Scrutinization of land subsidence rate using a supportive predictive model: Incorporating radar interferometry and ensemble soft-computing

Sub-Saharan Africa continues to shoulder the heaviest burden of malaria. The 2024 World Health Organisation (WHO) malaria report highlighted that Africa contributed an alarming 94% of the global cases and 95% of the deaths. In the WHO African region, progress towards elimination and management of malaria is hindered by weak health systems and a lack of traditional diagnostic methods such as microscopy and malaria Rapid Diagnostic Tests (mRDTs). The primary aim of this paper is to develop a machine learning (ML) ensemble model for malaria diagnosis using clinical and demographic data, tailored for resource-limited settings. A retrospective study was conducted using 637 patient records from Gutu Mission Hospital and Gweru Provincial Hospital in Zimbabwe. Clinical symptoms (fever, chills, abdominal pain, headache and diarrhea) and demographic features (age, gender, residence and travel history) were analysed. Data preprocessing included handling class imbalance using the Synthetic Minority Oversampling Technique (SMOTE) and feature selection using Recursive Feature Elimination (RFE). Seven individual ML models, including Logistic regression (LR), Random Forest (RF), Decision Trees (DT), Gradient Boosting (GB), K-Nearest Neighbour (KNN), Naive Bayes (NB) and XGBoost were trained and evaluated on the malaria dataset. The individual models were further combined to build, train and evaluate ensemble models such as Bagging, Stacking, Soft Voting and AdaBoost. Model performance was assessed using accuracy, precision, confusion matrices, recall, F1 score, and AUC-ROC metrics. Clinical symptoms (chills: p-value=0.001, fever: p=0.003, diarrhea: p=0.01, abdominal pain: p< 0.001) were statistically significant predictors of malaria. Of the demographic factors, only travel history (p=0.02) showed a significant association with malaria. Among the seven individual ML models, GB achieved the highest predictive performance (Accuracy =0.94), followed by RF (Accuracy =0.94) and XGBoost (Accuracy =0.93). The stacking ensemble model outperformed all individual ML models and other ensemble models (bagging, soft voting and AdaBoost), achieving accuracy =0.96, precision =0.95, recall =0.98, F1 score =0.96 and AUC-ROC =0.98. This study demonstrates that ML, particularly ensemble models, can be used to improve malaria diagnosis significantly. The integration of these models into a web-based application could provide a scalable and accessible diagnostic tool for healthcare workers in resource-limited settings.

BackgroundCerebral arteriovenous malformation (AVM) is a cerebrovascular disorder posing a risk for intracranial hemorrhage. However, there are few reliable quantitative indices to predict hemorrhage risk accurately. This study aimed to...

A novel nonlinear ensemble forecasting model incorporating GLAR and ANN for foreign exchange rates

Identification of the sensitive regions to soil erosion can be the foremost step in controlling the vigorous pace of soil degradation in agricultural lands. This study employs individual and ensemble models including compound factor (CF), analytical hierarchy process (AHP), and technique for order preference by similarity to ideal solution-analytical hierarchy process (TOPSIS-AHP) for identifying the regions susceptible to soil erosion in Dnyanganga watershed of India using morphometric attributes (MAs). Thirteen MAs were computed using the newly released TanDEM-X data, which is a bi-static X-band interferometric Synthetic Aperture Radar (SAR) acquisition. Weights for each of the MAs were calculated using AHP, which was employed in both AHP and TOPSIS-AHP models. The outcome of the AHP model displays that watershed slope (Sw), stream frequency (Fu), ruggedness number (Rn), and relative relief (Rr) greatly influences soil erosion in the study area. The priority ranking outcome of individual and ensemble models are compared using the Spearman correlation coefficient test (SCCT) and Kendall tau correlation coefficient test (KTCCT) to select the best model for soil erosion susceptibility mapping. The result of the non-parametric tests reveals that TOPSIS-AHP ensemble model is better in ranking alternatives than the individual models. According to TOPSIS-AHP ensemble model, the sub-watershed DNY16 is highly susceptibility to erosion, whereas DNY1 is least susceptible. Based on the ensemble model, 14.54%, 41.02%, and 44.43% of the area in Dnyanganga watershed fall under high, medium and low soil erosion susceptible classes, respectively. The results from this study will aid decision makers in better conservation planning of soil and water resources.

