Gradient Boosting Research Articles

Utilization finite element and machine learning methods to investigation the axial compressive behavior of elliptical FRP-confined concrete columns

Nowadays, elliptical sections are among the geometric shapes that are becoming more and more popular in the architecture world. Finite element analysis (FEA) software, such as ABAQUS, is used to simulate elliptical concrete columns confined with fiber reinforced polymer (FRP) jackets based on experimental data published in the literature. The validity of the finite element modeling (FEM) approach was confirmed by comparing it to experimental data obtained from 45 specimens in existing studies, demonstrating a favorable agreement. Additionally, a parametric study was also carried out, which included simulating 40 more specimens, resulting in a total of 85 specimens for analysis. The effect of various test variables such as sectional aspect ratio, the amount of FRP layers, and concrete strength on the elliptical column’s behavior is investigated. The axial compressive strength is predicted using current models from the literature. The outcomes revealed that the higher unconfined concrete strength decreased FRP confinement efficacy. Insufficient confinement with post-peak softening is more likely in FRP-confined high strength concrete columns, especially if the confinement was not stiff enough. Also, by adding more FRP layers, the stress and strain capabilities of elliptical concrete columns confined with FRP composites are improved. Physical experiments require a significant investment of time and resources, but they can produce valuable results. Finite element analysis, on the other hand, depends heavily on the modeler's competence and can minimize the number of experiments required by employing computer simulation, but it also requires high computer settings. In recent years, there has been a major advancement in the usage of machine learning (ML) algorithms in the applications of FRP composites with different concrete components. Taking this into consideration, this investigation is extended to estimate the compressive strength of elliptical FRP-confined columns using four tree-based ML algorithms including Decision Tree, Random Forest, Gradient Boosting and XGBoost. The XGBoost model achieved the highest predictive accuracy with the R2 values of 0.95 for both the compressive strength and axial strain targets.

Application of tongue image characteristics and oral-gut microbiota in predicting pre-diabetes and type 2 diabetes with machine learning

BackgroundThis study aimed to characterize the oral and gut microbiota in prediabetes mellitus (Pre-DM) and type 2 diabetes mellitus (T2DM) patients while exploring the association between tongue manifestations and the oral-gut microbiota axis in diabetes progression.MethodsParticipants included 30 Pre-DM patients, 37 individuals with T2DM, and 28 healthy controls. Tongue images and oral/fecal samples were analyzed using image processing and 16S rRNA sequencing. Machine learning techniques, including support vector machine (SVM), random forest, gradient boosting, adaptive boosting, and K-nearest neighbors, were applied to integrate tongue image data with microbiota profiles to construct predictive models for Pre-DM and T2DM classification.ResultsSignificant shifts in tongue characteristics were identified during the progression from Pre-DM to T2DM. Elevated Firmicutes levels along the oral-gut axis were associated with white greasy fur, indicative of underlying metabolic changes. An SVM-based predictive model demonstrated an accuracy of 78.9%, with an AUC of 86.9%. Notably, tongue image parameters (TB-a, perALL) and specific microbiota (Escherichia, Porphyromonas-A) emerged as prominent diagnostic markers for Pre-DM and T2DM.ConclusionThe integration of tongue diagnosis with microbiome analysis reveals distinct tongue features and microbial markers. This approach significantly improves the diagnostic capability for Pre-DM and T2DM.

