Nonlinear Artificial Neural Network Model Research Articles

AI-based non-linear models for mechanical and toughness properties of sustainable fiber-reinforced geopolymer concrete (FRGPC)

The incorporation of fibers into geopolymer concrete (GPC) produces FRGPC which mitigates brittle failure and restrains macro crack propagation; however, research on predicting the mechanical and toughness properties of FRGPC remains limited. This study addresses this gap by developing prediction models for compressive strength (CS), flexural strength (FS), flexural toughness (FT), and fracture toughness (FR). A dataset of 600 data points was compiled from the published literature, encompassing various constituent proportions, fiber shapes, fiber dosages, aspect ratios, and curing conditions. Employing an artificial neural network (ANN) methodology, 10 independent variables (g1, g2, …., g10) were utilized to predict four dependent variables (CS, FS, FT, and FR), resulting in the development of eight non-linear ANN models for both straight fibers (SFs) and hooked fibers (HFs). Each model has shown a higher R 2 value and lower root mean square error (RMSE) for training (70%), testing (15%), and validation (15%) datasets. The prediction model for the CS of FRGPC with HFs (CS-HF) and SF (CS-SF) showed R 2 values of 0.983 and 0.973, and RMSE of 2.088 and 2.435 MPa highlighting the accuracy of the ANN models. Similarly, the comparative analysis for FR models exhibited R 2 values of 0.997 and 0.987, and RMSE values of 0.045 Mpa √m and 0.032 Mpa √m for FR-HF and FR-SF, highlighting that fiber addition strongly impacts improving toughness properties. To identify the most influential independent variable(s), sensitivity analysis revealed g10, g1, g8, and g9 as the most influential parameters for mechanical and toughness properties with SFs. For HFs, g1, g2, and g3 were most influential for mechanical properties, and g8, g9, and g10 for toughness. This study also presented the mathematical formulation of the developed models for better interpretability and to facilitate optimal and economical FRGPC mix design selection, highlighting the potential of AI-based models in advancing sustainable construction materials.

On Hybrid Autoregressive Integrated Moving Average Artificial Neural Network Time Series Analysis of the Nigeria External Reserves

The amount of the external reserves possessed by a country is an essential component in the measure of its economic status. In essence, a country with substantial amount of money in its external reserves would experience a buoyant and friendly economy among other nations. This study employs three different time series models to examine the status of the Nigeria External reserves for 22 years (1960-2022). These include the Auto Regressive Integrated Moving Average (ARIMA), Artificial Neural Networks (ANN) and the hybrid ARIMA-ANN models. ARIMA (1,0,1)(0,0,2)12 emerged as the optimum model among other fitted linear models, having the smallest Akaike Information Criterion (AIC) value of 14365.84 and was used to fit the nonlinear Hybrid ARIMA-ANN model and for the prediction of the External Reserves series. The Mean Absolute Error (MAE), the Root Mean Square Error (RMSE) and the Mean Absolute Percentage Error (MAPE) values of (590.1479), (358.3421), and (24.10321) for the Hybrid ARIMA-ANN model were also the least in comparison with their corresponding values for the other models. These show that the Hybrid model has the least error value compared with those of the independent ARIMA and ANN models. Hence, the nonlinear Hybrid ARIMA-ANN model performed excellently in the estimation and the generation of the forecast values than the conventional linear ARIMA and the nonlinear ANN models.

