Search

This study assesses the impact of alkaline treatments and volume fractions on biocomposites composed of a high-density polyethylene (HDPE) matrix reinforced with date palm tree fibers (FPDS). Tensile tests were conducted on both untreated and NaOH-treated biocomposites. Additionally, fiber analysis was performed using scanning electron microscopy (SEM) and X-ray diffraction (XRD). The results reveal higher strength and stiffness compared to HDPE, albeit with limited plasticity making the material brittle. The NaOH treatment enhances certain mechanical properties. Further assessments encompassed hardness, density, melt index, and Izod impact tests. Two volume fractions, 20% and 25%, of FPDS were tested. The study establishes a correlation between empirical predictions and artificial neural network (ANN) models. Notably, an ANN architecture consisting of two input factors, 10 hidden nodes, and one output provides the analysis of mechanical properties. This investigation highlights the potential of FPDS-reinforced HDPE biocomposites, emphasizing their mechanical performance under various treatments and fiber levels. [ABSTRACT FROM AUTHOR]

Ensemble analysis is proven to provide advantages in climate change impact assessment based on outputs from climate models. Ensembled series are shown to outperform single-model assessments through increased consistency and stability. This study aims to test the improvement of precipitation estimates through the use of ensemble analysis for south and southwestern Turkey which is known to have complex climatic features due to varying topography and interacting climate forcings. The analysis covers an evaluation of the performance of eight regional climate models (RCMs) from the EUR-11 domain available from the CORDEX database. The historical outputs are evaluated for their representativeness of the current climate of the Mediterranean region and its surroundings in Turkey through a comparison with long-term monthly precipitation time series obtained from ground-based precipitation observations by the use of statistical performance indicators and Taylor diagrams. This is followed by a comparative evaluation of three ensemble methodologies, simple average of the models, multiple linear regression for superensemble, and artificial neural networks (ANN). The analysis results show that the overall performance of ensembled time series is better compared to individual RCMs. ANN generally provided the best performance when all RCMs are used as inputs. Improvement in the performance of ensembling due to the use of nonlinear models is further confirmed by fuzzy inference systems (FIS). Both ANN and FIS generated monthly precipitation time series with higher correlations with those of observations. However, extreme events are poorly represented in the ensembled time series, and this may result in inefficiency in the design of various water structures such as spillways and storm water drainage systems that are based on high return period events. [ABSTRACT FROM AUTHOR]

BACKGROUND: Rapid developments in the field of bioinformatics have provided new methods for the diagnosis of osteoarthritis. Artificial neural networks have powerful data computing and classification capabilities, which have shown better performance in disease diagnosis. OBJECTIVE: To establish a new diagnostic predictive model of osteoarthritis based on artificial neural network and to verify the diagnostic value of the model in osteoarthritis with an external dataset. METHODS: The eligible osteoarthritis-related data sets were downloaded through GEO database search and divided into Train group and Test group. The gene expression matrix of the Train group was analyzed to screen the differentially expressed genes. GO and KEGG enrichment analyses were performed on the differentially expressed genes. Through Lasso regression model, support vector machine model and random forest tree model, the key genes of osteoarthritis were further identified from the differentially expressed genes. The R software “Neuralnet” package was then used to construct the osteoarthritis diagnosis model based on artificial neural network, and the model performance was evaluated by the five-fold cross-validation. Two independent data sets in the Test group were used to verify their diagnostic results. RESULTS AND CONCLUSION: A total of 90 differentially expressed genes related to osteoarthritis were obtained by differential analysis, of which 33 were downregulated and 57 were up-regulated. GO enrichment analysis showed that the differentially expressed genes were mainly involved in the following biological processes, including leukocyte-mediated immunity, leukocyte migration in bone marrow and chemokine production. KEGG enrichment analysis showed that these genes were mainly enriched in rheumatoid arthritis, interleukin-17 signaling pathway and osteoclast differentiation pathway. Five key genes for the diagnosis of osteoarthritis, HMGB2, GADD45A, SLC19A2, TPPP3 and FOLR2, were identified by three machine learning methods. The artificial neural network model of five key genes in the Train group showed that the accuracy was 96.36% and the area under the curve was 0.997. The five-fold cross validation of the neural network model showed that the average area under the curve was greater than 0.9 and the model was of robustness. Two independent data sets in the Test group showed its area under the curve was 0.814 and 0.788 respectively. Therefore, the establishment of an artificial neural network model for the diagnosis of osteoarthritis has a certain diagnostic value. [ABSTRACT FROM AUTHOR]