Breast cancer (BC) is one of the most deadly cancers for the women across the world. Early detection of breast cancer is very important for life saving. Though it is one of the most critical diseases, yet permanent cure has not been developed till now. Artificial intelligence (AI) and machine learning (ML) have been playing a vital role for effective and quick detection of this disease and increasing the rate of survivals. Deep learning (DL) technologies are helping for the analysis of very important features affecting prediction and detection of serious breast cancer diseases. This research paper focuses on solving the problem of BC detection using Wisconsin Diagnosis Breast Cancer (WDBC) data set by applying different ML models after training and validation. Additionally, various types of performance metrics have been calculated and studied. Various ensemble models have also been developed for improved detection of BC. Recurrent neural network (RNN) and gated recurrent unit (GRU)-based DL models are used for feature extraction and training. Different classification models used in this literature are random forest (RF), support vector machine (SVM), k-nearest neighbor (KNN), logistic regression (LR). After classification, the majority voting and stacking ensemble models have been applied for better performances. After exhaustive simulation and analysis, the performance measures in terms of accuracy, precision, recall, F1-score and area under the curve (AUC) values of the ensembled model are 97.3%, 0.97, 0.971, 0.97 and 0.974, respectively. The proposed ensembled model gives higher performances compared to the individual ML models.

Land subsidence is a global problem occurring as a result of various reasons including excessive groundwater extraction. In Iran, over 300 plains are suffering from this phenomenon and the rate of land subsidence in some plains is critical. Tehran city, as the capital of Iran, is located in Tehran-Karaj plain, therefore, the occurrence of land subsidence in this plain can be crucial. Hence, using InSAR method, the paper focused on reconnaissance and evaluation of factors such as; rate of the land subsidence, cause of land subsidence and the time delay. The maximum cumulative value of land subsidence in southwest of Tehran city for a 15-year period of time, from 2004 to 2018, was measured to be 1.94 m which is high and remarkable. The paper further presents the influence of five parameters including; alluvial thickness, natural frequency of earth, land use, geology, and soil granulation on land subsidence. The results of this study showed that land use, geology and soil granulation are also effective parameters on land subsidence, beside groundwater extraction in the study area. Moreover, the time delay between groundwater extraction and land subsidence was determined by comparing the time of sudden decrease in groundwater level and the instant of the land subsidence. In addition, a time delay zonation map was presented for the study area. This map can be helpful in design and construction of geotechnical projects, such as subway tunnels, railways, roads, airports, etc., in the study area.

Crop yield prediction is of great importance for decision making, yet it remains an ongoing scientific challenge. Interactions among different genetic, environmental, and management factors and uncertainty in input values are making crop yield prediction complex. Building upon a previous work in which we coupled crop modeling with machine learning (ML) models to predict maize yields for three US Corn Belt states, here, we expand the concept to the entire US Corn Belt (12 states). More specifically, we built five new ML models and their ensemble models, considering the scenarios with and without crop modeling variables. Additional input values in our models are soil, weather, management, and historical yield data. A unique aspect of our work is the spatial analysis to investigate causes for low or high model prediction errors. Our results indicated that the prediction accuracy increases by coupling crop modeling with machine learning. The ensemble model overperformed the individual ML models, having a relative root mean square error (RRMSE) of about 9% for the test years (2018, 2019, and 2020), which is comparable to previous studies. In addition, analysis of the sources of error revealed that counties and crop reporting districts with low cropland ratios have high RRMSE. Furthermore, we found that soil input data and extreme weather events were responsible for high errors in some regions. The proposed models can be deployed for large-scale prediction at the county level and, contingent upon data availability, can be utilized for field level prediction.

