Multiple Machine Learning Methods Research Articles

A non-invasive method to determine core temperature for cats and dogs using surface temperatures based on machine learning

BackgroundRectal temperature (RT) is an important index of core temperature, which has guiding significance for the diagnosis and treatment of pet diseases.ObjectivesDevelopment and evaluation of an alternative method based on machine learning to determine the core temperatures of cats and dogs using surface temperatures.Animals200 cats and 200 dogs treated between March 2022 and May 2022.MethodsA group of cats and dogs were included in this study. The core temperatures and surface body temperatures were measured. Multiple machine learning methods were trained using a cross-validation approach and evaluated in one retrospective testing set and one prospective testing set.ResultsThe machine learning models could achieve promising performance in predicting the core temperatures of cats and dogs using surface temperatures. The root mean square errors (RMSE) were 0.25 and 0.15 for cats and dogs in the retrospective testing set, and 0.15 and 0.14 in the prospective testing set.ConclusionThe machine learning model could accurately predict core temperatures for companion animals of cats and dogs using easily obtained body surface temperatures.

Improvement of the ozone forecast over Beijing through combining the chemical transport model with multiple machine learning methods

Improvement of the ozone forecast over Beijing through combining the chemical transport model with multiple machine learning methods

Identification of prognostic signatures in remnant gastric cancer through an interpretable risk model based on machine learning: a multicenter cohort study

ObjectiveThe purpose of this study was to develop an individual survival prediction model based on multiple machine learning (ML) algorithms to predict survival probability for remnant gastric cancer (RGC).MethodsClinicopathologic data of 286 patients with RGC undergoing operation (radical resection and palliative resection) from a multi-institution database were enrolled and analyzed retrospectively. These individuals were split into training (80%) and test cohort (20%) by using random allocation. Nine commonly used ML methods were employed to construct survival prediction models. Algorithm performance was estimated by analyzing accuracy, precision, recall, F1-score, area under the receiver operating characteristic curve (AUC), confusion matrices, five-fold cross-validation, decision curve analysis (DCA), and calibration curve. The best model was selected through appropriate verification and validation and was suitably explained by the SHapley Additive exPlanations (SHAP) approach.ResultsCompared with the traditional methods, the RGC survival prediction models employing ML exhibited good performance. Except for the decision tree model, all other models performed well, with a mean ROC AUC above 0.7. The DCA findings suggest that the developed models have the potential to enhance clinical decision-making processes, thereby improving patient outcomes. The calibration curve reveals that all models except the decision tree model displayed commendable predictive performance. Through CatBoost-based modeling and SHAP analysis, the five-year survival probability is significantly influenced by several factors: the lymph node ratio (LNR), T stage, tumor size, resection margins, perineural invasion, and distant metastasis.ConclusionsThis study established predictive models for survival probability at five years in RGC patients based on ML algorithms which showed high accuracy and applicative value.

Biological characteristics of molecular subtypes of ulcerative colitis characterized by ferroptosis and neutrophil infiltration

Clinical ulcerative colitis (UC) is a heterogeneous condition. Moreover, medical interventions are nonspecific, and thus, treatment responses are inconsistent. The aim of this study was to explore the molecular subtypes and biological characteristics of UC based on ferroptosis and neutrophil gene sets. Multiple intestinal mucosa gene expression profiles of UC patients in the Gene Expression Omnibus (GEO) database were downloaded. Unsupervised clustering methods were used to identify potential molecular subtypes based on ferroptosis and neutrophil gene sets. Multiple immune infiltration algorithms were used to evaluate the biological characteristics of the molecular subtypes. Machine learning identifies hub genes for molecular subtypes and analyses their diagnostic efficacy for UC and predictive performance for drug therapy. The relevant conclusions were verified by clinical samples and animal experiments. Four molecular subtypes were identified according to the ferroptosis and neutrophil gene sets: neutrophil, ferroptosis, mixed and quiescent. The subtypes have different biological characteristics and immune infiltration levels. Multiple machine learning methods jointly identified four hub genes (FTH1, AQP9, STEAP3 and STEAP4). Receiver operating characteristic (ROC) curve analysis revealed that the four hub genes could be used as diagnostic markers for UC. The clinical response profile data of infliximab treatment patients showed that AQP9 and STEPA4 were reliable predictors of infliximab treatment response. In human samples the AQP9 and STEAP4 protein were shown to be increased in UC intestinal samples. In animal experiments, the ferroptosis and neutrophil phenotype were confirmed. Dual analysis of ferroptosis and neutrophil gene expression revealed four subgroups of UC patients. The molecular subtype-associated hub genes can be used as diagnostic markers for UC and predict infliximab treatment response.