Enhancing Forest Canopy Height Mapping in Kaziranga National Park, Assam, by Integrating LISS IV and SAR data with GEDI LiDAR data Using Machine Learning

Abstract. This study investigates the integration of spaceborne Global Ecosystem Dynamics Investigation (GEDI) LiDAR data with optical imagery from Linear Imaging Self Scanning Sensor (LISS IV), Sentinel-1 Synthetic Aperture Radar (SAR) and PALSAR data for continuous forest canopy height mapping in Kaziranga National Park (KNP), Assam for the years 2018 and 2022. Four machine learning models were trained and evaluated to assess the predictive ability of LISS IV data in conjunction with SAR variables. The mean canopy height was measured at approximately 8.58 m in 2018, which increased to about 9.07m in 2022. Results reveal Extreme Gradient Boosting (XGB) as the top-performing model, achieving an RMSE of 5.47m and an R2 of 0.55. In comparison, Random Forest (RF), Support Vector Machine (SVM), and k-Nearest Neighbors (kNN) achieved RMSE values of 5.49, 5.51, and 5.73, respectively. Analysis shows a prevalent occurrence of canopy heights below 5 meters in KNP (more than 35% of the area), while taller canopies beyond 20m can be found in less than 5% of the area. This finding underscores the importance of integrating satellite data and machine learning and highlights the novel application of LISS IV data in enhancing canopy height mapping. Furthermore, it represents the first comprehensive attempt to map canopy height in KNP, laying the groundwork for further research on biomass assessment and carbon sequestration in this vital biodiversity hotspot. Overall, the study highlights the potential of leveraging advanced remote sensing technologies and machine learning approaches for improved understanding and management of forest ecosystems.

Leveraging SAR and Optical Remote Sensing for Enhanced Biomass Estimation in the Amazon with Random Forest and XGBoost Models

Abstract. This study addresses the challenge of estimating above-ground biomass (AGB) in the Amazon rainforest by developing a reference geographical database, which provides the ground truth, and comparing the relative importance of using Synthetic Aperture Radar (SAR) and optical remote sensing data to automatically infer AGB. In the experiments reported in this article, we assessed how those two remote sensing data sources impact the accuracy of AGB estimates produced by regression models built with Random Forest (RF) and Extreme Gradient Boosting (XGBoost). The research involved compiling a comprehensive database from many available forest inventories, integrating parcel- and tree-level data to enable precise biomass estimation. The methodology included setting up a spatial data analysis environment, standardizing data, and implementing an experimental protocol with feature selection and leave-one-out cross-validation. The results demonstrate that both kinds of data, i.e., SAR and optical, and their combination can be used for estimating AGB, providing valuable insights for forest management and climate change mitigation efforts. The reference database is available upon request to the corresponding authors.

A fast prediction method of fatigue life for crane structure based on Stacking ensemble learning model

To quickly obtain the fatigue life of cranes in service, the metal structure that determines the crane life is anchored. Meanwhile, the fast prediction method of fatigue life of crane metal structures based on the Stacking ensemble learning model is proposed. Firstly, in line with the structural stress method, the global rough model of the metal structure is established by the co-simulation technology to obtain the fatigue damage regions of the structure. The local fine model is constructed by local cutting and boundary condition transplantation to determine the critical weld at the failure regions. Secondly, through weld definition, equivalent structural stress acquisition, and fatigue life calculation, the sample data set with lifting load and trolley running position as input and fatigue life cycle times as output is constructed. Then, the Stacking integrated learning model combining gradient boosting, ridge regression, Extra Trees, and linear is built. On this basis, combined with the Miner theory, the rapid prediction of crane fatigue life is realized. Finally, the proposed method is applied to the QD40t × 22.5 m × 9 m general bridge crane. The results show that the life sample set constructed by the structural stress method is more accurate and reasonable than the nominal, hot spot, and fracture mechanics methods. The life prediction results of the Stacking integration model were improved by 6.3 to 49.2% compared to the single model. The method has theoretical and practical significance in reducing accidents and ensuring the safe operation of cranes.

