Artificial Neural Model Research Articles

Plastic Deformation Behavior of Metal Materials: A Review of Constitutive Models

The deformation behavior of metal materials in plastic forming is intimately related to deformation conditions, which are greatly affected by deformation rate, forming temperature, and plastic variables. Macroscopic mechanical properties research is an important basis and technical means to analyze the process parameters and deformation process of metal plastic forming. Therefore, to reveal the influence mechanism of macroscopic mechanical properties of metal materials, and establish material constitutive models under different deformation conditions, it is of great significance to choose reasonable forming parameters and prevent forming defects. There are substantial variances in the macroscopic mechanical characteristics of different materials in the deformation process. In order to accurately predict its deformation behavior, the phenomenological constitutive model, the microscopic constitutive model reflecting the microscopic deformation mechanism, and the artificial neural network constitutive model based on the neural network were constructed respectively on the basis of macroscopic mechanical tests and microscopic microstructure tests. On the basis of the existing research results, the advantages and disadvantages of phenomenological constitutive model, microscopic constitutive model, and neural network constitutive model are compared and analyzed, respectively. The research results of this paper will provide support for the selection of constitutive models for reasonably predicting the deformation behavior of metal materials.

Application of gas chromatographic data and 2D molecular descriptors for accurate global mobility potential prediction

Mobility is a key feature affecting the environmental fate, which is of particular importance in the case of persistent organic pollutants (POPs) and emerging pollutants (EPs). In this study, the global mobility classification artificial neural networks-based models employing GC retention times (RT) and 2D molecular descriptors were constructed and validated. The high usability of RT was confirmed based on the feature selection step performed using the multivariate adaptive regression splines (MARS) tool. Although RT was found to be the most important, according to Kruskal-Wallis ANOVA analysis, it is insufficient to build a robust model, which justifies the need to expand the input layer with 2D descriptors. Therefore the following molecular descriptors: MPC10, WTPT-2, AATS8s, minaaCH, GATS7c, RotBtFrac, ATSC7v and ATSC1p, which were characterized by a high predicting potential were used to improve the classification performance. As a result of machine learning procedure ten of the most accurate neural networks were selected. The external validation showed that the final models are characterized by a high general accuracy score (85.71–96.43%). The high predicting abilities were also confirmed by the micro-averaged Matthews correlation coefficient (MAMCC) (0.73–0.88). To evaluate the applicability of the models, new retention times of selected POPs and EPs including pesticides, polycyclic aromatic hydrocarbons, pharmaceuticals, fragrances and personal care products were measured and used for mobility prediction. Further, the classifiers were used for photodegradation and chlorination products of two popular sunscreen agents, 2-ethyl-hexyl-4-methoxycinnamate and 2-ethylhexyl 4-(dimethylamino)benzoate.

Facile Synthesis of Regenerative Framework Adsorbent for Organic Dyes: Experimental and Artificial Neural Modeling Studies

AbstractThe primary challenge in the design and synthesis of materials having higher affinity, rapid uptake, and superior adsorption capacity is to achieve the balance between performance and economy for real‐world application. The present study unfolded the synthesis of a regenerative framework adsorbent employing greener condition, that shows excellent performance with an ultrahigh adsorption capacity and 99 % removal efficiency for a range of water polluting contaminants, i. e., organic dyes of industrial importance. A pseudo‐second‐order kinetics model shows a favorable fit for the adsorption process. The maximum adsorption capacity calculated by the Langmuir isotherm was found to be highest among many reports in the minimum concentration range, which directly indicates its outstanding efficiency even at lower contaminant concentration. The framework has good reusability and remarkable adsorption performance after being used for several repeated cycles. Consequently, the adsorption behavior of real effluent sample is also investigated using the current system. Additionally, artificial intelligence based on artificial neural network modeling has been executed successfully to estimate and model process parameters of dye removal by adsorption technique which results in mutual correlation between lab‐based techniques and neural based modeling for solving environmental concerns.