Unveiling Gender-Based Musculoskeletal Disorders in the Construction Industry: A Comprehensive Analysis

Without physically intensive building, modern infrastructure development would be impossible. Musculoskeletal diseases (MSDs) and other occupational health issues may arise in such a demanding environment. Construction workers often develop MSDs from repeated actions, uncomfortable postures, and heavy lifting. Musculoskeletal disorders may damage muscles, bones, tendons, ligaments, etc. The effect of MSDs is well known; occupational health studies increasingly include gender-specific aspects. Despite being in the minority, the number of female construction employees is growing. However, physiological variations and occupational activities and environments may provide distinct obstacles. Thus, identifying gender-specific MSDs in construction is essential for worker safety. This research proposes a gender-specific machine learning (ML)-based musculoskeletal disorder detection framework (GS-ML-MSD2F) in the construction industry. A simple random selection procedure chose 250 female and 250 male rebar workers with at least six months of experience for the dataset. In January and June 2023, face-to-face interviews and ergonomic evaluations were undertaken. The data were analyzed using different machine learning methods, and the effectiveness of the methods was studied. The data showed that 60% of female rebar workers had MSD symptoms. The lower back and shoulders accounted for 40% of cases. Multiple machine learning methods revealed two significant factors related to musculoskeletal disorders: lengthy working hours and uncomfortable postures, and long working hours had an adjusted odds ratio of 8.5%, whereas awkward posture had an adjusted odds ratio of 42.5%. These results emphasize the relevance of working hours and posture in MSD prevention for female rebar workers in the construction sector.

Analysis of the fatigue status of medical security personnel during the closed-loop period using multiple machine learning methods: a case study of the Beijing 2022 Olympic Winter Games

Using machine learning methods to analyze the fatigue status of medical security personnel and the factors influencing fatigue (such as BMI, gender, and wearing protective clothing working hours), with the goal of identifying the key factors contributing to fatigue. By validating the predicted outcomes, actionable and practical recommendations can be offered to enhance fatigue status, such as reducing wearing protective clothing working hours. A questionnaire was designed to assess the fatigue status of medical security personnel during the closed-loop period, aiming to capture information on fatigue experienced during work and disease recovery. The collected data was then preprocessed and used to determine the structural parameters for each machine learning algorithm. To evaluate the prediction performance of different models, the mean relative error (MRE) and goodness of fit (R2) between the true and predicted values were calculated. Furthermore, the importance rankings of various parameters in relation to fatigue status were determined using the RF feature importance analysis method. The fatigue status of medical security personnel during the closed-loop period was analyzed using multiple machine learning methods. The prediction performance of these methods was ranked from highest to lowest as follows: Gradient Boosting Regression (GBM) > Random Forest (RF) > Adaptive Boosting (AdaBoost) > K-Nearest Neighbors (KNN) > Support Vector Regression (SVR). Among these algorithms, four out of the five achieved good prediction results, with the GBM method performing the best. The five most critical parameters influencing fatigue status were identified as working hours in protective clothing, a customized symptom and disease score (CSDS), physical exercise, body mass index (BMI), and age, all of which had importance scores exceeding 0.06. Notably, working hours in protective clothing obtained the highest importance score of 0.54, making it the most critical factor impacting fatigue status. Fatigue is a prevalent and pressing issue among medical security personnel operating in closed-loop environments. In our investigation, we observed that the GBM method exhibited superior predictive performance in determining the fatigue status of medical security personnel during the closed-loop period, surpassing other machine learning techniques. Notably, our analysis identified several critical factors influencing the fatigue status of medical security personnel, including the duration of working hours in protective clothing, CSDS, and engagement in physical exercise. These findings shed light on the multifaceted nature of fatigue among healthcare workers and emphasize the importance of considering various contributing factors. To effectively alleviate fatigue, prudent management of working hours for security personnel, along with minimizing the duration of wearing protective clothing, proves to be promising strategies. Furthermore, promoting regular physical exercise among medical security personnel can significantly impact fatigue reduction. Additionally, the exploration of medication interventions and the adoption of innovative protective clothing options present potential avenues for mitigating fatigue. The insights derived from this study offer valuable guidance to management personnel involved in organizing large-scale events, enabling them to make informed decisions and implement targeted interventions to address fatigue among medical security personnel. In our upcoming research, we will further expand the fatigue dataset while considering higher precisionprediction algorithms, such as XGBoost model, ensemble model, etc., and explore their potential contributions to our research.