Predicting the Multiphotonic Absorption in Graphene by Machine Learning

This study analyzes the nonlinear optical properties exhibited by graphene, focusing on the nonlinear absorption coefficient and the nonlinear refractive index. The evaluation was conducted using the Z-scan technique with a 532 nm wavelength laser at various intensities. The nonlinear optical absorption and the nonlinear optical refractive index were measured. Four machine learning models, including linear regression, decision trees, random forests, and gradient boosting regression, were trained to analyze how the nonlinear optical absorption coefficient varies with variables such as spot radius, maximum energy, and normalized minimum transmission. The models were trained with synthetic data and subsequently validated with experimental data. Decision tree-based models, such as random forests and gradient boosting regression, demonstrated superior performance compared to linear regression, especially in terms of mean squared error. This work provides a detailed assessment of the nonlinear optical properties of graphene and highlights the effectiveness of machine learning methods in this context.

Can neotectonic faults influence soil air radon levels in the Upper Rhine Graben? An exploratory machine learning assessment

Radon (Rn) is a naturally occurring radioactive gas that poses a significant lung cancer risk. Subsurface fault zones can act as pathways for fluid and gas migration, potentially amplifying Rn accumulation. This study investigates the impact of fault zones on Rn concentrations within a 25 km2 area in the Northern Upper Rhine Graben, Germany — a region with available detailed geophysical exploration data and active neotectonic faulting. We conducted 597 Rn soil air measurements along precisely located fault zones, integrating a comprehensive range of environmental parameters. Utilizing the advanced machine learning model eXtreme Gradient Boosting (XGBoost) in conjunction with SHapley Additive exPlanations (SHAP) values, we dissected the influence of soil types, environmental factors, and proximity to fault zones on soil air Rn concentrations at a 1-meter depth. Our results reveal that clay-rich soils and cumulative 30-day precipitation are the primary drivers of elevated Rn levels. Proximity to fault zones also significantly influences Rn concentrations, though its impact is less pronounced than the factors mentioned above. Additionally, environmental factors such as wind speed, air pressure, and temperature exhibited lesser effects on Rn levels. The negligible influence of measuring devices and operating personnel increases confidence in data integrity in extensive environmental studies. This study demonstrates the effectiveness of integrating XGBoost with SHAP values to identify and quantify key factors influencing Rn concentrations. By providing a robust framework for enhancing Rn prediction models through machine learning, our findings contribute to improved risk assessments and mitigation strategies, thereby advancing public health and environmental management.

Common laboratory results-based artificial intelligence analysis achieves accurate classification of plasma cell dyscrasias

Background Plasma cell dyscrasias encompass a diverse set of disorders, where early and precise diagnosis is essential for optimizing patient outcomes. Despite advancements, current diagnostic methodologies remain underutilized in applying artificial intelligence (AI) to routine laboratory data. This study seeks to construct an AI-driven model leveraging standard laboratory parameters to enhance diagnostic accuracy and classification efficiency in plasma cell dyscrasias. Methods Data from 1,188 participants (609 with plasma cell dyscrasias and 579 controls) collected between 2018 and 2023 were analyzed. Initial variable selection employed Kruskal-Wallis and Wilcoxon tests, followed by dimensionality reduction and variable prioritization using the Shapley Additive Explanations (SHAP) approach. Nine pivotal variables were identified, including hemoglobin (HGB), serum creatinine, and β2-microglobulin. Utilizing these, four machine learning models (gradient boosting decision tree (GBDT), support vector machine (SVM), deep neural network (DNN), and decision tree (DT) were developed and evaluated, with performance metrics such as accuracy, recall, and area under the curve (AUC) assessed through 5-fold cross-validation. A subtype classification model was also developed, analyzing data from 380 cases to classify disorders such as multiple myeloma (MM) and monoclonal gammopathy of undetermined significance (MGUS). Results 1. Variable selection: The SHAP method pinpointed nine critical variables, including hemoglobin (HGB), serum creatinine, erythrocyte sedimentation rate (ESR), and β2-microglobulin. 2. Diagnostic model performance: The GBDT model exhibited superior diagnostic performance for plasma cell dyscrasias, achieving 93.5% accuracy, 98.1% recall, and an AUC of 0.987. External validation reinforced its robustness, with 100% accuracy and an F1 score of 98.5%. 3. Subtype Classification: The DNN model excelled in classifying multiple myeloma, MGUS, and light-chain myeloma, demonstrating sensitivity and specificity above 90% across all subtypes. Conclusions AI models based on routine laboratory results significantly enhance the precision of diagnosing and classifying plasma cell dyscrasias, presenting a promising avenue for early detection and individualized treatment strategies.