Digitization of Prior Authorization in Healthcare Management Using Machine Learning

Background: Prior Authorization is a widely used process by health insurance companies in the United States before they agree to cover prescribed medication under medical insurance. How-ever, the traditional approach includes long-length papers, leading to patients' delayed processing of their claims. This delay may deteriorate the patient’s medical condition. Also, due to man-made er-rors, there is a chance of incorrect decision-making on the claims. On the other hand, physicians are losing their time getting their prescribed medication approved. It is essential to reduce the wait time of patients and the tedious work of physicians for healthcare to be effective. This demands advanced technology that can boost the decision-making process of prior authorization methodology. Objective: This work aims to digitize the prior authorization process by implementing classification algorithms to classify the initial authorization applications into Accepted/Rejected/Partially Accepted classes. A web application that inputs prior authorization claim details and outputs the predicted class of the claim was proposed. Methods: Analyzed and collected significant features by implementing feature selection. Developed classification models using Artificial Neural Networks and Random Forest. Implemented model val-idation techniques to evaluate classifier performance. Results: From the research findings, generic medication cost, type of health insurance plan, addictive nature and side effects of the prescribed drug, patient physical qualities like Age/Gender/Current Medical condition are the significant attributes that impact the decision-making process in the prior authorization process. Then, implemented classifiers exhibited accurate performance on the Train and Test data. Amongst Artificial Neural Networks classification model portrayed higher accuracy. Further a confusion matrix was further analyzed for developed models. In addition, k-fold cross-validation and availed performance evaluation metrics were conducted to validate the model perfor-mance. Conclusion: Ameliorated Healthcare by removing time and location barriers in the Prior Authoriza-tion process while ensuring patients get quality and economical medication. The proposed web ap-plication with a machine learning predictive model as a backend automates the prior authorization process by classifying the applications in a few seconds.

Machine Learning-Assisted Carbon Dot Synthesis: Prediction of Emission Color and Wavelength.

Carbon dots (CDs) have attracted great attention in a range of applications due to their bright photoluminescence, high photostability, and good biocompatibility. However, it is challenging to design CDs with specific emission properties because the syntheses involve many parameters, and it is not clear how each parameter influences the CD properties. To help bridge this gap, machine learning, specifically an artificial neural network, is employed in this work to characterize the impact of synthesis parameters on and make predictions for the emission color and wavelength for CDs. The machine reveals that the choice of reaction method, purification method, and solvent relate more closely to CD emission characteristics than the reaction temperature or time, which are frequently tuned in experiments. After considering multiple models, the best performing machine learning classification model achieved an accuracy of 94% in predicting relative to actual color. In addition, hybrid (two-stage) models incorporating both color classification and an artificial neural network k-ensemble model for wavelength prediction through regression performed significantly better than either a standard artificial neural network or a single-stage artificial neural network k-ensemble regression model. The accuracy of the model predictions was evaluated against CD emission wavelengths measured from experiments, and the minimum mean average error is 25.8 nm. Overall, the models developed in this work can effectively predict the photoluminescence emission of CDs and help design CDs with targeted optical properties.

Construction of artificial neural network diagnostic model and analysis of immune infiltration for periodontitis.

Background: Periodontitis is a chronic inflammatory disease leading to tooth loss in severe cases, and early diagnosis is essential for periodontitis prevention. This study aimed to construct a diagnostic model for periodontitis using a random forest algorithm and an artificial neural network (ANN). Methods: Gene expression data of two large cohorts of patients with periodontitis, GSE10334 and GSE16134, were downloaded from the Gene Expression Omnibus database. We screened for differentially expressed genes in the GSE10334 cohort, identified key periodontitis biomarkers using a Random Forest algorithm, and constructed a classification artificial neural network model, using receiver operating characteristic curves to evaluate its diagnostic utility. Furthermore, patients with periodontitis were classified using a consensus clustering algorithm. The immune infiltration landscape was assessed using CIBERSOFT and single-sample Gene Set Enrichment Analysis. Results: A total of 153 differentially expressed genes were identified, of which 42 were downregulated. We utilized 13 key biomarkers to establish a periodontitis diagnostic model. The model had good predictive performance, with an area under the receiver operative characteristic curve (AUC) of 0.945. The independent cohort (GSE16134) was used to further validate the model's accuracy, showing an area under the receiver operative characteristic curve of 0.900. The proportion of plasma cells was highest in samples from patients with period ontitis, and 13 biomarkers were closely related to immunity. Two molecular subgroups were defined in periodontitis, with one cluster suggesting elevated levels of immune infiltration and immune function. Conclusion: We successfully identified key biomarkers of periodontitis using machine learning and developed a satisfactory diagnostic model. Our model may provide a valuable reference for the prevention and early detection of periodontitis.