Bond strength between recycled aggregate concrete and rebar: Interpretable machine learning modeling approach for performance estimation and engineering design

Recycled aggregate concrete has a large potential for application as a sustainable building material. To assist in guiding the engineering design of this low-carbon material, the complex relationship between its bond strength to the rebar and various design parameters was modeled. Comparative studies of multiple machine learning methods have shown that the extreme random tree model demonstrates outstanding performance with root mean square error of 1.982, mean absolute error of 1.202 and coefficient of determination of 0.954. To determine the role of design parameters in performance prediction, importance analysis and sensitivity analyses for parameters were used to explore the complex relationship in the model. The results indicated that rebar type, water-cement ratio, anchorage length, and compressive strength are crucial factors affecting the bond strength between recycled aggregate concrete and rebar. In addition, to further demonstrate the reliability and accuracy of machine learning models, explicit models in some specifications were used for comparative analysis. Due to variability in influencing factors, limited test data, and variability in material properties, the explicit models have limited accuracy for bond strength between recycled aggregate concrete and rebar. Therefore, the interpretable modeling approach can provide an accurate estimate of the bond strength between recycled concrete aggregates and rebar and guide the engineering design of this structure.

Mendelian randomization and transcriptome analysis identified immune-related biomarkers for osteoarthritis.

The immune microenvironment assumes a significant role in the pathogenesis of osteoarthritis (OA). However, the current biomarkers for the diagnosis and treatment of OA are not satisfactory. Our study aims to identify new OA immune-related biomarkers to direct the prevention and treatment of OA using multi-omics data. The discovery dataset integrated the GSE89408 and GSE143514 datasets to identify biomarkers that were significantly associated with the OA immune microenvironment through multiple machine learning methods and weighted gene co-expression network analysis (WGCNA). The identified signature genes were confirmed using two independent validation datasets. We also performed a two-sample mendelian randomization (MR) study to generate causal relationships between biomarkers and OA using OA genome-wide association study (GWAS) summary data (cases n = 24,955, controls n = 378,169). Inverse-variance weighting (IVW) method was used as the main method of causal estimates. Sensitivity analyses were performed to assess the robustness and reliability of the IVW results. Three signature genes (FCER1G, HLA-DMB, and HHLA-DPA1) associated with the OA immune microenvironment were identified as having good diagnostic performances, which can be used as biomarkers. MR results showed increased levels of FCER1G (OR = 1.118, 95% CI 1.031-1.212, P = 0.041), HLA-DMB (OR = 1.057, 95% CI 1.045 -1.069, P = 1.11E-21) and HLA-DPA1 (OR = 1.030, 95% CI 1.005-1.056, P = 0.017) were causally and positively associated with the risk of developing OA. The present study identified the 3 potential immune-related biomarkers for OA, providing new perspectives for the prevention and treatment of OA. The MR study provides genetic support for the causal effects of the 3 biomarkers with OA and may provide new insights into the molecular mechanisms leading to the development of OA.