Comparison between Linear and Non-linear ANN Models for Predicting Water Quality Parameters at Tigris River

In this research, Artificial Neural Networks (ANNs) technique was applied in an attempt to predict the water levels and some of the water quality parameters at Tigris River in Wasit Government for five different sites. These predictions are useful in the planning, management, evaluation of the water resources in the area. Spatial data along a river system or area at different locations in a catchment area usually have missing measurements, hence an accurate prediction. model to fill these missing values is essential.The selected sites for water quality data prediction were Sewera, Numania , Kut u/s, Kut d/s, Garaf observation sites. In these five sites models were built for prediction of the water level and water quality parameters. the following (Biological Oxygen Demand( ), Phosphate,( ) Sulfate(), Nitrate( ), Calcium(Ca), Magnesium(Mg), Total Hardness(TH), Potassium(K), Sodium (Na), Chloride (CL), Total Dissolved Solids (TDS), Electric conductivity (EC), Alkalinity(ALK)). The ANN models tried herein were the Multisite- Multivariate ANN models (5-sites, 14 variables), five models were built, one for each of the five stations as the missing data station. The linear ANN (traditional) models fail to make the prediction of all variables with high correlation coefficient simultaneously. Hence a non- linear input ANN model was developed herein and believed to be a new modification in ANN modeling. It was found that the ANNs have the ability to predict water level and water quality parameters at all the sites with a good degree of accuracy, the range of correlation coefficients obtained are (12.9%-97.2%) for linear models, while for this model with Non-linear terms, The range of correlation coefficients obtained is (71.8%-99.6%).

Simple and Rapid Discrimination of Methicillin-Resistant Staphylococcus aureus Based on Gram Staining and Machine Vision.

Methicillin-resistant Staphylococcus aureus (MRSA) is a clinical threat with high morbidity and mortality. Here, we describe a new simple, rapid identification method for MRSA using oxacillin sodium salt, a cell wall synthesis inhibitor, combined with Gram staining and machine vision (MV) analysis. Gram staining classifies bacteria as positive (purple) or negative (pink) according to the cell wall structure and chemical composition. In the presence of oxacillin, the integrity of the cell wall for methicillin-susceptible S. aureus (MSSA) was destroyed immediately and appeared Gram negative. In contrast, MRSA was relatively stable and appeared Gram positive. This color change can be detected by MV. The feasibility of this method was demonstrated in 150 images of the staining results for 50 clinical S. aureus strains. Based on effective feature extraction and machine learning, the accuracies of the linear linear discriminant analysis (LDA) model and nonlinear artificial neural network (ANN) model for MRSA identification were 96.7% and 97.3%, respectively. Combined with MV analysis, this simple strategy improved the detection efficiency and significantly shortened the time needed to detect antibiotic resistance. The whole process can be completed within 1 h. Unlike the traditional antibiotic susceptibility test, overnight incubation is avoided. This new strategy could be used for other bacteria and represents a new rapid method for detection of clinical antibiotic resistance. IMPORTANCE Oxacillin sodium salt destroys the integrity of the cell wall of MSSA immediately, appearing Gram negative, whereas MRSA is relatively stable and still appears Gram positive. This color change can be detected by microscopic examination and MV analysis. This new strategy has significantly reduced the time to detect resistance. The results show that using oxacillin sodium salt combined with Gram staining and MV analysis is a new, simple and rapid method for identification of MRSA.

Atmospheric Pattern–Based Predictions of S2S Sea Level Anomalies for Two Selected U.S. Locations

Abstract With climate change causing rising sea levels around the globe, multiple recent efforts in the United States have focused on the prediction of various meteorological factors that can lead to periods of anomalously high tides despite seemingly benign atmospheric conditions. As part of these efforts, this research explores monthly scale relationships between sea level variability and atmospheric circulation patterns and demonstrates two options for subseasonal to seasonal (S2S) predictions of anomalous sea levels using these patterns as inputs to artificial neural network (ANN) models. Results on the monthly scale are similar to previous research on the daily scale, with above-average sea levels and an increased risk of high-water events on days with anomalously low atmospheric pressure patterns and wind patterns leading to onshore or downwelling-producing wind stress. Some wind patterns show risks of high-water events to be over 6 times higher than baseline risk and exhibit an average water level anomaly of +94 mm above normal. In terms of forecasting, nonlinear autoregressive ANN models with exogenous input (NARX models) and pattern-based lagged ANN (PLANN) models show skill over postprocessed numerical forecast model output, and simple climatology. Damped-persistence forecasts and PLANN models show nearly the same skill in terms of predicting anomalous sea levels out to 9 months of lead time, with a slight edge to PLANN models, especially with regard to error statistics. This perspective on forecasting—using predefined circulation patterns along with ANN models—should aid in the real-time prediction of coastal flooding events, among other applications.

