A Hybrid Intelligence System Based on Relevance Vector Machines and Imperialist Competitive Optimization for Modelling Forest Fire Danger Using GIS

A hybrid artificial intelligence approach using GIS-based neural-fuzzy inference system and particle swarm optimization for forest fire susceptibility modeling at a tropical area

This study aims to develop a comprehensive approach including an analysis of the relationships between flood susceptibility and land-use change, based on the Relevance Vector Machine (RVM) and Coyote Optimization Algorithm (COA) models, applied to Gianh river watershed, Quang Binh province, Central Vietnam. Standard statistical indices, i.e. area under the curve (AUC), were used to assess the model performance. Comparative analyses emphasize that the COA successfully improves the performance of RVM model (AUC=0.99) and is also better than the reference models such as Support Vector Machine (AUC=0.98), Gradient Boosting Machine (AUC=0.97), Random Forest (AUC=0.99), Extra Trees Regressor (AUC=0.98), and Adaboost (AUC=0.96). The improved model, when used in conjunction with land use maps, is able to show that urbanization has increased in flood susceptible areas. The results highlighted that, urbanization has increased in the low and very low flood susceptibility area by 110% between 2005 and 2020, while in the high and very high area by 30 to 40%, despite urban and demographic growth.

Aims/Background Medullary thyroid carcinoma (MTC) is a rare thyroid malignancy with a high mortality rate. Early detection of cervical lymph node metastasis (LNM) is critical for improving prognosis for patients with MTC. This study aimed to investigate the predictive utility of ultrasound-based radiomics for preoperative prediction of cervical LNM in MTC patients. Methods The clinical, ultrasound, and pathological information of 193 patients with MTC were retrospectively examined. Radiomics features were obtained from the ultrasound images using PyRadiomics. The selected patients were randomly divided into training (n = 135) and validation (n = 58) cohorts. In the training dataset, radiomics features were selected using the least absolute shrinkage and selection operator (LASSO) regression, and the univariate and multivariate logistic regression tests were employed to identify the clinical independent predictors of cervical LNM. Three models were created: radiomics, clinical, and combined models, with the latter presented as a nomogram. The area under the curve (AUC) was calculated to evaluate the models' predictive performance. The differences in AUCs between the combined and approach-specific models were compared using the DeLong test. The clinical usefulness of the models was evaluated using decision curve analysis (DCA). Results Nineteen radiomics features were chosen, and the AUCs of the developed radiomics model in the training and validation datasets were 0.881 and 0.859, respectively. Tumour diameter, calcitonin (Ctn) level, tumour margin, and sonographers' suspicion of cervical LNM based on ultrasound findings were clinical independent predictors for cervical LNM. The AUCs of the clinical model built using these predictors were 0.800 and 0.805 in the training and validation datasets, whereas the combined model had much-improved AUCs, measuring 0.925 for the training dataset and 0.918 for the validation test. The DeLong test indicated a significant AUC difference between the combined and clinical models (training dataset p < 0.001, validation dataset p = 0.027), but the difference between the combined and radiomics models was significant only in the training dataset (training dataset p = 0.021, validation dataset p = 0.066). Furthermore, based on the DCA results, the combined model features the largest clinical net benefit. Conclusion The nomogram, the combined model merging the ultrasound-based radiomics with clinical independent predictors, effectively predicts preoperative cervical LNM in MTC patients, outperforming the radiomics and clinical models.