Comprehensive analysis of immunogenic cell death-related gene and construction of prediction model based on WGCNA and multiple machine learning in severe COVID-19

The death of coronavirus disease 2019 (COVID-19) is primarily due to from critically ill patients, especially from ARDS complications caused by SARS-CoV-2. Therefore, it is essential to contribute an in-depth understanding of the pathogenesis of the disease and to identify biomarkers for predicting critically ill patients at the molecular level. Immunogenic cell death (ICD), as a specific variant of regulatory cell death driven by stress, can induce adaptive immune responses against cell death antigens in the host. Studies have confirmed that both innate and adaptive immune pathways are involved in the pathogenesis of SARS-CoV-2 infection. However, the role of ICD in the pathogenesis of severe COVID-19 has rarely been explored. In this study, we systematically evaluated the role of ICD-related genes in COVID-19. We conducted consensus clustering, immune infiltration analysis, and functional enrichment analysis based on ICD differentially expressed genes. The results showed that immune infiltration characteristics were altered in severe and non-severe COVID-19. In addition, we used multiple machine learning methods to screen for five risk genes (KLF5, NSUN7, APH1B, GRB10 and CD4), which are used to predict COVID-19 severity. Finally, we constructed a nomogram to predict the risk of severe COVID-19 based on the classification and recognition model, and validated the model with external data sets. This study provides a valuable direction for the exploration of the pathogenesis and progress of COVID-19, and helps in the early identification of severe cases of COVID-19 to reduce mortality.

Construction and experimental validation of a novel ferroptosis-related gene signature for myelodysplastic syndromes.

Myelodysplastic syndromes (MDS) are clonal hematopoietic disorders characterized by morphological abnormalities and peripheral blood cytopenias, carrying a risk of progression to acute myeloid leukemia. Although ferroptosis is a promising target for MDS treatment, the specific roles of ferroptosis-related genes (FRGs) in MDS diagnosis have not been elucidated. MDS-related microarray data were obtained from the Gene Expression Omnibus database. A comprehensive analysis of FRG expression levels in patients with MDS and controls was conducted, followed by the use of multiple machine learning methods to establish prediction models. The predictive ability of the optimal model was evaluated using nomogram analysis and an external data set. Functional analysis was applied to explore the underlying mechanisms. The mRNA levels of the model genes were verified in MDS clinical samples by quantitative real-time polymerase chain reaction (qRT-PCR). The extreme gradient boosting model demonstrated the best performance, leading to the identification of a panel of six signature genes: SREBF1, PTPN6, PARP9, MAP3K11, MDM4, and EZH2. Receiver operating characteristic curves indicated that the model exhibited high accuracy in predicting MDS diagnosis, with area under the curve values of 0.989 and 0.962 for the training and validation cohorts, respectively. Functional analysis revealed significant associations between these genes and the infiltrating immune cells. The expression levels of these genes were successfully verified in MDS clinical samples. Our study is the first to identify a novel model using FRGs to predict the risk of developing MDS. FRGs may be implicated in MDS pathogenesis through immune-related pathways. These findings highlight the intricate correlation between ferroptosis and MDS, offering insights that may aid in identifying potential therapeutic targets for this debilitating disorder.

An Efficient Ensemble Machine Learning Model for Cardiovascular Disease Prediction Using Digital Health Records

Abstract: Detecting cardiovascular problems early is crucial for timely treatment. In our study, we employed machine learning to analyze a diverse set of information about individuals' lives and health, to predict cardiovascular disease. Ensuring data accuracy and addressing missing information were prioritized in our approach. Experimenting with different solo ML & ensemble ML methods, comprised of Random Forest and XGBoost with tuning, we achieved a notable 92% accuracy in identifying potential heart issues. Remarkably, combining multiple machine learning methods through ensemble learning proved even more effective than individual methods. Expanding our methodology to include Light GBM, Extra Tree, Decision Tree, SVM, Naive Bayes, QDA, & Adaboost enhanced the comprehensiveness of our analysis. Additionally, delving into ensemble learning methods such as bagging, boosting, tuning, & stacking further pushed the boundaries of predictive accuracy. In essence, our research outstands the potency of diverse ensemble machine-learning techniques and algorithms in early cardiovascular prediction. Ensemble methods, which combine different algorithms, emerged as powerful tools without relying on complex terminology.