A hybrid forecasting model based on the group method of data handling and wavelet decomposition for monthly rivers streamflow data sets

Abstract The natural streamflow of the River is encouraged to forecast through multiple methods. The impartiality of this study is the comparison of the forecast accuracy rates of the time-series (TS) hybrid model with the conventional model. The behavior of the natural monthly statistical chaotic streamflow to use in the forecasting models has been compiled by projecting two distinguished rivers, the Indus and Chenab of Pakistan. Therefore, this article is based on the monthly streamflow forecast analysis that has been reported using the group method of data handling with wavelet decomposition (WGMDH) as a new forecasting attribute. Discrete wavelets decompose the perceived data into sub-series and forecast hydrological variables; these fittingly have been endorsed as inputs in the hybrid model. The forecast efficiency and estimations of the hybrid model are measured by the appropriate statistical techniques such as mean absolute error (RME), root mean square error (RMSE), and correlation coefficients (R) and compared to the group method of data handling (GMDH), least-square support vector machine and artificial neural network conventional models. The comparative analysis shows that the hybrid WGMDH model is more stable and more potent for forecasting river flow than other predictive models and significantly proved that the hybrid model is a robust alternate forecasting tool for TS data sets.

Prediksi Hasil Belajar Hybrid Menggunakan Artificial Neural Network Dengan Multilayer Perceptron

The implementation of online distance learning affects the psychosocial aspects of students so that a combination of offline and face-to-face learning is carried out using Hybrid Learning. The problem in this research is the analysis of the competencies that will be generated, student characteristics, face-to-face interactions, online learning strategies, and lecturers by calculating the data from the Hybrid Learning evaluation to get the best level of accuracy using the Machine Learning technique method. The purpose of this study is to evaluate the results of Hybrid learning using the Artificial Neural Network (ANN) method which is expected to have the right level of accuracy and prediction and tolerate errors so that it can produce good predictions and can be used to model good relationships in finding patterns in the data. . Prediction of Hybrid Learning learning outcomes during the Covid 19 pandemic using Machine Learning techniques consisting of 12 attributes with a total of 1,231 datasets of Hybrid Learning learning outcomes in 2022. The Artificial Neural Network algorithm model uses Retrive operators, Set Roles, optimization parameters, Cross Validation, Apply Models, Performance and Logs. Accuracy results show 99.35% meaning that the results of the learning evaluation using Hybrid Learning with predictions that match and turn out to be appropriate are 1039, then those that do not match are 8. The predictions are very suitable and turn out to be very suitable in 104. Artificial Neural Network with Multilayer Perceptron, with two hidden layer and the format of the first hidden layer is 2 nodes, then the second hidden layer is 5 nodes with an output of 3 nodes

Task-technology fit model: Modelling and assessing the nurses’ satisfaction with health information system using AI prediction models

ABSTRACT Health information system (HIS) plays a significant and supportive role in improving the quality and efficiency of healthcare organizations and their services. The purpose of this paper is to predict the effectiveness of the health information system in terms of nurses’ satisfaction using an artificial neural model. To do so, the study focuses on predicting the nurses’ satisfaction with health information systems in the Gaza strip based on the task technology-fit model (TTF). A total of 164 nurses participated in the questionnaire survey from three hospitals in the Gaza strip to build the prediction dataset of the five task technology-fit factors including task characteristics, technology characteristics, attitude and task-technology fit, and nurses’ satisfaction. After comparing the performance of the neural and regression models, it was found that the neural network is better than the regression model in predicting the nurses’ satisfaction as the results indicated that the performance metrics including accuracy, precision, and recall of the neural model are 92.68%, 95.24%, and 90.90%; respectively. The findings will help the decision-makers and hospital managers to enhance the health information systems and nurses’ satisfaction.

