Support Vector Regression Algorithm Research Articles

DDCM: A Computational Strategy for Drug Repositioning Based on Support-Vector Regression Algorithm.

Computational drug-repositioning technology is an effective tool for speeding up drug development. As biological data resources continue to grow, it becomes more important to find effective methods to identify potential therapeutic drugs for diseases. The effective use of valuable data has become a more rational and efficient approach to drug repositioning. The disease-drug correlation method (DDCM) proposed in this study is a novel approach that integrates data from multiple sources and different levels to predict potential treatments for diseases, utilizing support-vector regression (SVR). The DDCM approach resulted in potential therapeutic drugs for neoplasms and cardiovascular diseases by constructing a correlation hybrid matrix containing the respective similarities of drugs and diseases, implementing the SVR algorithm to predict the correlation scores, and undergoing a randomized perturbation and stepwise screening pipeline. Some potential therapeutic drugs were predicted by this approach. The potential therapeutic ability of these drugs has been well-validated in terms of the literature, function, drug target, and survival-essential genes. The method's feasibility was confirmed by comparing the predicted results with the classical method and conducting a co-drug analysis of the sub-branch. Our method challenges the conventional approach to studying disease-drug correlations and presents a fresh perspective for understanding the pathogenesis of diseases.

MDs-NP: a property prediction model construction procedure for naphtha based on molecular dynamics simulation

In the context of carbon neutrality and carbon peaking, molecular management has become a focus of the petrochemical industry. The key to achieving molecular management is molecular reconstruction, which relies on rapid and accurate calculation of oil properties. Focusing on naphtha, we proposed a novel property prediction model construction procedure (MDs-NP) employing molecular dynamics simulations for property collections and gamma distribution from real analytical data for calculating mole fractions of simulation mixtures. We calculated 348 sets of mixture properties data in the range of 273 K–300 K by molecular dynamics simulations. Molecular feature extraction was based on molecular descriptors. In addition to descriptors based on open-source toolkits (RDKit and Mordred), we designed 12 naphtha knowledge (NK) descriptors with a focus on naphtha. Three machine learning algorithms (support vector regression, extreme gradient boosting and artificial neural network) were applied and compared to establish models for the prediction of the density and viscosity of naphtha. Mordred and NK descriptors + support vector regression algorithm achieved the best performance for density. The selected RDKFp and NK descriptors + artificial neural network algorithm achieved the best performance for viscosity. Using ablation studies, T, P_w and CC(C)C are three effective descriptors in NK that can improve the performance of the property prediction models. MDs-NP has the potential to be extended to more properties as well as more-complex petroleum systems. The models from MDs-NP can be used for rapid molecular reconstruction to facilitate construction of data-driven models and intelligent transformation of petrochemical processes.

Method for Monitoring Wheat Growth Status and Estimating Yield Based on UAV Multispectral Remote Sensing

An efficient and accurate estimation of wheat growth and yield is important for wheat assessment and field management. To improve the accuracy and stability of wheat growth and yield estimation, an estimation method based on a genetic algorithm-improved support vector regression (GA-SVR) algorithm was proposed in this study. The correlation analysis between vegetation indices calculated from spectral data and wheat growth phenotypes and yields was performed to obtain the optimal combination of vegetation indices with high correlation and good estimation performance. At the same time, the optimal model for wheat growth monitoring was screened and constructed in experiments with 12 wheat varieties and 3 gradient nitrogen fertilizer application levels. Then, the yield estimation model was established and its applicability was verified under different nitrogen fertilizer application levels. The results showed that the constructed models for the leaf area index, plant height, and yield estimation performed well, with coefficients of determination of 0.82, 0.71, and 0.70, and root mean square errors of 0.09, 2.7, and 68.5, respectively. This study provided an effective UAV remote sensing technique for monitoring wheat growth status and estimating yield. This study provides an effective unmanned aerial remote sensing technique for monitoring wheat growth and estimating yield, and provides technical support for wheat yield assessment and field management.