Machine learning models combined with wavelet transform and phase space reconstruction for groundwater level forecasting

Reliable forecasting of groundwater levels plays an important role in water resource management and the prevention of environmental problems. This study aims to introduce a novel hybrid model combining machine learning with wavelet transform (WT) and phase space reconstruction (PSR) for groundwater level forecasting. The development of the proposed hybrid model consists of two levels. Firstly, a popular data-driven nonlinear artificial neural network (ANN) model was selected as the base model. Thereafter, the WT is used to decompose the groundwater level time series. In addition, the PSR method is further adopted to select the suitable sub-series components resulting from the WT decomposition processes. At the first level, the hybrid model only combines the WT with ANN (WT-ANN). At the second level, a two-step preprocessing process, WT coupled with PSR, is implemented before the ANN, namely WT-PSR-ANN. Meanwhile, three mother wavelets (Daubechies, Symlets, and Dmeyer) are selected to further investigate the effect of wavelet type. Finally, the model's performance is evaluated using visual analysis and statistical metrics (root means square error, correlation coefficient, and the Nash-Sutcliffe efficiency coefficient). The results show that the WT-ANN model outperformed the vanilla ANN model. Furthermore, the WT-PSR-ANN model achieved the best accuracy for all selected mother wavelets. The most suitable mother wavelet is specific to each well, and there is no one-size-fits-all answer. This study highlights the effectiveness of PSR combined with WT as a data preprocessing step when the ANN model is applied for groundwater level forecasting.

Prediction of displacement in Reinforced concrete based on artificial neural networks using sensors

Although structural health monitoring (SHM) is widely used in civil engineering, researchers are working to enhance its accuracy. In previous laboratory experiments, four types of sensors, namely force resisting sensor (FSR), piezoelectric sensor (PZS), microelectromechanical system (MEMS) accelerometer sensor, and flex sensor (FLS), were employed to predict load-displacement in Reinforced Concrete (RC) beams subject to monotonic two-point loading. Static loading was performed in the laboratory. The results demonstrated that MEMS sensors outperformed the other three sensors. Consequently, only MEMS sensors were used in the ANN model to predict the error percentage. This study compared the static results of previous experiments to those obtained from an Artificial Neural Network (ANN) model that used the beam's position and load as input parameters to predict displacement. The results revealed that the non-linear ANN model produced positive outcomes. Therefore, SHM is likely to utilize MEMS sensors in various civil engineering applications.

A machine learning model for prediction of 30-day primary graft failure after heart transplantation

BackgroundPrimary graft failure (PGF) remains the most common cause of short-term mortality after heart transplantation. The main objective was to develop and validate a risk model for prediction of short-term mortality due to PGF after heart transplantation using the ISHLT Heart Transplant Registry. MethodsWe developed a non-linear artificial neural networks (ANN) model to evaluate the association between recipient-donor variables and post-transplant PGF. Patients in the ISHLT registry were randomly divided into derivation and an independent internal validation cohort. The primary endpoint was PGF defined as death within 30 days due to Graft failure or Cardiovascular causes or retransplant within 30 days for causes other than rejection. ResultsAmong 64,964 adult recipients transplanted between 1994 and 2013, mean age was 51 years and 22% were female. The incidence of PGF up to 30 days was 3.7%. The ANN model selected 33 of 77 risk variables as relevant for PGF prediction. The C-index in the test cohort was 0.70 (95% CI: 0.68-0.71). The risk variables which most influenced the PGF were underlying HF diagnosis, ischemia time and sex, while renal function had a lower influence. ConclusionAn ANN model to predict primary graft dysfunction was derived and independently validated. The good discrimination of the ANN model likely results from its flexibility to model potentially non-linear relationships and interactions. Whether this model with improved discrimination can assist in clinical decisions at the time of transplant should be tested.