Hydrogen is a vital resource in the fight against climate change, and it has the potential to revolutionize the energy sector. Our research focused on optimizing the production of high-purity hydrogen using coke oven gas (COG), a valuable hydrogen source in the steel industry. By leveraging advanced artificial neural networks (ANNs), we can predict the performance and exergy efficiency of a 6-bed 12-step vacuum pressure swing adsorption (VPSA) process accurately and efficiently. The Pareto fronts were addressed by combining the evolutionary algorithm with ANNs, and the effects of operating parameters were discussed in detail. Importantly, we found that our VPSA process can achieve a hydrogen purity of 99.99% with 45.2% exergy efficiency. We also demonstrated that using ANNs can significantly enhance VPSA process optimization, making it a valuable tool for extracting high-purity hydrogen from COG. • Multi-objective optimization is performed for the COG-to-hydrogen VPSA process. • Artificial neural networks are trained to predict performance as surrogate models. • The exergy efficiency of 100,000 unique samples is evaluated. • Exergy efficiency is up to 45.2% when producing hydrogen with 99.99% purity. • 9999% hydrogen can be produced at 0.27 bar vacuum pressure with 9 bar feed pressure. [ABSTRACT FROM AUTHOR]

Electro-discharge layering (EDL) is a coating technique, used for fabricating wear and corrosion resistance layer on the parts used in automobile, biomedical and structural applications. Eventhough, selecting the suitable parameters and levels in EDL process is difficult to attain better coating characteristics. We have studied the prediction and effect of process parameters on EDL of magnesium alloy using response surface methodology (RSM)-assisted feed forward back propagation artificial neural network (ANN). In this work, WC–Cu composite coating was deposited on AZ31B magnesium alloy using EDL viz. compaction pressure (CP), discharge current (DC) and pulse on time (PT). Influence of process parameters on electrode deposition rate (EDR) and coating roughness (CR) during EDL of magnesium alloys is studied. It was revealed that the correlation between the experimental values of RSM and predicted values of ANN was 0.991, which is closely to the working limit. Therefore, it was agreed that the established ANN model is suitable for predicting the EDR and CR. Furthermore, effect of parameters on CR and EDR are studied with support of mean effect plots generated by RSM tool. It was studied that the CR and EDR will increase, as increase in DC and PT at processing with low compaction pressured electrode. Conversely, it decreases with increase in CP of the electrode. Mechanism of coating, such as craters and globules were identified in the surface coated with higher DC and PT, resulting in higher surface roughness values. [ABSTRACT FROM AUTHOR]

The automotive S-rail plays a crucial role in frontal car crashes, aiming to absorb impact energy and reduce passenger injury. This paper innovatively presents an optimization approach to determine the optimal configuration of an S-rail featuring a tapered, multicell front section. The structural design of the S-rail is conceptualized within the design space of a heavy quadricycle vehicle and subsequently subjected to numerical investigation through LS-DYNA using nonlinear explicit dynamics analyses. The objective is to maximize energy absorption while minimizing the S-rail's initial peak force (IPF) and mass. An artificial neural network (ANN) is employed to construct surrogate models for the optimization process. The non dominated sorting genetic algorithm, integrated with the ANN, yields an optimal S-rail design. The Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method, using the mean weighting method on the Pareto frontier optimal solution set, is appropriately applied to select the optimal solution. The optimized S-rail shows improved specific energy absorption compared to the baseline model while maintaining a low IPF. In conclusion, this study highlights the superior predictability of an ANN over conventional quadratic response surface methodology. The results confirm the effectiveness of the ANN-based optimization approach and the selection of a compromise solution using the TOPSIS technique. The proposed procedure has substantial potential to enhance the safety and performance of automotive S-rails. [ABSTRACT FROM AUTHOR]

The reliable design of multiphase micro‐structured apparatus requires a precise knowledge of the internal flow regime. Previous research indicated that classifiers based on artificial neural networks (ANN) are relatively simple to develop and provide a reasonable accuracy when trained with data for specific inlet designs. This paper introduces advanced ANN classifiers capable of predicting all relevant flow regimes regardless of the inlet design with a recall of 94 % and above for Taylor, churn, dispersed, rivulet, and parallel flows, between 89 % and 94 % for annular and bubbly flows, and 83 % for Taylor‐annular flow. These classifiers were trained and validated by using more than 13,000 experimental data points extracted from 97 flow maps. [ABSTRACT FROM AUTHOR]