Machine Learning Modeling of Wheel and Non-Wheel Path Longitudinal Cracking

Roads degrade over time due to various factors such as traffic loads, environmental conditions, and the quality of materials used. Significant investments have been poured into road construction globally, necessitating regular evaluations and the implementation of maintenance and rehabilitation (M&R) strategies to keep the infrastructure performing at a satisfactory level. The development and refinement of performance prediction models are essential for forecasting the condition of pavements, especially to address longitudinal cracking distress, a major issue in thick asphalt pavements. This research leverages multiple machine learning methods to create models predicting non-wheel path (NWP) and wheel path (WP) longitudinal cracking using data from the Long-Term Pavement Performance (LTPP) program. This study highlights the marked differences in distress conditions between WP and NWP, underscoring the importance of precise models that cater to their unique features. Aging trends for both types of cracking were identified through correlation analysis, showing an increase in WP cracking with age and a higher initial International Roughness Index (IRI) linked to NWP cracking. Factors such as material characteristics, kinematic viscosity, pavement thickness, air voids, particle size distribution, temperature, KESAL, and asphalt properties were found to significantly influence both WP and NWP cracking. The Exponential Gaussian Process Regression (GPR) emerged as the best model for NWP cracking, showcasing exceptional accuracy with the lowest RMSE of 89.11, MSE of 7940.72, and an impressive R-Squared of 0.63. For WP cracking, the Squared Exponential GPR model was most effective, with the lowest RMSE of 12.00, MSE of 143.93, and a high R-Squared of 0.62. The GPR models, with specific kernels for each cracking type, proved their adaptability and efficiency in various pavement scenarios. A comparative analysis highlighted the superiority of our new machine learning model, which achieved an R2 of 0.767, outperforming previous empirical models, demonstrating the strength and precision of our machine learning approach in predicting longitudinal cracking.

Fully automated learning and predict price of aquatic products in Taiwan wholesale markets using multiple machine learning and deep learning methods

The aquatic market price is an essential indicator for fishermen in making decisions regarding conveyance, fish-catching, and wholesale strategies. As a guide for the aquaculture industry, understanding market prices, the geographical distribution of fish prices, and the prediction of fish prices are vital. In Taiwanese fisheries, however, predicting aquatic prices is challenging due to their drastic fluctuations caused by tropical and subtropical climate variations, export/import fish quantity changes, and political and economic situation uncertainty. Therefore, instead of using a single machine learning model to address the price fluctuations, we propose a hybrid fish price prediction model integrating multiple machine learning and deep learning models to identify and predict fish price dynamics in various wholesale markets. The extensive experimental results on actual market data from the Taiwan Council of Agriculture (COA) show that the proposed models can achieve >90% accuracy in predicting future aquatic prices. Additionally, we developed a fully automated learning and prediction system architecture on the cloud, allowing us to acquire data automatically and fine-tune the AI models continuously to achieve better performance with long-term operability.

Hybrid Theory‐Guided Data Driven Framework for Calculating Irrigation Water Use of Three Staple Cereal Crops in China

AbstractCurrent irrigation water use (IWU) estimation methods confront uncertainties warranting further attention, primarily stemming from constraints within model structure and data quality. This study proposes a hybrid framework that integrates multiple machine learning (ML) methods with theory‐guided strategies to calculate IWU for three principal cereal crops within the Chinese agricultural landscape. We generated high resolution time series data sets of evapotranspiration and surface soil moisture (SM) using remote sensing resources. ML techniques, along with the Bayesian three‐cornered hat ensemble, were employed to drive multiple remote sensing‐derived data sets in IWU calculation. We applied two theory‐guided mechanisms to quantify irrigation signals: first, converting original SM values into logarithmic terms, and second, extracting process‐based SM residuals. Proposed framework has been validated at 12 field stations across China, yielding coefficient of determination (R2) ranging from 0.54 to 0.70, and root mean square error (RMSE) spanning 278–335 mm/yr. Our framework demonstrates considerable strength in IWU estimation when compared to reported IWU values form 341 cities across China. Specifically, for rice, wheat, and maize, the R2 values range from 0.78 to 0.83, 0.68 to 0.76, and 0.53 to 0.64, respectively, with corresponding RMSE measuring 0.22–0.25, 0.10–0.12, and 0.11–0.13 km3/yr, respectively. These findings highlight the effectiveness of theory‐guided strategies in discerning irrigation‐related information, thereby improving overall model performance. Attention should be directed toward the uncertainties in evapotranspiration and precipitation products on model performance, which remained modest, with a relative change of less than 5%.