Predictors and Profile of Severe Infectious Complications in Multiple Myeloma Patients Treated with Daratumumab-Based Regimens: A Machine Learning Model for Pneumonia Risk

Background: Daratumumab (Dara) is the first monoclonal antibody introduced into clinical practice to treat multiple myeloma (MM). It currently forms the backbone of therapy regimens in both newly diagnosed (ND) and relapsed/refractory (RR) patients. However, previous reports indicated an increased risk of infectious complications (ICs) during Dara-based treatment. In this study, we aimed to determine the profile of ICs in MM patients treated with Dara-based regimens and establish predictors of their occurrence. Methods: This retrospective, real-life study included MM patients treated with Dara-based regimens between July 2019 and March 2024 at our institution. Infectious events were evaluated using the Terminology Criteria for Adverse Events (CTCAE) version 5.0. Results: The study group consisted of a total of 139 patients, including 49 NDMM and 90 RRMM. In the RR setting, the majority (60.0%) of patients received the Dara, bortezomib, and dexamethasone (DVd) regimen, whereas ND patients were predominantly (98%) treated with the Dara, bortezomib, thalidomide, and dexamethasone (DVTd) regimen. Overall, 55 patients (39.6%) experienced ICs. The most common IC was pneumonia (37.5%), followed by upper respiratory tract infections (26.8%). Finally, twenty-five patients had severe ICs (grade ≥ 3) and required hospitalization, and eight patients died due to ICs. In the final multivariable model adjusted for setting (ND/RR) and age, hemoglobin level (OR 0.77, 95% CI: 0.61–0.96, p = 0.0037), and Eastern Cooperative Oncology Group (ECOG) >1 (OR 4.46, 95% CI: 1.63–12.26, p = 0.0037) were significant factors influencing severe IC occurrence. Additionally, we developed predictive models using the J48 decision tree, gradient boosting, and random forest algorithms. After conducting 10-fold cross-validation, these models demonstrated strong performance in predicting the occurrence of pneumonia during treatment with daratumumab-based regimens. Conclusions: Simple clinical and laboratory assessments, including hemoglobin level and ECOG scale, can be valuable in identifying patients vulnerable to infections during Dara-based regimens, facilitating personalized prophylactic strategies.

Prediction of bearing capacity of geogrid-reinforced sand over stone columns in soft clay

It is a challenge to estimate the ultimate bearing capacity (qrs) of a geogrid-reinforced sandy bed on vertical stone columns in soft clay due to their complex geometry and uncertain parameters. Predicting qrs can be costly and challenging, so creating an accurate prediction model is important for real-world applications. The research aims to use ensemble techniques like K Nearest Neighbors (KNN), Random Forest (RF), and Extreme Gradient Boosting (XGBoost) to develop a model for estimating the bearing capacity of geogrid-reinforced stone columns (GRBS) on a sandy bed with vertical stone columns in soft clay. A dataset of 245 experimental observations is used to train the models. The present study highlights the potential use of the XGBoost model as a useful tool that can assist in predicting the bearing capacity. The results of the study reveal that this model has performed well in predicting the bearing capacity with high correlation coefficients of 0.9947. Furthermore; a SHAP dependency analysis was conducted to ascertain the significance of each parameter.