An Interwell Connectivity Assessment Model for Polymer Flooding Short-Term Development Data Based on A-LSTM and EFAST Methods

Interwell connectivity assessment in polymer-driven reservoirs is critical for setting appropriate injection rates and improving oil recovery. Traditional deep learning techniques often lack accuracy and reliability when applied to short-term oilfield production data. In response, the A-LSTM algorithm is proposed, which integrates the attention mechanism with a long- and short-term memory network (LSTM). The predictive accuracy of A-LSTM is assessed and juxtaposed with LSTM and support vector regression (SVR) algorithms for short-term single-well daily oil production analysis. The Huber loss function was utilized to quantify the difference between predicted and actual results, resulting in a dynamic production prediction model. An interwell connectivity (IWC) assessment model is then obtained by fusing the dynamic production prediction model with the EFAST method, thus demonstrating the superior prediction accuracy of A-LSTM in oil production prediction and connectivity assessment. Moreover, the credibility of the assessment is further corroborated through numerical simulations and interwell tracer tests. The study results showed that the interwell connectivity evaluation model based on the A-LSTM algorithm and EFAST method is not only capable of accurately predicting the single-well daily oil production using a small sample dataset but also a highly reliable method for interwell connectivity evaluation, and the application of the interwell connectivity assessment model can further guide polymer flooding work in oilfields.

Assessing Pipe Condition in Water Distribution Networks

The condition assessment of water distribution pipes is of utmost importance for the prioritization of rehabilitation interventions. However, the application of available methodologies for condition assessment by water utilities with limited human, technological and financial resources is becoming increasingly complex. The current paper aims at the development and application of a methodology for the prediction of the physical condition of water distribution pipes without the need for visual inspection. The methodology includes the development and application of three different algorithms (heuristic, linear regression and support vector regression). The methodology is applied to a water distribution network located in the Algarve region, Portugal. The results obtained from each algorithm are compared with a well-known performance indicator, the ratio of useful life, and present significant differences in its overall pipe condition classification. Results have demonstrated the following: the ratio of useful life tends to distribute pipe classification more equally in the three classes (i.e., good, average and unsatisfactory); the heuristic algorithm classifies most pipes as average condition; and the linear regression algorithm classifies with unsatisfactory conditions. The support vector regression algorithm stands out as the main classifier for identifying pipes in good condition when compared to other algorithms.

Energy-saving and low-carbon leather production: AI-assisted chrome tanning process optimization

Reducing CO2 emissions is a global effort to improve industrial cost efficiency, enhance product compatibility, and fight climate change. For leather-making industry, the optimization of tanning process is crucial to fulfill the goal of low-carbon manufacturing. Artificial intelligence (AI) is an emerging technology for its ability to reduce errors, increase efficiency, and help realizing precise optimization of manufacturing processes. Herein, an AI-assisted method was first proposed to optimize chrome tanning process for energy-saving and carbon reduction. By comparing the performance of four machine learning (ML) algorithms, i.e., multiple linear regression (MLR), polynomial regression (PR), support vector regression (SVR), and backpropagation-artificial neural network (BP-ANN), it was found that machine learning models are superior in identifying the optimization of process parameters (pH and time) which lead to maintain key product quality indexes in the penetration procedure and binding procedure. Out of the four ML models, the BP-ANN algorithm and SVR algorithm are proven to have good predictive ability. Thus, we systematically optimized the chrome tanning process for energy saving, with key product quality as the constraints. The optimal process parameters, namely pickling pH of 2.86, penetration time of 90 min, pH after basifying of 4.08, and binding time of 90 min, were generated by solving the optimization models of the chrome tanning process via the genetic algorithm (GA) with powerful global optimization capability. Finally, it was validated that the optimized process was highly reliable by evaluating the wet blue product performances, such as the evenness of chrome distribution, thermal stability, surface reactivity, morphology, as well as physical and mechanical properties. Notably, compared to conventional chrome tanning, the optimized process will save 8353.33 kgce of energy and reduce 87295.6 kgCO2 of CO2 emission in an average tannery with an annual production capacity of 300,000 standard sheets of upper cowhide leather. This approach can precisely optimize conventional leather-making process, providing a fesible way to achieve energy-saving and low-carbon leather production.