Remote sensing estimation of rice chlorophyll content based on UAV image feature selection and PSO-optimized ensemble learning

Background: Feature selection techniques are important factors for improving machine learning models because they increase prediction accuracy and decrease the time to create a model. Recently, feature selection techniques have been employed on software quality prediction problems with different results and no clear indication of which techniques are frequently used.Objective: This study aims to conduct a systematic review of the application of feature selection techniques in software quality prediction and answers eight research questions.Method: The review evaluates 15 papers in 9 journals and 6 conference proceedings from 2007 to 2017 using the standard systematic literature review method.Results: The results obtained from this study reveal that the filter feature selection method was the most commonly used in the studies (60%) and RELIEF was the most employed among this method, and a limited number of studies employed an ensemble method. Several studies used public datasets available in the PROMISE software project repository (60%). Most studies focused on software defect prediction (classification problem) using area under curve (AUC) as a primary evaluation measure, whereas only two studies focused on software maintainability prediction (regression problem) using mean magnitude of relative error (MMRE) as a primary evaluation measure. All selected studies performed k-fold cross-validation to evaluate model accuracy. Individual prediction models were mostly employed and ensemble models appeared only in three studies. Naive Bayes was the most investigated among individual models, whereas Random forest was the most investigated among ensemble models.Conclusion: Feature selection techniques used by selected primary studies have a positive impact on the performance of the prediction models. Further, both ensemble feature selection method and ensemble models have the ability for increasing prediction accuracy over single methods or individual models and have reported improvement in the prediction accuracy; however, the application of these techniques in software quality prediction is still limited.

Enhancing estimation of water quality index using stacking machine learning techniques: The case of Southern Bug River.

A novel total ensemble (TE) algorithm was developed and compared with random forest optimization (RFO), gradient boosted machines (GBM), partial least squares (PLS), Cubist and Bayesian additive regression tree (BART) algorithms to predict numerous soil health indicators in soils with diverse climate-smart land uses at different soil depths. The study investigated how land-use practices affect several soil health indicators. Good predictions using the ensemble method were obtained for total carbon (R2 = 0.87; RMSE = 0.39; RPIQ = 1.36 and RPD = 1.51), total nitrogen (R2 = 0.82; RMSE = 0.03; RPIQ = 2.00 and RPD = 1.60), and exchangeable bases, m3. Cu, m3. Fe, m3. B, m3. Mn, exchangeable Na, Ca (R2 > 0.70). The performances of algorithms were in order of TE > Cubist > BART > PLS > GBM > RFO. Soil properties differed significantly among land uses and between soil depths. In Kenya, however, soil pH was not significant, except at depths of 45–100 cm, while the Fe levels in Tanzanian grassland were significantly high at all depths. Ugandan agroforestry had a substantially high concentration of ExCa at 0–15 cm. The total ensemble method showed better predictions as compared to other algorithms. Climate-smart land-use practices to preserve soil quality can be adopted for sustainable food production systems.

Land subsidence is a process of land movement based on a certain datum (geodetic reference frame) in which there are various variables causing it. This phenomenon occurs in many big cities in Indonesia, which is the city of Surabaya. The Surabaya city has a type of soil in the form of alluvial soil which is prone to movement. The accumulation of land subsidence that occurs will certainly cause its own losses. As a result, it can damage urban infrastructure so that in the future it can cause disruption to the environment, economic activities, and social life in the region. In addition, the Surabaya area is also a coastal area. Coastal areas have characteristics where this area will be strongly influenced by natural events. One of the natural events that affect the oceans and coastal areas is global temperature changes. The phenomenon of global temperature changes can affect sea level rise. If sea level rise occurs accompanied by an increase in the rate of land subsidence, this can result in flooding or inundation in coastal areas. Considering that the city of Surabaya is a metropolitan city that has a high population density and a continuous level of development, mitigation efforts are important in following up on land subsidence in this area. One of the efforts that can be applied is by conducting GPS observations. The outcomes of the GPS observations were then processed using scientific software in the form of GAMIT/GLOBK. According to the study’s findings indicated that Surabaya is experiencing land subsidence in the northern and eastern coastal areas with a range of 0.2 – 83.3 mm/year. The point that experienced a significant decline was in the border area of Tandes and Asemrowo, namely station BM35. Furthermore, land subsidence conditions influence the extent of tidal inundation, where in 2032 tidal inundation will inundate 105.411 km2.