Classification and Explanation of Iron Deficiency Anemia from Complete Blood Count Data Using Machine Learning

Background: Currently, discriminating Iron Deficiency Anemia (IDA) from other anemia requires an expensive test (serum ferritin). Complete Blood Count (CBC) tests are less costly and more widely available. Machine learning models have not yet been applied to discriminating IDA but do well for similar tasks. Methods: We constructed multiple machine learning methods to classify IDA from CBC data using a US NHANES dataset of over 19,000 instances, calculating accuracy, precision, recall, and precision AUC (PR AUC). We validated the results using an unseen dataset from Kenya, using the same model. We calculated ranked feature importance to explain the global behavior of the model. Results: Our model classifies IDA with a PR AUC of 0.87 and recall/sensitivity of 0.98 and 0.89 for the original dataset and an unseen Kenya dataset, respectively. The explanations indicate that low blood level of hemoglobin, higher age, and higher Red Blood Cell distribution width were most critical. We also found that optimization made only minor changes to the explanations and that the features used remained consistent with professional practice. Conclusions: The overall high performance and consistency of the results suggest that the approach would be acceptable to health professionals and would support enhancements to current automated CBC analyzers.

Kinome-Wide Virtual Screening by Multi-Task Deep Learning.

Deep learning is a machine learning technique to model high-level abstractions in data by utilizing a graph composed of multiple processing layers that experience various linear and non-linear transformations. This technique has been shown to perform well for applications in drug discovery, utilizing structural features of small molecules to predict activity. Here, we report a large-scale study to predict the activity of small molecules across the human kinome-a major family of drug targets, particularly in anti-cancer agents. While small-molecule kinase inhibitors exhibit impressive clinical efficacy in several different diseases, resistance often arises through adaptive kinome reprogramming or subpopulation diversity. Polypharmacology and combination therapies offer potential therapeutic strategies for patients with resistant diseases. Their development would benefit from a more comprehensive and dense knowledge of small-molecule inhibition across the human kinome. Leveraging over 650,000 bioactivity annotations for more than 300,000 small molecules, we evaluated multiple machine learning methods to predict the small-molecule inhibition of 342 kinases across the human kinome. Our results demonstrated that multi-task deep neural networks outperformed classical single-task methods, offering the potential for conducting large-scale virtual screening, predicting activity profiles, and bridging the gaps in the available data.

Investigation of bacterial DNA gyrase Inhibitor classification models and structural requirements utilizing multiple machine learning methods.

Infections from multidrug-resistant (MDR) bacteria have emerged as a paramount global health concern, and the therapeutic effectiveness of current treatments is swiftly diminishing. An urgent need exists to explore innovative strategies for countering drug-resistant bacteria. Bacterial DNA gyrase, functioning as an ATP-dependent enzyme, plays a pivotal role in the intricate processes of transcription, replication, and chromosome segregation within bacterial DNA. This renders it a prime target for the development of innovative antibacterial agents. However, the experimental identification of bacterial DNA gyrase inhibitors faces multifaceted challenges due to current methodological constraints. Recognizing its significance, this study developed 56 computational models designed for predicting bacterial DNA gyrase inhibitors. These models employed seven distinct molecular fingerprints and eight machine learning algorithms. Among these models, Model_2D, created using KlekotaRoth fingerprints and the SVM algorithm, stands out as the most robust performer (ACC = 0.86, MCC = 0.63, G-mean = 0.82). Moreover, given the limited exploration of structural fragments required for DNA Gyrase B inhibitors, crucial structural fingerprints influencing DNA Gyrase B inhibitors were identified through Bayesian classification. Subsequently, we conducted molecular docking to reveal the binding modes between these crucial structural fingerprints and the active site of DNA gyrase B. In conclusion, the present study aimed to develop the optimal classification model for bacterial DNA gyrase inhibitors, offering invaluable support to medicinal chemists creating innovative DNA gyrase inhibitors.