In this letter, attempt has been made to successfully design a pipelined VLSI architecture for the computation of genetic algorithm (GA). The concept of stochastic computing is uniquely exploited in the proposed pipelined GA architecture to attain significant area and power efficiency with reasonably high speed of operation. The prototype 8-bit fixed point GA architecture is realised using VHDL on Xilinx Vivado 2020.3 and implemented on Zynq Ultrascale+ MPSoC (XCZU7EV-2FFVC1156) to train an arbitrary 4:3:2 fully connected neural network in real-time. The performance of the prototype GA architecture in case of real-time training of the neural network outshines the software and other existing GA architectures. The proposed GA-trained 4:3:2 network exhibits 6X reduction in training time and 720X increase in power efficiency, only at the cost of 0.06 % reduction in accuracy with respect to other existing works and software in case of the image classification of MNIST data-set. [ABSTRACT FROM AUTHOR]

Isothermal compression tests for ERNiCrMo-3 alloy specimens were conducted under temperatures of 990–1170 °C with strain rates of 0.01–10 s−1. Hot deformation behavior of ERNiCrMo-3 was analyzed based on the obtained flow stress curves. Three constitutive models were developed to anticipate the flow stress: the Arrhenius model compensated by strain, the Arrhenius model optimized by genetic algorithm (GA), and the artificial neuron network (ANN) model. The correlation coefficient (R) and average absolute relative error (AARE) were used to assess the predictive ability for these three models. The R values of the strain-compensated, GA optimized, and ANN models were 91.04, 93.54, and 99.01%, respectively, while the AARE values were 12.30, 8.19, and 2.92%, respectively. Furthermore, the absolute relative error of the ANN model was the most concentrated and mainly around 0. Therefore, the ANN model provides the maximum predictability and accuracy for flow stress. The ANN model was implemented via subroutine USRMTR in DEFORM 2D to simulate the hot compression process of the ERNiCrMo-3 alloy specimens. The numerical results were in good agreement with the experimental results, indicating a high potential for applicability to actual production processes. Finally, hot working characteristic of ERNiCrMo-3 alloy was analyzed by processing map and microstructural investigation. According to the result, the suitable thermal processing domain for ERNiCrMo-3 alloy should be at a temperature range of 1130–1170 °C and a strain rate range of 0.03–0.36 s−1. [ABSTRACT FROM AUTHOR]

Background: Parkinson's disease (PD) ranks as the second most prevalent neurodegenerative disorder globally, and its incidence is rapidly rising. The diagnosis of PD relies on clinical characteristics. Although current treatments aim to alleviate symptoms, they do not effectively halt the disease's progression. Early detection and intervention hold immense importance. This study aimed to establish a new PD diagnostic model. Methods: Data from a public database were adopted for the construction and validation of a PD diagnostic model with random forest and artificial neural network models. The CIBERSORT platform was applied for the evaluation of immune cell infiltration in PD. Quantitative real-time PCR was performed to verify the accuracy and reliability of the bioinformatics analysis results. Results: Leveraging existing gene expression data from the Gene Expression Omnibus (GEO) database, we sifted through differentially expressed genes (DEGs) in PD and identified 30 crucial genes through a random forest classifier. Furthermore, we successfully designed a novel PD diagnostic model using an artificial neural network and verified its diagnostic efficacy using publicly available datasets. Our research also suggests that mast cells may play a significant role in the onset and progression of PD. Conclusion: This work developed a new PD diagnostic model with machine learning techniques and suggested the immune cells as a potential target for PD therapy. [ABSTRACT FROM AUTHOR]

The use of mechanoregulatory schemes based on finite element (FE) analysis for the evaluation of bone ingrowth around porous surfaces is a viable approach but requires significant computational time and effort. The aim of this study is to develop a combined macro–micro FE and artificial neural network (ANN) framework for rapid and accurate prediction of the site-specific bone ingrowth around the porous beaded-coated tibial implant for total ankle replacement (TAR). A macroscale FE model of the implanted tibia was developed based on CT data. Subsequently, a microscale FE model of the implant-bone interface was created for performing bone ingrowth simulations using mechanoregulatory algorithms. An ANN was trained for rapid and accurate prediction of bone ingrowth. The results predicted by ANN are well comparable to FE-predicted results. Predicted site-specific bone ingrowth using ANN around the implant ranges from 43.04 to 98.24%, with a mean bone ingrowth of around 74.24%. Results suggested that the central region exhibited the highest bone ingrowth, which is also well corroborated with the recent explanted study on BOX®. The proposed methodology has the potential to simulate bone ingrowth rapidly and effectively at any given site over any implant surface. [ABSTRACT FROM AUTHOR]