An artificial neural network visible mathematical model for real-time prediction of multiphase flowing bottom-hole pressure in wellbores

Accurate prediction of multiphase flowing bottom-hole pressure (FBHP) in wellbores is an important factor required for optimal tubing design and production optimization. Existing empirical correlations and mechanistic models provide inaccurate FBHP predictions when applied to real-time field datasets because they were developed with laboratory-dependent parameters. Most machine learning (ML) models for FBHP prediction are developed with real-time field data but presented as black-box models. In addition, these ML models cannot be reproduced by other users because the dataset used for training the machine learning algorithm is not open source. These make using the ML models on new datasets difficult. This study presents an artificial neural network (ANN) visible mathematical model for real-time multiphase FBHP prediction in wellbores. A total of 1001 normalized real-time field data points were first used in developing an ANN black-box model. The data points were randomly divided into three different sets; 70% for training, 15% for validation, and the remaining 15% for testing. Statistical analysis showed that using the Levenberg-Marquardt training optimization algorithm (trainlm), hyperbolic tangent activation function (tansig), and three hidden layers with 20, 15 and 15 neurons in the first, second and third hidden layers respectively achieved the best performance. The trained ANN model was then translated into an ANN visible mathematical model by extracting the tuned weights and biases. Trend analysis shows that the new model produced the expected effects of physical attributes on FBHP. Furthermore, statistical and graphical error analysis results show that the new model outperformed existing empirical correlations, mechanistic models, and an ANN white-box model. Training of the ANN on a larger dataset containing new data points covering a wider range of each input parameter can broaden the applicability domain of the proposed ANN visible mathematical model.

Construction of a combined random forest and artificial neural network diagnosis model to screening potential biomarker for hepatoblastoma

The purpose of our study is to identify potential biomarkers of hepatoblastoma (HB) and further explore the pathogenesis of it. Differentially expressed genes (DEGs) were incorporated into the combined random forest and artificial neural network diagnosis model to screen candidate genes for HB. Gene set enrichment analysis (GSEA) was used to analyze the ARHGEF2. Student's t test was performed to evaluate the difference of tumor-infiltrating immune cells (TIICs) between normal and HB samples. Spearson correlation analysis was used to calculate the correlation between ARHGEF2 and TIICs. ARHGEF2, TCF3, TMED3, STMN1 and RAVER2 were screened by the new model. The GSEA of ARHGEF2 included cell cycle pathway and antigen processing presenting pathway. There were significant differences in the composition of partial TIICs between HB and normal samples (p < 0.05). ARHGEF2 was significantly correlated with memory B cells (Cor = 0.509, p < 0.05). These 5 candidate genes contribute to the molecular diagnosis and targeted therapy of HB. And we found "ARHGEF2-RhoA-Cyclin D1/CDK4/CDK6-EF2" is a key mechanism regulating cell cycle pathway in HB. This will be helpful in the treatment of HB. The occurrence of HB is related to abnormal TIICs. We speculated that memory B cells play an important role in HB.

Visual Information Computing and Processing Model Based on Artificial Neural Network.

This paper analyzes the parallel and serial information processing structure of visual system and proposes a visual information processing model with three layers: visual receptor layer, visual information conduction and relay layer, and information processing layer of visual information computing and processing area. Based on the analysis, abstraction, and simplification of the biological prototype of each layer in the visual system, a framework model of an artificial neural system corresponding to the visual system is proposed. An artificial neural network model is proposed to simulate the mechanism of visual attention. A network model is formed by introducing the saliency mask map as additional information on the benchmark network, and the selective enhancement operation is performed on the extracted features in different regions according to the mask map. The experimental results show that the visual computing processing network model can effectively improve the classification performance of the network when the appropriate saliency mask is used. The visual information computing and processing model network can work effectively for different data sets and different structures of the benchmark network, which is a universal network model. The complexity of visual information computing and processing model network is very small, and the improvement of network performance is not at the cost of increasing model complexity, but in the way of improving network efficiency. The performance of artificial neural network visual information computation and processing model is directly related to the performance of saliency map used as mask map.

