drift. Better climate predictions in the study region will thus require improving the initialization step, as well as enhancing process representation in the climate models.

Improving the Forecasting of Winter Wheat Yields in Northern China with Machine Learning–Dynamical Hybrid Subseasonal-to-Seasonal Ensemble Prediction

Subseasonal-to-seasonal (S2S) prediction of winter wheat yields is crucial for farmers and decision-makers to reduce yield losses and ensure food security. Recently, numerous researchers have utilized machine learning (ML) methods to predict crop yield, using observational climate variables and satellite data. Meanwhile, some studies also illustrated the potential of state-of-the-art dynamical atmospheric prediction in crop yield forecasting. However, the potential of coupling both methods has not been fully explored. Herein, we aimed to establish a skilled ML–dynamical hybrid model for crop yield forecasting (MHCF v1.0), which hybridizes ML and a global dynamical atmospheric prediction system, and applied it to northern China at the S2S time scale. In this study, we adopted three mainstream machining learning algorithms (XGBoost, RF, and SVR) and the multiple linear regression (MLR) model, and three major datasets, including satellite data from MOD13C1, observational climate data from CRU, and S2S atmospheric prediction data from IAP CAS, used to predict winter wheat yield from 2005 to 2014, at the grid level. We found that, among the four models examined in this work, XGBoost reached the highest skill with the S2S prediction as inputs, scoring R2 of 0.85 and RMSE of 0.78 t/ha 3–4 months, leading the winter wheat harvest. Moreover, the results demonstrated that crop yield forecasting with S2S dynamical predictions generally outperforms that with observational climate data. Our findings highlighted that the coupling of ML and S2S dynamical atmospheric prediction provided a useful tool for yield forecasting, which could guide agricultural practices, policy-making and agricultural insurance.

Combining Dynamical and Statistical Modeling to Improve the Prediction of Surface Air Temperatures 2 Months in Advance: A Hybrid Approach

A new type of hybrid prediction system (HPS) of the land surface air temperature (SAT) is described and its skill evaluated for one particular application. This approach utilizes sea-surface temperatures (SST) forecast by a dynamical prediction system, SINTEX-F2, to provide predictors of the SAT to a statistical modeling system consisting of a set of nine different machine learning algorithms. The statistical component is aimed to restore teleconnections between SST and SAT, particularly in the mid-latitudes, which are generally not captured well in the dynamical prediction system. The HPS is used to predict the SAT in the central region of Japan around Tokyo (Kantō) as a case study. Results show that at 2-month lead the hybrid model outperforms both persistence and the SINTEX-F2 prediction of SAT. This is also true when prediction skill is assessed for each calendar month separately. Despite the model's strong performance, there are also some limitations. The limited sample size makes it more difficult to calibrate the statistical model and to reliably evaluate its skill.

Dynamic prediction of renal survival among deeply phenotyped kidney transplant recipients using artificial intelligence: an observational, international, multicohort study

Kidney allograft failure is a common cause of end-stage renal disease. We aimed to develop a dynamic artificial intelligence approach to enhance risk stratification for kidney transplant recipients by generating continuously refined predictions of survival using updates of clinical data. In this observational study, we used data from adult recipients of kidney transplants from 18 academic transplant centres in Europe, the USA, and South America, and a cohort of patients from six randomised controlled trials. The development cohort comprised patients from four centres in France, with all other patients included in external validation cohorts. To build deeply phenotyped cohorts of transplant recipients, the following data were collected in the development cohort: clinical, histological, immunological variables, and repeated measurements of estimated glomerular filtration rate (eGFR) and proteinuria (measured using the proteinuria to creatininuria ratio). To develop a dynamic prediction system based on these clinical assessments and repeated measurements, we used a Bayesian joint models-an artificial intelligence approach. The prediction performances of the model were assessed via discrimination, through calculation of the area under the receiver operator curve (AUC), and calibration. This study is registered with ClinicalTrials.gov, NCT04258891. 13 608 patients were included (3774 in the development cohort and 9834 in the external validation cohorts) and contributed 89 328 patient-years of data, and 416 510 eGFR and proteinuria measurements. Bayesian joint models showed that recipient immunological profile, allograft interstitial fibrosis and tubular atrophy, allograft inflammation, and repeated measurements of eGFR and proteinuria were independent risk factors for allograft survival. The final model showed accurate calibration and very high discrimination in the development cohort (overall dynamic AUC 0·857 [95% CI 0·847-0·866]) with a persistent improvement in AUCs for each new repeated measurement (from 0·780 [0·768-0·794] to 0·926 [0·917-0·932]; p<0·0001). The predictive performance was confirmed in the external validation cohorts from Europe (overall AUC 0·845 [0·837-0·854]), the USA (overall AUC 0·820 [0·808-0·831]), South America (overall AUC 0·868 [0·856-0·880]), and the cohort of patients from randomised controlled trials (overall AUC 0·857 [0·840-0·875]). Because of its dynamic design, this model can be continuously updated and holds value as a bedside tool that could refine the prognostic judgements of clinicians in everyday practice, hence enhancing precision medicine in the transplant setting. MSD Avenir, French National Institute for Health and Medical Research, and Bettencourt Schueller Foundation.