Predicting Fecal-Indicator Organisms in Coastal Waters Using a Complex Nonlinear Artificial Intelligence Model

High levels of fecal-indicator organisms (FIOs) at bathing water sites can cause disease and impose threats to public health. There is a need for predicting FIO levels to inform the public and reduce exposure. Data-driven models are one of the main tools being considered as predictive models. However, identifying the main inputs of the data-driven models is a major challenge in developing FIO predictor models. This paper develops a data-driven model for FIO concentration prediction based on a limited number of critical input variables. Essential variables were identified with be a combination of the gamma test and Genetic Algorithm (Gamma-GA test). Artificial neural networks (ANNs) and linear regression models were developed using these two variable identification approaches for comparison. The models were applied to a case study, and it was found that the model using the Gamma-GA test has a high potential to predict FIO levels more accurately, although this requires further investigation with different case studies. A correlation analysis was required prior to the variable identification approaches in this study. The need of this analysis highlights the significance of understanding the waterbody and the data set in the development and application of data-driven models. Models using a Gamma-GA test were more capable of predicting extreme (high) FIO concentrations, making a Gamma-GA test more suitable for a bathing water quality early warning system. The importance of nonlinearity in such predictive models was also demonstrated by the better performance of nonlinear ANN models compared with linear regression models regardless of the variable identification approaches used. This paper highlights the importance of nonlinearity in bathing water quality prediction and encourages further utilization of nonlinear models for this application.

Near-Infrared Spectroscopy Coupled with Chemometrics and Artificial Neural Network Modeling for Prediction of Emulsion Droplet Diameters.

There is increased interest in the food industry for emulsions as delivery systems to preserve the stability of sensitive biocompounds with the aim of improving their bioavailability, solubility, and stability; maintaining their texture; and controlling their release. Emulsification in continuously operated microscale devices enables the production of emulsions of controllable droplet sizes and reduces the amount of emulsifier and time consumption, while NIR, as a nondestructive, noninvasive, fast, and efficient technique, represents an interesting aspect for emulsion investigation. The aim of this work was to predict the average Feret droplet diameter of oil-in-water and oil-in-aqueous mint extract emulsions prepared in a continuously operated microfluidic device with different emulsifiers (PEG 1500, PEG 6000, and PEG 20,000) based on the combination of near-infrared (NIR) spectra with chemometrics (principal component analysis (PCA) and partial least-squares (PLS) regression) and artificial neural network (ANN) modeling. PCA score plots for average preprocessed NIR spectra show the specific grouping of the samples into three groups according to the emulsifier used, while the PCA analysis of the emulsion samples with different emulsifiers showed the specific grouping of the samples based on the amount of emulsifier used. The developed PLS models had higher R2 values for oil-in-water emulsions, ranging from 0.6863 to 0.9692 for calibration, 0.5617 to 0.8740 for validation, and 0.4618 to 0.8692 for prediction, than oil-in-aqueous mint extract emulsions, with R2 values that were in range of 0.8109-0.8934 for calibration, 0.5017-0.6620, for validation and 0.5587-0.7234 for prediction. Better results were obtained for the developed nonlinear ANN models, which showed R2 values in the range of 0.9428-0.9917 for training, 0.8515-0.9294 for testing, and 0.7377-0.8533 for the validation of oil-in-water emulsions, while for oil-in-aqueous mint extract emulsions R2 values were higher, in the range of 0.9516-0.9996 for training, 0.9311-0.9994 for testing, and 0.8113-0.9995 for validation.