ARIMAX—Artificial neural network hybrid model for predicting semilooper (Chrysodeixis acuta) incidence on soybean

ARIMAX—Artificial neural network hybrid model for predicting semilooper (Chrysodeixis acuta) incidence on soybean

An artificial neural network model based on autophagy-related genes in childhood systemic lupus erythematosus

BackgroundChildhood systemic lupus erythematosus (cSLE) is a multisystemic, life-threatening autoimmune disease. Compared to adults, SLE in childhood is more active, can cause multisystem involvement including renal, neurological and hematological, and can cause cumulative damage across systems more rapidly. Autophagy, one of the core functions of cells, is involved in almost every process of the immune response and has been shown to be associated with many autoimmune diseases, being a key factor in the interplay between innate and adaptive immunity. Autophagy influences the onset, progression and severity of SLE. This paper identifies new biomarkers for the diagnosis and treatment of childhood SLE based on an artificial neural network of autophagy-related genes.MethodsWe downloaded dataset GSE100163 from the Gene Expression Omnibus database and used Protein–protein Interaction Network (PPI) and Least Absolute Shrinkage and Selection Operator (LASSO) to screen the signature genes of autophagy-related genes in cSLE. A new artificial neural network model for cSLE diagnosis was constructed using the signature genes. The predictive efficiency of the model was also validated using the dataset GSE65391. Finally, "CIBERSORT" was used to calculate the infiltration of immune cells in cSLE and to analyze the relationship between the signature genes and the infiltration of immune cells.ResultsWe identified 37 autophagy-related genes that differed in cSLE and normal samples, and finally obtained the seven most relevant signature genes for cSLE (DDIT3, GNB2L1, CTSD, HSPA8, ULK1, DNAJB1, CANX) by PPI and LASOO regression screening, and constructed an artificial neural network diagnostic model for cSLE. Using this model, we plotted the ROC curves for the training and validation group diagnoses with the area under the curve of 0.976 and 0.783, respectively. Finally, we performed immunoassays on cSLE samples, and the results showed that Plasma cells, Macrophages M0, Dendritic cells activated and Neutrophils were significantly infiltrated in cSLE.ConclusionWe constructed an artificial neural network diagnostic model of seven autophagy-related genes that can be used for the diagnosis of cSLE. Meanwhile, the characteristic genes affect the immune infiltration of cSLE, which may provide new perspectives for the exploration of cSLE treatment and related mechanisms.

Predicting mosquito repellents for clothing application from molecular fingerprint-based artificial neural network SAR models

ABSTRACT Spraying repellents on clothing limits toxicity and allergy problems that can occur when the repellents are directly applied to skin. This also allows the use of higher doses to ensure longer lasting effects. As the number of repellents available on the market is limited, it is necessary to propose new ones, especially by using in silico methods that reduce costs and time. In this context SAR models were built from a dataset of 2027 chemicals for which repellent activity on clothing was measured against Aedes aegypti. The interest of using either the ECFP or MACCS fingerprints as input neurons of a three-layer perceptron was evaluated. Transformation of MACCS bit strings into disjunctive tables led to interesting results. Models obtained with both types of fingerprints were compared to a model including physicochemical and topological descriptors.

Transient Modeling of Induction Machine Using Artificial Neural Network Surrogate Models

A transient model of an induction machine (IM) is developed in this work using an artificial neural network (ANN) surrogate model. The model is suitable to be used for direct-on-line IMs. The finite-element (FE)-based model of IM is used to generate the training, validation, and testing datasets. Different inputs and model configurations are investigated to find an optimal solution in developing the transient model. The proposed transient model is suitable to be used in digital twin services since it can estimate the current and torque accurately in real time based on only voltage and measured shaft speed.

Strong energy component is more important than spectral selectivity in modeling responses of midbrain auditory neurons to wide-band environmental sounds

Modeling central auditory neurons in response to complex sounds not only helps understanding neural processing of speech signals but can also provide insights for biomimetics in neuro-engineering. While modeling responses of midbrain auditory neurons to synthetic tones is rather good, modeling those to environmental sounds is less satisfactory. Environmental sounds typically contain a wide range of frequency components, often with strong and transient energy. These stimulus features have not been examined in the conventional approach of auditory modeling centered on spectral selectivity. To this end, we firstly compared responses to an environmental sound of auditory midbrain neurons across 3 subpopulations of neurons with frequency selectivity in the low, middle and high ranges; secondly, we manipulated the sound energy, both in power and in spectrum, and compared across these subpopulations how their modeled responses were affected. The environmental sound was recorded when a rat was drinking from a feeding bottle (called the ‘drinking sound’). The sound spectrum was divided into 20 non-overlapping frequency bands (from 0 to 20 kHz, at 1 kHz width) and presented to an artificial neural model built on a committee machine with parallel spectral inputs to simulate the known tonotopic organization of the auditory system. The model was trained to predict empirical response probability profiles of neurons to the repeated sounds. Results showed that model performance depended more on the strong energy components than on the spectral selectivity. Findings were interpreted to reflect general sensitivity to rapidly changing sound intensities at the auditory midbrain and in the cortex.