A framework for estimating the matric suction in unsaturated soils using multiple artificial intelligence techniques

AbstractImplementation of the state‐of‐the‐art understanding of the mechanics of unsaturated soils into geotechnical engineering practice is partly limited due to the lack of quick, reliable, and economical techniques for matric suction measurement. Matric suction is one of the key stress state variables that significantly influences the hydro‐mechanical behavior of unsaturated soils. In this paper, to address this objective, two artificial intelligence (AI) models were developed for estimating matric suction in unsaturated soils based on the particle swarm optimization support vector regression (PSO‐SVR) and multivariate adaptive regression spline (MARS) algorithms. The results suggest that both these models can reasonably estimate matric suction. Compared to the MARS model, the PSO‐SVR model can achieve higher accuracy. Nonetheless, the MARS model facilitates the sensitivity analysis and the selection of essential inputs. A novel integrated framework is proposed and validated, leveraging the strengths, and alleviating the limitations of the PSO‐SVR and MARS algorithms for reliable and rapid estimation of matric suction in the range of 0–1500 kPa for low plastic soils (0 < Ip ≤ 7). Six inputs are required to use this model successfully; some can be measured using conventional laboratory tests, and others can be calculated from mass‐volume relationships.

Optimized design of composition and brazing process for Cu-Ag-Zn-Mn-Ni-Si-B-P alloy brazing material based on machine learning strategy to improve brazing properties

As brazing devices become more sophisticated and service environments become more demanding, Cu-Ag-Zn-Mn-Ni-Si-B-P brazing material are subjected to higher wettability and brazed joint strength requirements. In this paper, a machine learning strategy is proposed. Firstly, the composition-process-properties dataset is obtained based on high-throughput experimental techniques. Then, the composition features are constructed by elemental chemical and physical parameters instead of the direct input of alloy composition, and the proposed novel feature selection method is used to effectively identify the key composition feature combinations that affect brazing properties. Then, the support vector regression algorithm was used to establish a joint strength model with an error of less than 3.47 % and a spreading area model with an error of less than 8.37 % using the key alloy composition features and brazing process features as inputs, respectively. Finally, two multi-objective optimization strategies, Bayesian optimization and NSGA-II genetic algorithm, were used for the integrated optimization design of composition and brazing process. The experimental results show that the Cu-14Ag-4Zn-24Mn-13.5Ni-0.4Si-0.3B-0.3 P alloy and the matched brazing process (945 °C and 12 min) were optimally designed using NSGA-II genetic algorithm, corresponding to the measured joint strength and spread area of 356 MPa and 412 mm2, respectively, which outperforms the high-throughput experimentally optimized data as well as commercialized data. The machine learning strategy established in this work provides a new research idea for the comprehensive improvement of brazing properties of brazing alloys with small data sets.

Predicting Scale Deposition in Oil Reservoirs Using Machine Learning Optimization Algorithms

Scale deposition, a form of formation damage, not only affects the reservoir but also damages the well and equipment. This phenomenon occurs due to changes in temperature, pressure, and the injection of incompatible salt water, leading to ionic reactions. This study investigated permeability reduction due to scale deposition and examined how parameters such as temperature, pressure drop, and ion concentration affect the prediction accuracy. The scale deposits investigated in this study include CaSO4, BaSO4, and SrSO4. This paper uses Python to employ different machine-learning algorithms to predict the results. Each machine learning model has certain hyper-parameters that need adjustment. Failure to do so will result in reduced accuracy and incomplete interpretation of input data. The accuracy of the support vector regression (SVR) algorithm was significantly affected by the variation of the epsilon parameter in the dataset used. Therefore, before hyperparameter optimization, SVR had the lowest accuracy at 0.575. After adjusting the hyper-parameters, our findings show that SVR had the highest increase in R-squared value, which was 0.900, and the most minor growth in KNN, which went from 0.995 to 0.996. Additionally, the highest accuracy value for K-Nearest Neighbor is 0.996. Furthermore, most errors were related to SVR and XGBoost algorithms, while the most negligible errors were for the Decision Tree and KNN algorithms.