Establishment of a Combined Model for Ozone Concentration Simulation with Stepwise Regression Analysis and Artificial Neural Network

With the development of industrialization and the increase in the number of motor vehicles in megacities in China, ozone pollution has become a prominent problem. Although different models have been used on ozone concentration simulation, the accuracy of different models still varies. In this study, the performance of two models including a linear stepwise regression (SR) model and a non-linear artificial neural network (ANN) model on the simulation of ozone concentration were analyzed in the Jing-Jin-Ji region, which is one of the most polluted areas in China. Results showed that the performance of the ANN model (adjusted R2 = 0.8299, RMSE = 22.87, MAE = 16.92) was better than the SR model (adjusted R2 = 0.7324, RMSE = 28.61, MAE = 22.30). The performance of the ANN on simulating an ozone pollution event was better than the SR model since a higher probability of detection (POD) and threat score (TS) values were obtained by the ANN model. The model performance for spring, autumn and winter was generally higher than that for summer, which may because the weights of factors on simulating high and low ozone concentrations were different. The method proposed by this study can be used in ozone concentration estimation.

Modelling Complex Relationships Between Sustainable Competitiveness and Digitalization

Digitalization is the core component of future development in the 4.0 industrial era. It represents a powerful mechanism for enhancing the sustainable competitiveness of economies worldwide. Diverse triggering effects shape future digitalization trends. Thus, the main research goal in this study is to use sustainable competitiveness pillars (such as social, economic, environmental and energy) to evaluate international digitalization development. The proposed empirical model generates comprehensive knowledge of the sustainable competitiveness-digitalization nexus. For that purpose, a nonlinear regression has been applied on gathered annual data that consist of 33 European countries, ranging from 2010 to 2016. The dataset has been deployed using Bernoulli’s binominal distribution to derive training and testing samples and the entire analysis has been adjusted in that context. The empirical findings of artificial neural networks (ANN) suggest strong effects of the economic and energy use indicators on the digitalization progress. Nonlinear regression and ANN model summary report valuable results with a high degree of coefficient of determination (R2>0.9 for all models). Research findings state that the digitalization process is multidimensional and cannot be evaluated as an isolated phenomenon without incorporating other relevant factors that emerge in the environment. Indicators report the consumption of electrical energy in industry and households and GDP per capita to achieve the strongest effect.

Hybrid Artificial Neural Networks for Modeling Shallow-Water Bathymetry via Satellite Imagery

Accurate bathymetric mapping for shallow-water areas is essential for coastal and maritime engineering applications. However, traditional multibeam or light detection and ranging (LiDAR) survey techniques used to produce high-quality bathymetric maps are expensive. Satellite-derived bathymetry provides a fast and inexpensive method for the large-scale mapping of shallow-water areas and can overcome the complexities of traditional bathymetric mapping methods in these areas. Traditionally, linear regression models, most commonly the Stumpf model, are used for satellite-based bathymetric modeling. However, nonlinear artificial neural network (ANN) models have been recently developed and implemented for satellite-based bathymetric modeling and are under significant investigation to develop the most accurate and optimal model. This article proposes two new hybrid ANN-based models for bathymetric modeling and investigates their performance using satellite imagery data and “truth” depth data for a coastal shallow-water study area. Two-hybrid ANN algorithms are developed, namely, particle swarm optimization (PSO)-ANN and optimally pruned extreme learning machine (OPELM), and their results are compared with the traditional Stumpf method and current state-of-the-art ANN model. The study area dataset comprises the “truth” depth data from a nautical chart of the Alqumriyah Island study area in Saudi Arabia and the corresponding spectral reflection values of green, blue, and near-infrared bands from the free-of-charge Level-1C product of Sentinel-2A images used to train and validate the two newly developed models and the traditional models. The results show that the developed OPELM method can accurately derive the bathymetry and is superior to the developed PSO-ANN model, the current state-of-the-art ANN model, and the traditional Stumpf model by 12.10%, 18.76%, and 32.46%, respectively. The OPELM model can also be used for bathymetric modeling of shallow-water areas with depths up to 30 m with a high level of accuracy compared with the current state-of-the-art ANN and traditional methods. The significant contribution of this research is that it is the first investigation of the artificial intelligence-based hybrid OPELM method for accurate bathymetric modeling and will certainly encourage further investigations of hybrid models. Moreover, this research explores whether these developed hybrid models can meet the International Hydrographic Organization standards for hydrographic survey applications.