Predicting Urgent Dialysis at Ambulance Transport to the Emergency Department Using Machine Learning Methods.

Hemodialysis patients frequently require ambulance transport to the hospital for dialysis. Some patients require urgent dialysis (UD) within 24 hours of transport to hospital to avoid morbidity and mortality. UD is not available in all hospitals; therefore, predicting patients who need UD prior to hospital transport can help paramedics with destination planning. In this paper, we developed machine learning models for paramedics to predict whether a patient needs UD based on patient characteristics available at the time of ambulance transport. This paper presented a study based on ambulance data collected in Halifax, Canada. Given that relatively few patients need UD, a class imbalance problem is addressed by up-sampling methods and prediction models are developed using multiple machine learning methods. The achieved prediction scores are F1-score=0.76, sensitivity=0.76, and specificity=0.97, confirming that models can predict UD with limited patient characteristics.

Comparison of Multiple Machine Learning Methods for Correcting Groundwater Levels Predicted by Physics-Based Models

Accurate groundwater level (GWL) prediction is crucial in groundwater resource management. Currently, it relies mainly on physics-based models for prediction and quantitative analysis. However, physics-based models used for prediction often have errors in structure, parameters, and data, resulting in inaccurate GWL predictions. In this study, machine learning algorithms were used to correct the prediction errors of physics-based models. First, a MODFLOW groundwater flow model was created for the Hutuo River alluvial fan in the North China Plain. Then, using the observed GWLs from 10 monitoring wells located in the upper, middle, and lower parts of the alluvial fan as the test standard, three algorithms—random forest (RF), extreme gradient boosting (XGBoost), and long short-term memory (LSTM)—were compared for their abilities to correct MODFLOW’s predicted GWLs of these 10 wells under two sets of feature variables. The results show that the RF and XGBoost algorithms are not suitable for correcting predicted GWLs that exhibit continuous rising or falling trends, but the LSTM algorithm has the ability to correct them. During the prediction period, the LSTM2 model, which incorporates additional source–sink feature variables based on MODFLOW’s predicted GWLs, can improve the Pearson correlation coefficient (PR) for 80% of wells, with a maximum increase of 1.26 and a minimum increase of 0.02, and can reduce the root mean square error (RMSE) for 100% of the wells with a maximum decrease of 1.59 m and a minimum decrease of 0.17 m. And it also outperforms the MODFLOW model in capturing the long-term trends and short-term seasonal fluctuations of GWLs. However, the correction effect of the LSTM1 model (using only MODFLOW’s predicted GWLs as a feature variable) is inferior to that of the LSTM2 model, indicating that multiple feature variables are superior to a single feature variable. Temporally and spatially, the greater the prediction error of the MODFLOW model, the larger the correction magnitude of the LSTM2 model.

Individual Tree-Level Monitoring of Pest Infestation Combining Airborne Thermal Imagery and Light Detection and Ranging

The infestation of pine shoot beetles (Tomicus spp.) in the forests of Southwestern China has inflicted serious ecological damages to the environment, causing significant economic losses. Therefore, accurate and practical approaches to detect pest infestation have become an urgent necessity to mitigate these harmful consequences. In this study, we explored the efficiency of thermal infrared (TIR) technology in capturing changes in canopy surface temperature (CST) and monitoring forest health at the scale of individual tree crowns. We combined data collected from TIR imagery and light detection and ranging (LiDAR) using unmanned airborne vehicles (UAVs) to estimate the shoot damage ratio (SDR), which is a representative parameter of the damage degree caused by forest infestation. We compared multiple machine learning methods for data analysis, including random forest (RF), partial least squares regression (PLSR), and support vector machine (SVM), to determine the optimal regression model for assessing SDR at the crown scale. Our findings showed that a combination of LiDAR metrics and CST presents the highest accuracy in estimating SDR using the RF model (R2 = 0.7914, RMSE = 15.5685). Our method enables the accurate remote monitoring of forest health and is expected to provide a novel approach for controlling pest infestation, minimizing the associated damages caused.