Prediction of subthalamic stimulation efficacy on isolated dystonia via support vector regression

IntroductionDeep brain stimulation (DBS) of subthalamic nucleus (STN) has been well-established and increasingly applied in patients with isolated dystonia. Nevertheless, the surgical efficacy varies among patients. This study aims to explore the factors affecting clinical outcomes of STN-DBS on isolated dystonia and establish a well-performed prediction model. MethodsIn this prospective study, thirty-two dystonia patients were recruited and received bilateral STN-DBS at our center. Their baseline characteristics and up to one-year follow-up outcomes were assessed. Implanted electrodes of each subject were reconstructed with their contact coordinates and activated volumes calculated. We explored correlations between distinct clinical characteristics and surgical efficacy. Those features were then trained for the model in outcome prediction via support vector regression (SVR) algorithm and testified through cross-validation. ResultsPatients demonstrated an average clinical improvement of 56 ± 25 % after STN-DBS, significantly affected by distinct symptom forms and activated volumes. The optimal targets and activated volumes were concentratedly located at the dorsal posterior region to STN. Most patients had a rapid response to STN-DBS, and their motor score improvement within one week was highly associated with long-term outcomes. The trained SVR model, contributed by distinct weights of features, could reach a maximum prediction accuracy with mean errors of 11 ± 7 %. ConclusionSTN-DBS demonstrated significant and rapid therapeutic effects in patients with isolated dystonia, by possibly affecting the pallidofugal fibers. Early improvement highly indicates the ultimate outcomes. SVR proves valid in outcome prediction. Patients with predominant phasic and generalized symptoms, shorter disease duration, and younger onset age may be more favorable to STN-DBS in the long run.

Development of a Novel HS-GC/MS Method Using the Total Ion Spectra Combined with Machine Learning for the Intelligent and Automatic Evaluation of Food-Grade Paraffin Wax Odor Level.

The intensity of the odor in food-grade paraffin waxes is a pivotal quality characteristic, with odor panel ratings currently serving as the primary criterion for its assessment. This study presents an innovative method for assessing odor intensity in food-grade paraffin waxes, employing headspace gas chromatography with mass spectrometry (HS/GC-MS) and integrating total ion spectra with advanced machine learning (ML) algorithms for enhanced detection and quantification. Optimization was conducted using Box-Behnken design and response surface methodology, ensuring precision with coefficients of variance below 9%. Analytical techniques, including hierarchical cluster analysis (HCA) and principal component analysis (PCA), efficiently categorized samples by odor intensity. The Gaussian support vector machine (SVM), random forest, partial least squares regression, and support vector regression (SVR) algorithms were evaluated for their efficacy in odor grade classification and quantification. Gaussian SVM emerged as superior in classification tasks, achieving 100% accuracy, while Gaussian SVR excelled in quantifying odor levels, with a coefficient of determination (R2) of 0.9667 and a root mean square error (RMSE) of 6.789. This approach offers a fast, reliable, robust, objective, and reproducible alternative to the current ASTM sensory panel assessments, leveraging the analytical capabilities of HS-GC/MS and the predictive power of ML for quality control in the petrochemical sector's food-grade paraffin waxes.

Design and optimization of haze prediction model based on particle swarm optimization algorithm and graphics processor