Background Machine learning (ML) algorithm support vector machine (SVM) performed better than Thrombolysis in Myocardial Infarction (TIMI) score for ASIAN STEMI patients. However, Deep Learning (DL) effectiveness in the multiethnic ASIAN population has yet to be determined. DL has automatic learning of the feature from a given dataset without the need to conduct feature selection. However, the selected features by the algorithm is black box. Identifying features associated with mortality is essential to recognize characteristics of patients with high risk for better patient management. Purpose To develop a DL algorithm for in-hospital mortality in multiethnic STEMI patients using predictors identified from the SVM algorithm. To investigate DL performance constructed using predictors from SVM feature extraction and expert-recommended predictors. Methods We constructed four algorithms; a) DL and SVM algorithms with predictors identified from the SVM variable importance b) DL and SVM using predictors based on expert recommendation. We used registry data from the National Cardiovascular Disease Database of 11397 patient's. Fifty parameters including demographics, cardiovascular risk, medications and clinical variables were considered. The Area under the curve (AUC) is the performance evaluation metric. Algorithms were validated against the TIMI and tested using the same validation data. SVM variable importance with backward elimination was used to select and rank important variables. Results DL algorithms outperform SVM and TIMI on the validation dataset; i) DL with SVM selected predictors (15 predictors, AUC = 0.97), ii) DL with expert-recommended predictors (16 predictors, AUC = 0.96), iii) SVM with selected predictors (15 predictors, AUC = 0.92), iv) SVM with expert-recommended predictors (AUC = 0.89) and TIMI (AUC = 0.82). Common predictors across SVM feature selection, expert-recommendation and TIMI are: age, heart rate, Killip class, fasting blood glucose, systolic blood pressure, comorbid diseases and ST-elevation. SVM feature selection also identified diuretics, PCI and pharmacotherapy drugs as predictors that improve mortality prediction in STEMI patients. Our findings suggest that the TIMI score underestimates patients risk of mortality. DL algorithm using selected predictors classified 35% of nonsurvival patients as high risk (risk probabilities &amp;gt;50%) compared to only 12.7% nonsurvival patients by TIMI (score &amp;gt;5) (Figure below). Conclusions In the ASIAN population, patients with STEMI can be better classified using the DL algorithm compared to the ML and TIMI score. Combining ML feature selection with DL allows the identification of distinct factors in a unique ASIAN population for better mortality prediction than relying solely on an expert recommendation as it is a very subjective approach. Continuous validation on population-specific algorithms using DL and ML is needed before implementing in a real clinical setting. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Technology Development Fund 1 TIMI performance on validation setDL performance on validation set

BackgroundMachine learning (ML) algorithm support vector machine (SVM) performed better than Thrombolysis in Myocardial Infarction (TIMI) score for ASIAN STEMI patients. However, Deep Learning (DL) effectiveness in the multiethnic ASIAN population has yet to be determined. DL has automatic learning of the feature from a given dataset without the need to conduct feature selection. However, the selected features by the algorithm is black box. Identifying features associated with mortality is essential to recognize characteristics of patients with high risk for better patient management.PurposeTo develop a DL algorithm for in-hospital mortality in multiethnic STEMI patients using predictors identified from the SVM algorithm. To investigate DL performance constructed using predictors from SVM feature extraction and expert-recommended predictors.MethodsWe constructed four algorithms; a) DL and SVM algorithms with predictors identified from the SVM variable importance b) DL and SVM using predictors based on expert recommendation. We used registry data from the National Cardiovascular Disease Database of 11397 patient's. Fifty parameters including demographics, cardiovascular risk, medications and clinical variables were considered. The Area under the curve (AUC) is the performance evaluation metric. Algorithms were validated against the TIMI and tested using the same validation data. SVM variable importance with backward elimination was used to select and rank important variables.ResultsDL algorithms outperform SVM and TIMI on the validation dataset; i) DL with SVM selected predictors (15 predictors, AUC = 0.97), ii) DL with expert-recommended predictors (16 predictors, AUC = 0.96), iii) SVM with selected predictors (15 predictors, AUC = 0.92), iv) SVM with expert-recommended predictors (AUC = 0.89) and TIMI (AUC = 0.82). Common predictors across SVM feature selection, expert-recommendation and TIMI are: age, heart rate, Killip class, fasting blood glucose, systolic blood pressure, comorbid diseases and ST-elevation. SVM feature selection also identified diuretics, PCI and pharmacotherapy drugs as predictors that improve mortality prediction in STEMI patients. Our findings suggest that the TIMI score underestimates patients risk of mortality. DL algorithm using selected predictors classified 35% of nonsurvival patients as high risk (risk probabilities >50%) compared to only 12.7% nonsurvival patients by TIMI (score >5) (Figure below).ConclusionsIn the ASIAN population, patients with STEMI can be better classified using the DL algorithm compared to the ML and TIMI score. Combining ML feature selection with DL allows the identification of distinct factors in a unique ASIAN population for better mortality prediction than relying solely on an expert recommendation as it is a very subjective approach. Continuous validation on population-specific algorithms using DL and ML is needed before implementing in a real clinical setting.Funding AcknowledgementType of funding sources: Public grant(s) – National budget only. Main funding source(s): Technology Development Fund 1TIMI performance on validation setDL performance on validation set