Deep learning reveals cuproptosis features assist in predict prognosis and guide immunotherapy in lung adenocarcinoma.

BackgroundCuproptosis is a recently found non-apoptotic cell death type that holds promise as an emerging therapeutic modality in lung adenocarcinoma (LUAD) patients who develop resistance to radiotherapy and chemotherapy. However, the Cuproptosis’ role in the onset and progression of LUAD remains unclear.MethodsCuproptosis-related genes (CRGs) were identified by a co-expression network approach based on LUAD cell line data from radiotherapy, and a robust risk model was developed using deep learning techniques based on prognostic CRGs and explored the value of deep learning models systematically for clinical applications, functional enrichment analysis, immune infiltration analysis, and genomic variation analysis.ResultsA three-layer artificial neural network risk model was constructed based on 15 independent prognostic radiotherapy-related CRGs. The risk model was observed as a robust independent prognostic factor for LUAD in the training as well as three external validation cohorts. The patients present in the low-risk group were found to have immune “hot” tumors exhibiting anticancer activity, whereas the high-risk group patients had immune “cold” tumors with active metabolism and proliferation. The high-risk group patients were more sensitive to chemotherapy whereas the low-risk group patients were more sensitive to immunotherapy. Genomic variants did not vary considerably among both groups of patients.ConclusionOur findings advance the understanding of cuproptosis and offer fresh perspectives on the clinical management and precision therapy of LUAD.

Modelling of liquid loading in gas wells using a software-based approach

Liquid loading is the most common operational problem influencing gas well productivity for the petroleum operator. Liquid loading is defined as an operational constraint that is associated with gas wells where the major driving mechanism for hydrocarbon production is by the associated gas-driven mechanisms. Liquid loading occurs when liquid accumulated in the tubing or casing results in the gas velocity lower than the critical value (the minimum velocity required for gas to push the liquid out of the gas well), which overtime leads to a hydrostatic back pressure greater than the formation pressure of the well, thereby limiting the flow of gas into the well. The continuous build-up of pressure from liquid loading eventually minimizes well productivity and expensive work over operations. However, current mathematical models to predict liquid loading are flawed with varying inaccuracies and depending on the models deployed will ultimately lead to loss of production time and well productivity. In our work we present prediction of liquid loading using a software-based model incorporating the particle swarm optimization algorithm, genetic algorithm, and artificial neural network and Bayesian neural network algorithms applications. The results of our research findings show that artificial neural network software-based model with a simulated accuracy of 93% and 92% for test and trained data, respectively, outperformed the particle swarm optimization data-driven model with a simulated sensitivity accuracy of 92% and 83%, and genetic algorithm data-driven models with a simulated accuracy of 89% and 83%. The Bayesian neural network was postulated as a robust model because of its simplicity shown to have simulated accuracy of 77% and 73% for train and test data, respectively. Thus software-based code environment and data-driven model developed and presented in this paper may resolve many of current deficiencies and gaps in the current technical literature to predict liquid loading with high precision offering saving in millions of dollars to the operators.

Prediction of mechanical properties for acrylonitrile-butadiene-styrene parts manufactured by fused deposition modelling using artificial neural network and genetic algorithm

ABSTRACT This paper aims to model and predict the mechanical properties of acrylonitrile-butadiene-styrene parts manufactured by fused deposition modelling. An L27 Taguchi array was used to associate the peculiar parameters (nozzle diameter, layer height, fill density, printing velocity, raster orientation and infill pattern) to perform experiments. A three-point bending test was executed on all samples in accordance with the ASTM D7264 standard, aimed to obtain Flexural strength and the 0.2% offset yield strength. An artificial neural network multi-parameter regression model was formed and then the model of the input-output relations developed by this network was optimised by genetic algorithm. This modelling and optimisation suggest a direct relation between choosing process parameters correctly and enhancing performance fused deposition modelling. This method provides optimal solutions for getting allotted output values. Infill pattern plays a critical role in providing strength. Infill patterns can be recommended for desired output within constraints of input factors.