Forecasting stock returns on the Amman Stock Exchange: Do neural networks outperform linear regressions?

Different models have been used in the finance literature to predict the stock market returns. However, it remains an open question whether non-linear models can outperform linear models while providing accurate predictions for future returns. This study examines the prediction of the non-linear artificial neural network (ANN) models against the baseline linear regression models. This study aims specifically to compare the prediction performance of regression models with different specifications and static and dynamic ANN models. Thus, the analysis was conducted on a growing market, namely the Amman Stock Exchange. The results show that the trading volume and interest rates on loans tend to explain the monthly returns the most, compared to other predictors in the regressions. Moreover, incorporating more variables is not found to help in explaining the fluctuations in the stock market returns. More importantly, using the root mean square error (RMSE), as well as the mean absolute error statistical measures, the static ANN becomes the most preferred model for forecasting. The associated forecasting errors from these metrics become equal to 0.0021 and 0.0005, respectively. Lastly, the analysis conducted with the dynamic ANN model produced the highest RMSE value of 0.0067 since November 2018 following the amendment to the Jordanian income tax law. The same observation is also seen since the emerging of the COVID-19 outbreak (RMSE = 0.0042).

Resistance of blended alkali-activated fly ash-OPC mortar to mild-concentration sulfuric and acetic acid attack.

The traditional cementitious product is prone to suffer from a high degree of deterioration in the case of exposure to acid solutions because of the decomposition of the binder network. However, the degradation of concrete structures in service by mild concentrations of acid under conditions involving sewage, industrial waters, and acid rain is more common and results in a significant environmental problem. The utilization of alkali-activated materials has been seen to potentially offer an attractive option with regard to acceptable durability and a low carbon footprint. With the aid of visual observation, mass loss, compressive strength tests, X-ray diffraction, Fourier transform infrared spectroscopy, and field-emission scanning electron microscopy with energy-dispersive X-ray spectroscopy, the acid resistance of alkali-activated fly ash mortars in which the precursor was partially replaced (0-30% by mass proportion) with ordinary Portland cement (OPC) was evaluated after 180 days of exposure to mild-concentration sulfuric and acetic solutions (pH = 3). A conventional cement mortar (100% OPC) was used as a reference group. The results demonstrate that the addition of OPC into the alkali-activated system causes a significant increase in compressive strength (around 16.08-36.61%) while showing an opposite influence on durability after acid attack. Based on a linear mean value and nonlinear artificial neural network model simulation, the mass losses of the specimens were evaluated, and the alkali-activated pure-fly ash mortar demonstrated the lowest value (i.e., a maximum of 5.61%) together with the best behavior in the aspect of discreteness at 180 days. The results from microstructure analysis show that the coexistence of the N-A-S-H and C-S-H networks in the blend system occurred by both OPC hydration and FA. However, the formation of the gypsum deposition within the fly ash-OPC blend systems at sulfuric acid was found to impose internal disintegrating stresses, causing a significant area of delamination and cracks. In addition, alkali metal ion leaching, dealumination, as well as the disappearance of some crystalline phases occurred in specimens immersed in both types of acids.

PM2.5 concentration prediction during COVID-19 lockdown over Kolkata metropolitan city, India using MLR and ANN models

Kolkata is the third densely populated city of India and Kolkata stands in the World's 25 most polluted cities along with 10 worse polluted cities in India. The relevant study claims that due to the imposition of lockdown during COVID-19 pandemic, the atmospheric pollution level has been significantly reduced over the metropolitan city Kolkata like other cities of the world. The main objective of this study is to predict the concentration of PM2.5 using multiple linear regression (MLR) and artificial neural network (ANN) models and similarly, to compare the accuracy level of two models. The concentration of PM2.5 data has been obtained from state pollution control board, Govt. of West Bengal and daily meteorological data have been collected from the world weather website. The results show that non-linear artificial neural network model is more rational compared with multiple linear regression model due to its high precision and accuracy level (in respect to RMSE, MAE and R2). In this research artificial neural network (ANN) model exhibited higher accuracy during the training and testing phases (root mean square error (RMSE), mean absolute error (MAE) and R2 indicate 3.74, 1.14 and 0.91 respectively in training phase and 2.55, 4.32 and 0.69 in testing phase respectively). This model (ANN)) can be applied to predict the concentration of PM2.5 during the execution of urban air quality management plan.