Detection of metastasis of mediastinal lymph nodes in lung cancer patients with an artificial intelligence model.

This study aims to develop and validate a machine learning-based mortality risk prediction model for V-A ECMO patients to improve the precision of clinical decision-making. This multicenter retrospective cohort study included 280 patients receiving V-A ECMO from the Second Affiliated Hospital of Guangxi Medical University, Yulin First People’s Hospital, and the MIMIC-IV database. The data from the Second Affiliated Hospital of Guangxi Medical University and the MIMIC-IV database were merged and randomly divided in a 7:3 ratio into a training set and an internal validation set, respectively. The dataset from Yulin First People’s Hospital was reserved as an external validation cohort. The primary study outcome was defined as in-hospital mortality.Feature selection was conducted using Lasso regression, followed by the development of six machine learning models: Logistic Regression, Random Forest (RF), Deep Neural Network (DNN), Support Vector Machine (SVM), LightGBM, and CatBoost. Model performance was assessed using the Area Under the Curve (AUC), accuracy, sensitivity, specificity, and F1 score. Model validation was performed through calibration and decision curve analysis. Feature importance was evaluated using SHAP, and subgroup analysis was conducted to assess the model’s applicability across different clinical scenarios. In internal validation, the Logistic Regression model performed the best, with an AUC of 0.86 (95% CI: 0.77–0.93), accuracy of 0.76, sensitivity of 0.73, specificity of 0.79, and an F1 score of 0.73. It outperformed other models (RF: AUC = 0.79, DNN: AUC = 0.78, SVM: AUC = 0.76, LightGBM: AUC = 0.71, CatBoost: AUC = 0.77). External validation yielded consistent results, with the Logistic Regression model’s AUC at 0.75 (95% CI: 0.56–0.92), accuracy of 0.69, sensitivity of 0.64, specificity of 0.73, and an F1 score of 0.66. Calibration curve analysis revealed that the Logistic Regression model had the lowest Brier score (0.1496), indicating the most reliable predicted probabilities. Decision curve analysis demonstrated that the model provided the highest net benefit across most decision thresholds. SHAP analysis identified lactate, age, and albumin as key predictors of mortality, with lactate and age positively correlated, and albumin negatively correlated. Subgroup analysis revealed better performance in the cardiac arrest group (AUC = 0.81), non-sepsis group (AUC = 0.75), and non-diabetes group (AUC = 0.78). The Logistic Regression-based mortality risk prediction model for V-A ECMO patients demonstrated comparable or even favorable performance to more complex machine learning models, with the advantage of higher interpretability.By explicitly incorporating lactate, age, and albumin as the principal predictors, this model facilitates precise risk stratification and provides practical support for clinical decision-making in ECMO management.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-025-25423-4.