During recent years, groundwater exploitation and thereby decreasing hydraulic head in the compressible sedimentary aquifer which is placed in the district 19 of Tehran have been caused noticeable land subsidence. The land subsidence has been damaging the infrastructures which have been being built in the south of Tehran Basin, especially in the district 19 of Tehran. A finite-difference groundwater flow model (MODFLOW) and a synthetic aperture radar (SAR) method have been used to estimate and predict the rate of land subsidence in this area, and help hydrogeologists manage the vital groundwater resource correctly. The data have been imported into the model, and the change of the amount of land subsidence and head have been obtained for 39 years. Then the available radar images have been processed. Afterwards, the head calibration and subsidence calibration have been done. The results of the calibrations confirmed the accuracy of the results obtained by the model. Finally, this study suggests that 118 mm of land subsidence and an 11.6 m piezometric head decline are likely to occur from 2014 until 2043.

The ancient city of Xi’an, China, has been suffering severe land subsidence and ground fissure hazards since the 1960s, mainly due to the over-withdrawal of groundwater and large-scale urban construction. This has threatened and will continue to threaten the stability of urban infrastructure, such as the construction and operation of high buildings and subway lines. It is necessary to map the spatiotemporal variations of land subsidence over Xi’an, and to analyze their causes and the correlation with underground water level changes and ground fissure deformation. Time series of land subsidence were observed with the interferometric synthetic aperture radar (InSAR) technique, using multi-sensor SAR datasets from 2012 to 2018. Four land subsidence rate maps over Xi’an city were retrieved from TerraSAR-X, ALOS/PALSAR2, and Sentinel-1 data, each with different tracks. The InSAR derived results were then cross-validated with three independent SAR data stacks, and calibrated with GPS and leveling observations. Next, the spatiotemporal evolutions of three main regional land subsidence zones were quantitatively analyzed in detail, and the surface deformation of the Xi’an subway network was spatially analyzed. Third, the correlations between land subsidence and ground water withdrawal, ground fissure deformation, landforms, and faults were intensively analyzed. Finally, a flat lying sill model with distributed contractions was implemented to model the InSAR deformation over one typical subsidence zone, which further suggested that the ground deformation was mainly caused by groundwater withdrawal. This systematic research can provide sound evidence to serve decision-making for land subsidence mitigation in Xi’an, and may also guide land subsidence research in other cities.

Modern Yellow River Delta has become one of the areas with serious land subsidence disaster in the plain area of Shandong Province. In this paper, Small Baseline Subsets - Interferometry Synthetic Aperture Radar (SBAS-InSAR) technology was used to process the Sentinel-1A data from August 2018 to October 2019 in the modern Yellow River Delta, in order to obtain the range and rate of land subsidence in the study area. At the same time, use the Leveling results of the same period to verify the accuracy. By selecting four larger land subsidence rate as feature points to analyze the characteristics of surface time series deformation, the influencing factors of land subsidence in this area are found out. The results show that during the past 2018-2019 years, the land subsidence in the the Yellow River Delta area was mainly distributed in Xianhe Town, southern the Huanghekou Town, Hubin Town, Guangrao County and its surrounding areas, DaWang Town, Yingli Town, Yangkou Town and the western part of Binhai Township, the most serious subsidence area is located in Xianhe Town, with an annual settlement rate of -204.7 mm/a. The land subsidence in the Yellow River Delta is mainly affected by groundwater overdraft, and the distribution characteristics of land subsidence funnel and deep groundwater depression funnel are basically consistent. In addition, natural factors such as consolidation of Quaternary loose sediments and regional tectonic movement also exacerbate the occurrence of land subsidence.