Non-valvular atrial fibrillation recurrence after sinus rhythm restoring at different follow-up periods: phenotype-genotype high-risk groups, considering rs10465885 polymorphism in connexin-40 gene

Abstract Funding Acknowledgements Type of funding sources: None. Purpose to provide risk stratification of non-valvular atrial fibrillation (AF) recurrence after sinus rhythm restoring (SRR) in patients (pts) with AF at different follow-up periods (3 months (AFr3m), 12 months (AFr12m) and 18 months [AFr18m]), based on phenotype-genotype high-risk groups, considering rs10465885 polymorphism in connexin-40 gene (SNP-Cx40). Methods. We enrolled 186 pts (mean age (55 ± 10) years; males 123 [66,1%]) with AF (paroxysmal – 86, persistent – 72, stable – 28 pts; first onset (FO) AF – 48 pts). Clinical, laboratory and echocardiographic data were analyzed. SNP-Cx40 was genotyped by real time PCR (T – reference, C – minor allele) in 112 pts. The genotypes were distributed as follows: TT – 25,9% (n = 29); CT – 49,1% (n = 55); CC – 25,0% (n = 28). SRR was performed in 112 cases (102 pts) with non-permanent AF: 30 – pharmacological cardioversion (PCV), 62 – direct-current cardioversion (DCV), 20 cases – radiofrequency ablation (RFA). AFr3m occurred in 53 (43,4%) of 122 available cases; AFr12m – 65,5% (76/116); AFr18m – 75,2% (79/105). The Artificial Neural networks (ANNs) analysis was performed to select the AF recurrence predictors. We considered the ANN activation function value (Y) and its relation to Y cut-off value (Ycrit). In case of Y > Ycrit, the AF recurrence risk was considered as «high». Results. We built three nonlinear ANN models (multilayer perceptrons) for AFr3m (Ycrit = 0,496), AFr12m (Ycrit = 0,503) and AFr18m (Ycrit = 0,720) risk prediction. In case of SNP-Cx40 CC genotype carriage, we determined the additional increase of AFr3m risk after PCV in pts with CHA2DS2-VASc score «0» and normal (Y = 0,629) or mildly decreased (Y = 0,616) left ventricular mid-wall fractional shortening, and in the case of its moderate decrease – both after PCV (Y = 0,585) and DCV (Y = 0,627). The CC genotype was associated with AFr12m high risk in pts without heart failure (HF) and mildly increased left atrial dimension (LAD) – both after PCV (FO AF with known precise event duration (PED); Y = 0,906) or DCV (FO AF with unknown PED; Y = 0,911). Additionally, CC genotype was associated with AFr12m high risk after RFA in pts without HF and normal or mildly increased LAD (Y = 0,912), and in pts with HF B or C1 stage (according to modified AHA/ACC classification) with moderately increased LAD (Y = 0,912). The high-risk groups of AFr18m in pts with CC genotype were as follows (Y = 0,913): after RFA in pts with recurrent AF and presence of episodes lasting ≥7 days; after PCA in case of FO AF with unknown PED and index episode lasting ≥1 month; after DCV in case of FO AF with unknown PED and index episode lasting ≥12 months. Conclusion. AFr3m, AFr12m and AFr18m, besides SNP-Cx40, were non-linearly associated with SRR type, and certain clinical and echocardiographic phenotypic parameters, which could be used for AF recurrence risk stratification, with the selection of phenotype-genotype high-risk groups, considering SNP-Cx40.