Frequent forest fires are causing severe harm to the natural environment, such as decreasing air quality and threatening different species; therefore, developing accurate prediction models for forest fire danger is vital to mitigate these impacts. This research proposes and evaluates a new modeling approach based on TensorFlow deep neural networks (TFDeepNN) and geographic information systems (GIS) for forest fire danger modeling. Herein, TFDeepNN was used to create a forest fire danger model, whereas the adaptive moment estimation (ADAM) optimization algorithm was used to optimize the model, and GIS with Python programming was used to process, classify, and code the input and output. The modeling focused on the tropical forests of the Phu Yen Province (Vietnam), which incorporates 306 historical forest fire locations from 2019 to 2023 and ten forest-fire-driving factors. Random forests (RF), support vector machines (SVM), and logistic regression (LR) were used as a baseline for the model comparison. Different statistical metrics, such as F-score, accuracy, and area under the ROC curve (AUC), were employed to evaluate the models’ predictive performance. According to the results, the TFDeepNN model (with F-score of 0.806, accuracy of 79.3%, and AUC of 0.873) exhibits high predictive performance and surpasses the performance of the three baseline models: RF, SVM, and LR; therefore, TFDeepNN represents a novel tool for spatially predicting forest fire danger. The forest fire danger map from this study can be helpful for policymakers and authorities in Phu Yen Province, aiding sustainable land-use planning and management.

This research aims at proposing a new artificial intelligence approach (namely RVM-ICA) which is based on the Relevance Vector Machine (RVM) and the Imperialist Competitive Algorithm (ICA) optimization for landslide susceptibility modeling. A Geographic Information System (GIS) spatial database was generated from Lang Son city in Lang Son province (Vietnam). This GIS database includes a landslide inventory map and fourteen landslide conditioning factors. The suitability of these factors for landslide susceptibility modeling in the study area was verified by the Information Gain Ratio (IGR) technique. A landslide susceptibility prediction model based on RVM-ICA and the GIS database was established by training and prediction phases. The predictive capability of the new approach was evaluated by calculations of sensitivity, specificity, accuracy, and the area under the Receiver Operating Characteristic curve (AUC). In addition, to assess the applicability of the proposed model, two state-of-the-art soft computing techniques including the support vector machine (SVM) and logistic regression (LR) were used as benchmark methods. The results of this study show that RVM-ICA with AUC = 0.92 achieved a high goodness-of-fit based on both the training and testing datasets. The predictive capability of RVM-ICA outperformed those of SVM with AUC = 0.91 and LR with AUC = 0.87. The experimental results confirm that the newly proposed model is a very promising alternative to assist planners and decision makers in the task of managing landslide prone areas.

Clinical prediction rules could help identify children at risk of slowed growth after an episode of diarrheal illness, and these rules may be generalizable to all children, regardless of diarrhea status.

Introduction: Machine learning models hold potential for improved cardiometabolic disease prediction and more efficacious prevention and intervention efforts. Current models are limited by use of cross-sectional datasets and variables (features) consisting of diagnostic criteria (such as blood glucose or hemoglobin A1c in type 2 diabetes models) and overlook health behaviors and social determinants of health (HB-SDH). Hypothesis: Machine learning models built with HB-SDH variables will accurately predict type 2 diabetes. Methods: We used data from 2,493 participants without diabetes in the Adolescent to Adult Health study (mean age = 15.5 years at baseline (1994), 40.9% male) who contributed blood samples and anthropometric measures in wave 4 (2008) or 5 (2016-2018). We used an 80/20 split to form separate training and test datasets; the training dataset was used to build random forest and XGBoost models, and performance with respect to classifying diabetes was evaluated in the test dataset. We first built a model using 15 demographic and biomarker features as a comparator for our second model, which included 34 HB-SDH features measured during adulthood but no biomarkers. Results: Five-hundred sixteen participants developed type 2 diabetes. In XGBoost, the comparator model was found to predict diabetes with 92.6% accuracy (95% CI: 89.8%, 94.8%) and area under the curve (AUC) = 0.8165. The HB-SDH model had 88.7% accuracy (95% CI: 85.4, 91.4%) and AUC = 0.7631. Using random forest, the comparator model had 93.0% accuracy (95% CI: 90.3, 95.2%) and AUC = 0.8376, while the HB-SDH model had 89.8% (95% CI: 88.6, 92.4%) accuracy and AUC = 0.7650. The top 10 variables in ascertaining classifications in the random forest model included self-reported general health, waist circumference, self-reported anxiety diagnosis, self-reported hypertension diagnosis, income, BMI, education level, change in weight, financial scarcity, and experienced stigma, while in the XGBoost model these were self-reported general health, experienced stigma, financial scarcity, self-reported anxiety diagnosis, experienced weight stigma, waist circumference, income, self-reported hypertension diagnosis, sugar-sweetened beverage intake, and BMI. Conclusions: Machine learning models built with behavioral and social determinants of health variables perform comparably to models with biomarkers and identify salient features for diabetes risk. The features that factored into classification performance were similar between random forest and XGBoost models. In this population, HB-SDH of relevance included socioeconomic indicators, sugary beverages, anxiety, and experiencing stigma, in addition to adiposity and hypertension. This finding needs to be replicated in diverse cohorts that may experience different HB-SDH.