Development of New Electricity System Marginal Price Forecasting Models Using Statistical and Artificial Intelligence Methods

The System Marginal Price (SMP) is the cost of the last unit of electricity supplied to the grid, reflecting the supply–demand equilibrium and serving as a key indicator of market conditions. Accurate SMP forecasting is essential for ensuring market stability and economic efficiency. This study addresses the challenges of SMP prediction in Turkey by proposing a comprehensive forecasting framework that integrates machine learning, deep learning, and statistical models. Advanced feature selection techniques, such as Minimum Redundancy Maximum Relevance (mRMR) and Maximum Likelihood Feature Selector (MLFS), are employed to refine model inputs. The framework incorporates time series methods like Multilayer Perceptron (MLP), Long Short-Term Memory (LSTM), Bidirectional LSTM (Bi-LSTM), and Convolutional LSTM (ConvLSTM) to capture complex temporal patterns, alongside models such as Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), and Extreme Learning Machine (ELM) for modeling non-linear relationships. Model performance was evaluated using the Mean Absolute Percentage Error (MAPE) across regular weekdays, weekends, and public holidays. XGBoost combined with MLFS consistently achieved the lowest MAPE values, demonstrating exceptional accuracy and robustness. Among all of the models, XGBoost combined with MLFS consistently achieved the lowest MAPE values, demonstrating superior accuracy and robustness. The results highlight the inadequacy of traditional models like ARIMA and SARIMA in capturing non-linear and highly volatile patterns, reinforcing the necessity of using advanced techniques for effective SMP forecasting. Overall, this study presents a novel and comprehensive approach tailored for complex electricity markets, significantly enhancing predictive reliability by incorporating economic indicators and sophisticated feature selection methods.

A Study on the Classification of Shrubs and Grasses on the Tibetan Plateau Based on Unmanned Aerial Vehicle Multispectral Imagery

The ecosystem of the Qinghai–Tibet Plateau is highly fragile due to its unique geographical conditions, with vegetation playing a crucial role in maintaining ecological balance. Thus, accurately monitoring the distribution of vegetation in the plateau region is of paramount importance. This study employs UAV multispectral imagery in combination with four machine-learning models—Support Vector Machine (SVM), Decision Tree (DT), Extreme Gradient Boosting (XGBoost), and Random Forest (RF)—to investigate the impact of different features and their combinations on the fine classification of shrubs and grasses on the Qinghai–Tibet Plateau, including Salix psammophila, Populus simonii Carrière, Kobresia tibetica, and Kobresia pygmaea. The results indicate that near-infrared spectral information can improve classification accuracy, with improvements of 5.21%, 1.65%, 6.64%, and 5.03% for Salix psammophila, Populus simonii Carrière, Kobresia tibetica, and Kobresia pygmaea, respectively. Feature selection effectively reduces redundant information and enhances model classification accuracy, with all four machine-learning models achieving the best performance on the optimized feature set. Furthermore, the RF model performs best on the optimized feature set, achieving an overall accuracy (OA) of 95.32% and a kappa coefficient of 0.94. This study provides important scientific support for the fine classification and ecological monitoring of plateau vegetation.

Phosphate-solubilizing fungus (PSF) - mediated phosphorous solubilization and validation through Artificial intelligence computation.

Phosphate-solubilizing fungus (PSF) strain alaromyces funiculosus was investigated for phosphorus solubilization, utilizing a range of pH levels and phosphate sources, followed by data confirmation through artificial intelligence modeling. T. funiculosus strain was exposed to five different phosphate sources [Ca3(PO4)2, FePO4, CaHPO4, AlPO4, and phytin] at different pH levels (4.5, 5.5, 6.5, 7.0, and 7.5). ANOVA, Pareto charts, and normal plots were used for analyzing the data. Artificial intelligence-based multilayer perceptron (MLP), random forest (RF) and extreme gradient boosting (XGBoost) models were used for data validation and prediction. Five-fold more phosphate (P) solubility by T. funiculosus was registered as compared to the control. The maximum soluble P was found at pH 4.5 (318324 ppb) and CaHPO4 (444045 ppb). Combination of phytin × 4.5 pH yielded the highest dissolved phosphorus (1537988 ppb), followed by 127458 ppb from the control × 4.5 pH. Pareto chart and normal plot analysis showedthe negative impact of pH (B), pH × F/C (fungus/control) × P-Source (ABC), and F/C (A) factor. Whereas pH × P-Source (AC) and P-Source (C) has positive impact on P solubility. The maximum R2 scores showed the order of RF (0.944) > MLP (0.938) > XGBoost (0.899). T. funiculosus strain has a grain potential for sustainable use for different types of phosphate sources. Application AI/ML models based on different performance metrics predicted the validated the attained results. In future research, it is recommended to check the efficacy of developed strategy under field conditions and to check the impact on soil and plant.