Analysis of Critical Velocity of Cold Spray Based on Machine Learning Method with Feature Selection

Cold spraying has a potential application prospect in the field of repairing and additive manufacturing. The critical velocity of the cold spray is a key factor that determines the adhesion of particles during the cold spraying process, and it only depends on the particle parameters under the same working conditions. In the present study, the relationship between particle parameters and critical velocity is investigated using a feature selection method to obtain the influence weight of different particle parameters. Based on the results of feature selection, linear and nonlinear artificial neural networks are established to predict the critical velocity, respectively. The results of the feature selection show that the mechanical parameters of the material have a higher influence weight on the critical velocity than thermal parameters. In the prediction model, the ANN (artificial neural network) method shows a good prediction, and the nonlinear ANN model achieves better generalization ability than the linear ANN model and empirical formula with 95.24% prediction accuracy on the original data set and 96.45% prediction accuracy on the new data set.

Radial basis function-based node localization for unmanned aerial vehicle-assisted 5G wireless sensor networks

Localization or positioning of wireless sensor nodes is an essential task for a wide range of applications in wireless sensor networks-based fifth generation (5G) networks. Node localization using mobile aerial beacon nodes (MABNs) provides high localization accuracy and less deployment cost compared to the localization using fixed ground beacon nodes because MABN deployed in unmanned aerial vehicles sends signals to unknown nodes (UNs) through reliable air to ground (AG) channel link. The classical received signal strength indicator (RSSI)-based multilateration and optimization-based least square localization (OLSL) schemes result high localization error because of the nonlinear distortions induced in the wireless channel. So, the highly nonlinear artificial neural network (ANN) models such as multilayer perceptron (MLP) and radial basis function (RBF) are used effectively for nonlinear node localization problems. ANN-based localization techniques are also capable of localizing mobile UNs. Further, the RBF with Gaussian activation has a little edge over the MLP with sigmoid activation because the wireless channel usually modelled with a Gaussian random variable. The simulation results included in this article also justify the efficacy of the proposed RBF localization with performance gain of about (7–70) % over MLP, OLSL and RSSI localization techniques.

Artificial neural network (ANN), M5P-tree, and regression analyses to predict the early age compression strength of concrete modified with DBC-21 and VK-98 polymers

In this study, the effect of two water-reducer polymers with smooth and rough surfaces on the workability, and the compression strength of concrete from an early age (1 day) up to 28 days of curing was investigated. The polymer contents used in this study varied from 0 to 0.25% (wt%). The initial ratio between water and cement ( $$ \frac{w}{c} $$ ) was 60%, and it slowly reduced to 0.46 by increasing the polymer contents. The compression strength of concrete was increased significantly with increasing the polymer contents by 24–95% depending on the polymer type, polymer content, $$ \frac{w}{c} $$ , and curing age. Because of a fiber net (netting) in the concrete when the polymers were added which leads to a decrease void between the particles, binding the cement particles, therefore, increased rapidly the viscosity for the fresh concrete and the compression strength of the hardened concrete. This study also aims to establish systematic multiscale models to predict the compression strength of concrete containing polymers and to be used by the construction projects with no theoretical restrictions. For that purpose, 88 concrete samples modified with two types of polymer (44 samples for each modification) has been tested, analyzed, and modeled. Linear, nonlinear regression, M5P-tree, and artificial neural network (ANN) approaches were used for the qualifications. In the modeling process, the most relevant parameters affect the strength of concrete, i.e., polymer incorporation ratio (0–0.25% of cement’s mass), water-to-cement ratio (0.46–0.6), and curing ages (1–28 days). Among the used approaches and based on the training data set, the model made based on the nonlinear regression, ANN, and M5P-tree models seem to be the most reliable models. The sensitivity investigation concludes that the curing time is the most dominating parameter for the prediction of the maximum stress (compression strength) of concrete with this dataset.