Comparing accuracy of machine learning approaches to identifying parathyroid adenomas: Lessons and new directions

Novel Radiomics-Clinical Model for the Noninvasive Prediction of New Fractures After Vertebral Augmentation

Bus headway regularity heavily affects transit riders’ attitude for choosing public transportation and also serves as an important indicator for transit performance evaluation. Therefore, an accurate estimate of bus headway can benefit both transit riders and transit operators. This paper proposed a relevance vector machine (RVM) algorithm to predict bus headway by incorporating the time series of bus headways, travel time, and passenger demand at previous stops. Different from traditional computational intelligence approaches, RVM can output the probabilistic prediction result, in which the upper and lower bounds of a predicted headway within a certain probability are yielded. An empirical experiment with two bus routes in Beijing, China, is utilized to confirm the high precision and strong robustness of the proposed model. Five algorithms [support vector machine (SVM), genetic algorithm SVM, Kalman filter, k-nearest neighbor, and artificial neural network] are used for comparison with the RVM model and the result indicates that RVM outperforms these algorithms in terms of accuracy and confidence intervals. When the confidence level is set to 95%, more than 95% of actual bus headways fall within the prediction bands. With the probabilistic bus headway prediction information, transit riders can better schedule their trips to avoid late and early arrivals at bus stops, while transit operators can adopt the targeted correction actions to maintain regular headway for bus bunching prevention.

Gully erosion susceptibility mapping (GESM) is a valuable tool for sustainable land use management and reducing soil erosion. Gully erosion and its formation are a natural process; it greatly threatens agriculture, environment, ecosystem disruption, and natural resources. The objective of this present study is to develop a GESM by implementation of well acceptable SVM learning algorithm in Golestan Province, Kalaleh Township, Iran. Primarily, gully sites were obtained by comprehensive field observations. After that, 12 gully erosion predisposing factors were selected to assess the gully erosion susceptibility map. The 12 conditioning factors were aspect, altitude, drainage density, lithology, slope angle, slope length, distance from river, profile curvature, drainage density, TWI, distance from road, and plan curvature. Finally, gully erosion susceptibility map was prepared using the SVM model in “R” environment. In the final stage, assessment of the prediction accuracy of the susceptibility model with the help of training (70%) and validation datasets (30%) of gully location was done. The predicted susceptibility map was validated with the help of receiver operating characteristic (ROC) curve, true skill statistics (TSS), and deviance value. The results indicated that the areas under the curve (AUC) were calculated as 94.3% and 97.0% based on validation and training dataset, respectively. Furthermore, the TSS, deviance, and correlation values were 0.84, 0.50, and 0.85, respectively. So, the results of other indices including, sensitivity, specificity, and Cohen’s Kappa (CK) showed that SVM model has reasonable prediction accuracy for the cases of gully erosion susceptibility assessment. As regards the SVM model, a total area of 11.66% was identified as the hazard prone area of the mentioned Town ship. So, it is concluded that the gully erosion map serves as an important tool for protective action and watershed management, specifically at the initiation of the gully to protect the development of land degradation.