Precision medicine in hepatology: harnessing IoT and machine learning for personalized liver disease stage prediction

<span>In this research, we used a dataset from Siksha ‘O’ Anusandhan (S’O’A) University Medical Laboratory containing 6,780 samples collected manually and through internet of things (IoT) sensor sources from 6,780 patients to perform a thorough investigation into liver disease stage prediction. The dataset was carefully cleaned before being sent to the machine learning pipeline. We utilised a range of machine learning models, such as Naïve Bayes (NB), sequential minimal optimisation (SMO), K-STAR, random forest (RF), and multi-class classification (MCC), using Python to predict the stages of liver disease. The results of our simulations demonstrated how well the SMO model performed in comparison to other models. We then expanded our analysis using different machine learning boosting models with SMO as the base model: adaptive boosting (AdaBoost), gradient boost, extreme gradient boosting (XGBoost), CatBoost, and light gradient boosting model (LightGBM). Surprisingly, gradient boost proved to be the most successful, producing an astounding 96% accuracy. A closer look at the data showed that when AdaBoost was combined with the SMO base model, the accuracy results were 94.10%, XGBoost 90%, CatBoost 92%, and LightGBM 94%. These results highlight the effectiveness of proposed model i.e. gradient boosting in improving the prediction of liver disease stage and provide insightful information for improving clinical decision support systems in the field of medical diagnostics.</span>

Machine learning - assisted prediction of yield strength in irradiated type 316 stainless steels

The investigation into irradiation hardening and embrittlement is of critical importance in nuclear material subject. This work explores the applications of machine learning (ML) to predict the yield strength of irradiated type 316 stainless steels. A dataset comprising 354 samples is compiled through an extensive review of prior experimental studies. Each sample has 23 potentially influential features. Five distinct machine learning models are trained and evaluated. Among these models, the Gradient Boosting (GB) model demonstrates superior prediction performance and robust stability. The prominent factors identified by the GB model are in reasonable agreement with established knowledge regarding the determinants of yield strength in irradiated type 316 stainless steels. These findings provide critical insights into the mechanical properties of irradiated type 316 stainless steels.

Machine learning prediction of stalk lignin content using Fourier transform infrared spectroscopy in large scale maize germplasm

Lignin has been recognized as a major factor contributing to lignocellulosic recalcitrance in biofuel production and attracted attentions as a high-value product in the biorefinery field. As the traditional wet chemical methods for detecting lignin content are labor-intensive, time-consuming and environment-toxic, it is an urgent need to develop high-throughput and environment-friendly techniques for large-scale crop germplasms screening. In this study, we conducted a Fourier transform infrared (FTIR) assay on 150 maize germplasms with a diverse lignin composition to build predictive models for lignin content in maize stalk. Principal component analysis (PCA) was applied to the FTIR spectra for use as model inputs. Classification and advanced gradient boosting machine (GBM) algorithms demonstrated higher predictive accuracy (0.82–0.96) compared to traditional linear and regularization algorithms (0.03–0.04) in the training set. Notably, two optimal models, built using the extreme gradient boosting (XGBoost) and light gradient boosting machine (LightGBM) algorithms, achieved R2 values of over 0.91 in the training set and over 0.82 in the test set. Overall, the combination of FTIR and machine learning (ML) algorithms offers a high-throughput and efficient method for predicting lignin content. This approach holds significant potential for genetic breeding and the effective utilization of maize in industrial production.