With the rapid expansion of industrialization and urbanization, fine Particulate Matter (PM2.5) pollution has escalated into a major global environmental crisis. This pollution severely affects human health and ecosystem stability. Accurately predicting PM2.5 levels is essential. However, air quality forecasting currently faces challenges in processing vast data and enhancing model accuracy. Deep learning models are widely applied for their superior learning and fitting abilities in haze prediction. Yet, they are limited by optimization challenges, long training periods, high data quality needs, and a tendency towards overfitting. Furthermore, the complex internal structures and mechanisms of these models complicate the understanding of haze formation. In contrast, traditional Support Vector Regression (SVR) methods perform well with complex non-linear data but struggle with increased data volumes. To address this, we developed CUDA-based code to optimize SVR algorithm efficiency. We also combined SVR with Genetic Algorithms (GA), Sparrow Search Algorithm (SSA), and Particle Swarm Optimization (PSO) to identify the optimal haze prediction model. Our results demonstrate that the model combining intelligent algorithms with Central Processing Unit-raphics Processing Unit (CPU-GPU) heterogeneous parallel computing significantly outpaces the PSO-SVR model in training speed. It achieves a computation time that is 6.21–35.34 times faster. Compared to other models, the Particle Swarm Optimization-Central Processing Unit-Graphics Processing Unit-Support Vector Regression (PSO-CPU-GPU-SVR) model stands out in haze prediction, offering substantial speed improvements and enhanced stability and reliability while maintaining high accuracy. This breakthrough not only advances the efficiency and accuracy of haze prediction but also provides valuable insights for real-time air quality monitoring and decision-making.

A new algorithm of critical control rod position search in MC code

Delayed Material Replacement algorithm is used in control rod movement simulation. Then, critical control rod position search is studied in Monte Carlo criticality calculation. Support Vector Regression algorithm is used in fitting neutron absorption as a function of rod positions. Fission iteration batches are reorganized as blocks in order to avoid fission source perturbation caused by control rod movement. The keff result in AP1000 reactor calculation shows discrepancy less than 20 pcm, though stable fluctuations are found in rod position values during fission iteration.

A design of power prediction algorithm based on health assessment optimized neural network

Wind power prediction holds significant value for the stability of the electrical grid when wind power is connected to the grid. Using neural networks for wind power prediction may have some limitations, such as slow speed and low accuracy. This paper proposes to enhance the power prediction accuracy and speed by optimizing the neural network through health assessment wind turbines. Firstly, based on wind turbine actual operating data, a health assessment is conducted to obtain a health matrix of wind turbine. Then, by calculating the weights of the matrix, the power prediction strategy of the network is optimized. Following that, matrix approximation hyperparameters are utilized to expedite the optimization process. Finally, some tests are conducted on neural network power prediction, act as optimized back propagation (BP) neural network and whale swarm algorithm–support vector regression (WSA-SVR) neural networks are employed for wind power prediction. Results show noticeable optimization: after optimizing the BP network, power prediction accuracy increased by about 40%, and prediction speed rose by about 20%; after optimizing the WSA-SVR network, power prediction accuracy improved by 10%, and prediction speed surged by about 45%. Further analysis shows that this method can improve the accuracy and speed of most neural network wind power prediction algorithms.

An Investigation of Real-Time Robotic Polishing Motion Planning Using a Dynamical System

When addressing the technical challenges of achieving precise force tracking during the local polishing process of polishing robots, controlling the contact state between the robot and the workpiece surface is essential. To this end, a contact motion-planning strategy based on dynamic systems is designed to generate trajectory routes during local polishing. The trajectory simulation of the local modulation dynamic system is achieved through the employment of the support vector regression (SVR) algorithm with a Gaussian kernel, facilitating the learning process. The feasibility and stability of planning local paths are validated using the local modulation dynamic system. To maintain a constant contact force between the end-effector polishing robot and the workpiece, an integral adaptive impedance control strategy is utilized, enabling the robot’s compliant control. Subsequently, an experimental system for the polishing robot is constructed in order to verify the effectiveness of the motion-planning system. The experimental results demonstrate that the proposed motion-planning approach is applicable in practical polishing processes, ensuring smooth contact and maintaining the desired contact force when scanning nonlinear surfaces, and thus showcasing stability and practicality.