Exploring blood-brain barrier passage using atomic weighted vector and machine learning.

This study investigates the potential of leveraging molecular properties, as determined by MD-LOVIs software and machine learning techniques, to predict the ability of compounds to cross the blood-brain barrier (BBB). Accurate prediction of BBB permeation is critical for the development of central nervous system (CNS) drugs. The study applies various machine learning models, including both classification and regression techniques, to predict BBB passage and molecular activity. Notably, classification models such as GBM-AWV (accuracy = 0.801), GLM-CN (accuracy = 0.808), SVMPoly-means (accuracy = 0.980), SVMPoly-AC (accuracy = 0.980), SVMPoly-MI_TI_SI (accuracy = 0.900), SVMPoly-GI (accuracy = 0.900), RF-means (accuracy = 0.870), and GLM-means (accuracy = 0.818) demonstrate high accuracy in predicting BBB passage. In contrast, regression models like ES-RLM-AG (R2 = 0.902), IB-IBK (R2 = 0.82), IB-Kstar (R2 = 0.834), IB-MLP (R2 = 0.843), and DRF-AWV (R2 = 0.810) exhibit strong performance in predicting molecular activity. The results show that classification models like GBM-AWV, GLM-CN, and SVMPoly variants, as well as regression models like ES-RLM-AG and IB-MLP, achieve high performance, demonstrating the effectiveness of machine learning in predicting BBB permeability. The computational methods employed in this study include the MD-LOVIs software for generating molecular descriptors and several machine learning algorithms, including gradient boosting machines (GBM), generalized linear models (GLM), support vector machines (SVM) with polynomial kernels, random forests (RF), ensemble regression models, and instance-based learning algorithms. These models were trained and validated using various datasets to predict BBB passage and molecular activity, with the performance metrics reported for each model. Standard computational techniques were employed throughout, ensuring the reliability of the predictions.

Leveraging machine learning for preoperative prediction of supramaximal ablation in laser interstitial thermal therapy for brain tumors.

Maximizing safe resection in neuro-oncology has become paramount to improving patient survival and outcomes. Laser interstitial thermal therapy (LITT) offers similar survival benefits to traditional resection, alongside shorter hospital stays and faster recovery times. The extent of ablation (EOA) achieved using LITT is linked to patient outcomes, with greater EOA correlating with improved outcomes. However, the preoperative predictors for achieving supramaximal ablation (EOA ≥ 100%) are not well understood. By leveraging machine learning (ML) techniques, this study aimed to identify these predictors to enhance patient selection and therefore outcomes. The objective was to explore preoperative predictors for supramaximal EOA using ML in patients with glioblastoma. A retrospective study was conducted on the medical records of 254 patients undergoing LITT from 2013 to 2023 at a single tertiary center. Cohort criteria included age ≥ 18 years, diagnosis of glioblastoma, single-trajectory ablation, and a complete dataset. The study assessed preoperative clinical and radiographic factors, using EOA ≥ 100% as the endpoint. Five ML models were used: logistic regression, random forest (RF), gradient boosting, Gaussian naive Bayes, and support vector machine. Training and testing cohorts were subsequently assessed across ML models with fivefold cross-validation. Models were optimized using hyperparameter tuning. Performance was primarily quantified using the area under the curve (AUC) of the receiver operating characteristic curve. The final cohort consisted of 72 patients. Among the ML models, RF achieved the highest AUC (mean ± SD 0.94 ± 0.06). The leading models identified that lower preoperative volume, history of prior radiation therapy, history of prior craniotomy, preoperative neurological deficits, history of preoperative seizures, and distance from intracranial heat sinks were predictive of successful ablations in patients. Additionally, RF had the best mean metrics: accuracy 0.88, precision 0.87, specificity 0.87, and sensitivity 0.89. This is the first study to investigate the role of ML for optimizing ablation volumes in LITT. These ML models suggest that low preoperative volumes, previous craniotomy, previous radiation therapy, no previous neurological deficits, larger catheter-heat sink distance, and the presence of preoperative seizures are important prognostic factors for predicting successful supramaximal ablations with LITT.

A Tropical Cyclone Risk Prediction Framework using Flood Susceptibility and Tree-based Machine Learning Models: County-level Direct Economic Loss Prediction in Guangdong Province

Tropical cyclones (TCs), characterized by strong winds, heavy rainfall, storm surges, and flooding, have caused significant economic losses and fatalities in coastal regions globally. However, existing TC risk prediction frameworks often fail to adequately account for the direct impacts of flooding. In this study, we propose integrating flood susceptibility, a critical component of flood early warning systems, into TC risk prediction frameworks. Focusing on Guangdong Province, we employ four tree-based machine learning (ML) models (random forest, extreme gradient boosting, light gradient boosting machine, and categorical boosting) to predict county-level direct economic losses (DELs) based on flood susceptibility, oceanographic-meteorological data, and vulnerability data. These ML models are trained and tested on a dataset of 896 samples, achieving high prediction accuracies, with Pearson correlation coefficients exceeding 0.81 between the predicted and observed DEL values. Among the four models, the light gradient boosting machine demonstrates the best performance, achieving the highest values of R and R2, and the lowest values of MSE, MAE, and MedAE. The integration of flood susceptibility is validated by comparing it with traditional methods that directly incorporate environmental factors. Furthermore, the proposed TC risk prediction framework is applied to forecast the impacts of Super Typhoon Mangkhut in 2018, illustrating its potential ability for “real-time” TC risk assessments. These “real-time” DEL predictions not only estimate potential losses but also facilitate timely interventions, thereby enhancing the practical value of the model for disaster prevention and response.

Machine Learning Models for Predicting Dysphonia Following Anterior Cervical Discectomy and Fusion – A Swedish Registry Study

BACKGROUNDDysphonia is one of the more common complications following anterior cervical discectomy and fusion (ACDF). ACDF is the gold standard for treating degenerative cervical spine disorders, and identifying high-risk patients is therefore crucial. PURPOSEThis study aimed to evaluate different machine learning models to predict persistent dysphonia after ACDF. STUDY DESIGNA retrospective review of the nationwide Swedish spine registry (Swespine) PATIENT SAMPLEAll adults in the Swespine registry who underwent elective ACDF between 2006 and 2020. OUTCOME MEASURESThe primary outcome was self-reported dysphonia lasting at least one month after surgery. Predictive performance was assessed using discrimination and calibration metrics. METHODS: Patients with missing dysphonia data at the one-year follow-up were excluded. Data preprocessing involved one-hot encoding categorical variables, scaling continuous variables, and imputing missing values. Four machine learning models (logistic regression, random forest (RF), gradient boosting, K-nearest neighbor) were employed. The models were trained and tested using an 80:20 data split and 5-fold cross-validation, with performance metrics guiding the selection of the best model for predicting persistent dysphonia. RESULTSIn total, 2,708 were included in the study. Twelve key predictors were identified. Four machine learning models were tested, with the RF model achieving the best performance (AUC = 0.794). The most significant predictors across models included preoperative NDI, EQ5Dindex, preoperative neurology, number of operated levels, and use of a fusion cage. The RF model, chosen for its superior performance, showed high sensitivity and consistent accuracy, but a low specificity and positive predictive value. CONCLUSIONSIn this study, machine learning models were employed to identify predictors of persistent dysphonia following ACDF. Among the models tested, the RF classifier demonstrated superior performance, with an AUC value of 0.790. The RF model identified NDI, EQ5Dindex, and number of fused vertebrae as key variables. These findings underscore the potential of machine learning models in identifying patients at increased risk for dysphonia persisting for more than one month after surgery.