Predicting individual autistic symptoms for patients with autism spectrum disorder using interregional morphological connectivity

Intelligent predictive models for autistic symptoms based on neuroimaging datasets were beneficial for the precise intervention of patients with ASD. The goals of this study were twofold: investigating predictive models for autistic symptoms and discovering the brain connectivity patterns for ASD-related behaviors. To achieve these goals, we obtained a cohort of patients with ASD from the ABIDE project. The autistic symptoms were measured using the Autism Diagnostic Observation Schedule (ADOS). The anatomical MRI datasets were preprocessed using the Freesurfer package, resulting in regional morphological features. For each individual, the interregional morphological network was constructed using a novel feature distance-based method. The predictive models for autistic symptoms were built using the support vector regression (SVR) algorithm with feature selection method. The predicted autistic symptoms (i.e., ADOS social score, ADOS behavior) were significantly correlated to the original measures. The most predictive features for ADOS social scores were located in the bilateral fusiform. The most predictive features for ADOS behavior scores were located in the temporal pole and the lingual gyrus. In summary, the autistic symptoms could be predicted using the interregional morphological connectivity and machine learning. The interregional morphological connectivity could be a potential biomarker for autistic symptoms.

Machine learning modeling of base flow generation potential: A case study of the combined application of BWM and Fallback bargaining algorithm

The study of base flow is essential for sustainable water management. By understanding the dynamics of base flow, policymakers can make informed decisions to ensure the long-term health and availability of water resources. This study was conducted with the aim of modeling the spatial changes of Base Flow Generation Potential (BFGP) using integrated Machine Learning Algorithms (MLAs) including Decision Tree Regression (DTR), K-Nearest Neighbors (KNN), Naïve Bayes (NB), Random Forest (RF), Simple Linear Regression (SLR), Support Vector Machine (SVM) and Support Vector Regression (SVR) and optimal Multi-Criteria Decision Making (MCDM) methods including Best-Worst Method (BWM) and Fallback bargaining in the Cheshmeh-Kileh watershed, Iran. BFGP conditioning factors were selected for the Sub-Watershed (SW) and weighed using MCDM (BWM and Fallback bargaining algorithm) methods. BFGP maps of five classes were generated and analyzed. MCDM methods were also integrated with ML algorithms for optimum performance evaluation of the integrated unit. Based on the results, in the integrated model of BWM and Fallback bargaining algorithm with ML algorithms, SVR and RF algorithms was selected as the optimal model, respectively. In BWM in the Sehezar River, the upstream had the highest BFGP. The pattern of spatial changes based on the Fallback bargaining algorithm in the upstream of Sehezar River was similar to BWM. The combined approach of RF-GTA with the factors of minimum temperature (Tn), NDSI and elevation showed the highest correlation. It was observed that that the Fallback bargaining algorithm and BWM of the MCDM exhibited similarity in output of regarding the spatial change patterns of the BFGP. Also, Fallback bargaining exhibited a 67% similarity with the integrated RF-GTA in BFGP prioritization of the SWs, while BWM performed poorly. This suggests that the Fallback bargaining algorithm was more effective than BWM of the MCDM methods for SWs BFGP prioritization. The general summary indicated that the combined use of MCDM and MLAs were valuable tools in BF studies and helped in better understanding the complex hydrological processes. The findings of our research may be of help to Iran Water Resources Research Company using hydrological data arising from base flow studies to support their analysis. In conclusion, understanding the BF characteristics of SWs allows watershed managers to effectively prioritize conservation and restoration efforts aimed at protecting and enhancing the natural flow regime.

Accurate blood glucose level monitoring using microwave imaging

Painless and non-invasive detection techniques are needed to replace finger-prick blood collection for people with diabetes. A first-of-its-kind, noninvasive, and continuous blood glucose level (BGL) detection method based on microwave imaging is introduced in this paper. This method avoids the complex task of frequency choice for the design of electromagnetic sensors. A radar-based microwave imaging technology combined with an improved very-deep super-resolution (VDSR-BL) method is presented to obtain high-resolution (HR) microwave images. After image super-resolution reconstruction by VDSR-BL, the peak signal-to-noise ratio and structural similarity index of HR images reach 35.4461 dB and 0.9761, respectively. Then, an ensemble learning strategy based on support vector regression and random forest algorithms is proposed to identify HR microwave images for BGL estimation. The developed detection system has been verified on the medium under tests with different glucose solutions. The final detection results obtain a root mean squared error of 0.1394 mg ml−1 and a mean absolute relative difference of 8.02%, which show good accuracy with clinical acceptance. Meanwhile, we also conducted human trials. A high correlation coefficient (R) of 0.9254 was achieved between the results of microwave imaging and invasive BGL. Together, these results show that microwave imaging offers a promising new approach for noninvasive BGL monitoring.

A data-driven combined prediction method for the demand for intensive care unit healthcare resources in public health emergencies

BackgroundPublic health emergencies are characterized by uncertainty, rapid transmission, a large number of cases, a high rate of critical illness, and a high case fatality rate. The intensive care unit (ICU) is the “last line of defense” for saving lives. And ICU resources play a critical role in the treatment of critical illness and combating public health emergencies.ObjectiveThis study estimates the demand for ICU healthcare resources based on an accurate prediction of the surge in the number of critically ill patients in the short term. The aim is to provide hospitals with a basis for scientific decision-making, to improve rescue efficiency, and to avoid excessive costs due to overly large resource reserves.MethodsA demand forecasting method for ICU healthcare resources is proposed based on the number of current confirmed cases. The number of current confirmed cases is estimated using a bilateral long-short-term memory and genetic algorithm support vector regression (BILSTM-GASVR) combined prediction model. Based on this, this paper constructs demand forecasting models for ICU healthcare workers and healthcare material resources to more accurately understand the patterns of changes in the demand for ICU healthcare resources and more precisely meet the treatment needs of critically ill patients.ResultsData on the number of COVID-19-infected cases in Shanghai between January 20, 2020, and September 24, 2022, is used to perform a numerical example analysis. Compared to individual prediction models (GASVR, LSTM, BILSTM and Informer), the combined prediction model BILSTM-GASVR produced results that are closer to the real values. The demand forecasting results for ICU healthcare resources showed that the first (ICU human resources) and third (medical equipment resources) categories did not require replenishment during the early stages but experienced a lag in replenishment when shortages occurred during the peak period. The second category (drug resources) is consumed rapidly in the early stages and required earlier replenishment, but replenishment is timelier compared to the first and third categories. However, replenishment is needed throughout the course of the epidemic.ConclusionThe first category of resources (human resources) requires long-term planning and the deployment of emergency expansion measures. The second category of resources (drugs) is suitable for the combination of dynamic physical reserves in healthcare institutions with the production capacity reserves of corporations. The third category of resources (medical equipment) is more dependent on the physical reserves in healthcare institutions, but care must be taken to strike a balance between normalcy and emergencies.

Using sentinel-2 satellite images and machine learning algorithms to predict tropical pasture forage mass, crude protein, and fiber content.

Grasslands cover approximately 24% of the Earth's surface and are the main feed source for cattle and other ruminants. Sustainable and efficient grazing systems require regular monitoring of the quantity and nutritive value of pastures. This study demonstrates the potential of estimating pasture leaf forage mass (FM), crude protein (CP) and fiber content of tropical pastures using Sentinel-2 satellite images and machine learning algorithms. Field datasets and satellite images were assessed from an experimental area of Marandu palisade grass (Urochloa brizantha sny. Brachiaria brizantha) pastures, with or without nitrogen fertilization, and managed under continuous stocking during the pasture growing season from 2016 to 2020. Models based on support vector regression (SVR) and random forest (RF) machine-learning algorithms were developed using meteorological data, spectral reflectance, and vegetation indices (VI) as input features. In general, SVR slightly outperformed the RF models. The best predictive models to estimate FM were those with VI combined with meteorological data. For CP and fiber content, the best predictions were achieved using a combination of spectral bands and meteorological data, resulting in R2 of 0.66 and 0.57, and RMSPE of 0.03 and 0.04g/g dry matter. Our results have promising potential to improve precision feeding technologies and decision support tools for efficient grazing management.