Interstitial high-dose-rate brachytherapy (HDR-BT) is an effective therapy modality for patients with localized prostate carcinoma. The objectives of the study were to optimise the therapy regime variables using two models: response surface methodology (RSM) and artificial neural network (ANN). Thirty-one studies with 5651 patients were included (2078 patients presented as low-risk, 3077 patients with intermediate-risk, and 496 patients with high-risk). A comparison of these therapy schedules was carried out using an effective biologically effective dose (BED ef) that was calculated assuming the number of treatment days and dose (D) per day. The modelling and optimization of therapy parameters (BED ef and risk level) in order to obtain the maximum biochemical free survival (BFS) were carried out by the RSM and ANN models. An optimal treatment schedule (BFS = 97%) for patients presented with low-risk biochemical recurrence would be D = 26 Gy applied in one application, 2 fractions at least 6 h apart, within an overall treatment time of 1 day (BED ef = 251 Gy) by the RSM and ANN model. For patients presented with intermediate- or high-risk an optimal treatment regime (BFS = 94% and 90%, respectively) would be D = 38 Gy applied in one application, 4 fractions at least 6 h apart, with an overall treatment time of 2 days (BED ef = 279 Gy) by the RSM and ANN models. The RSM and ANN models determine almost the same optimal values for the set of predicted therapy parameters that make a feasible selection of an optimal treatment regime. • HDR-BT is an effective therapy modality for patients with localised prostate carcinoma. • Optimisation of HDR-BT regime variables was carried out by the RSM and ANN models. • The RSM and ANN models determined almost the same optimal values. [ABSTRACT FROM AUTHOR]

We explore the potential for using a nonsmooth loss function based on the max-norm in the training of an artificial neural network without hidden layers. We hypothesise that this may lead to superior classification results in some special cases where the training data are either very small or the class size is disproportional. Our numerical experiments performed on a simple artificial neural network with no hidden layer appear to confirm our hypothesis. [ABSTRACT FROM AUTHOR]

High-energy photon interrogation is a nondestructive technique that is used to detect special nuclear materials and characterize nuclear waste. The development of such systems is complex and requires Monte Carlo simulations to optimize system performance. Monte Carlo simulations rely on various scattering, absorption, and photonuclear cross-section data. While the scattering and absorption cross-section data have been extensively studied and validated with experiments, the results obtained from photonuclear simulations are often found to underpredict measured results, indicating uncertainties in the photonuclear cross sections themselves. Thus, there is a need for new measured results that can be used to quantify underpredictions in simulations using photonuclear cross-section data. In the present work, we interrogated depleted uranium with a 9-MV electron linac and detected photoneutrons with trans-stilbene organic scintillators. The measurement of photoneutrons with organic scintillators is challenging due to the presence of the intense photon flux, which causes issues such as pulse pile-up, detector saturation, and poor signal-to-background ratio. To mitigate these challenges, we used iron and polyethylene shielding of varying thicknesses around the depleted uranium target and a neural network–based digital pulse processing algorithm to recover neutron and photon information from piled-up events. Our goal was to compare the measured photoneutron count rate with the simulated rate obtained using the MCNPX-PoliMi transport code. For a light output window of 0.28 to 2.67 MeVee (1.66- to 6.85-MeV proton recoil energy), we found that the simulated count rate obtained using the ENDF/B-VII photonuclear cross-section library underpredicts the measured rate by 32.8% $$ \pm $$ ± 3.2%. Additionally, we compared the simulated and measured photoneutron light output distributions. For the least thicknesses of shielding, the simulation was found to underpredict measurements in the 0.70- to 2.67-MeVee light output window. For the greatest thicknesses of shielding, the simulation was found to underpredict the measurement across the entire light output window of 0.28 to 2.67 MeVee. [ABSTRACT FROM AUTHOR]

First-principles-based modelings have been extremely successful in providing crucial insights and predictions for complex biological functions and phenomena. However, they can be hard to build and expensive to simulate for complex living systems. On the other hand, modern data-driven methods thrive at modeling many types of high-dimensional and noisy data. Still, the training and interpretation of these data-driven models remain challenging. Here, we combine the two types of methods to model stochastic neuronal network oscillations. Specifically, we develop a class of artificial neural networks to provide faithful surrogates to the high-dimensional, nonlinear oscillatory dynamics produced by a spiking neuronal network model. Furthermore, when the training data set is enlarged within a range of parameter choices, the artificial neural networks become generalizable to these parameters, covering cases in distinctly different dynamical regimes. In all, our work opens a new avenue for modeling complex neuronal network dynamics with artificial neural networks. [ABSTRACT FROM AUTHOR]

• A model to explore the multiphysics characteristic of semiconductors is presented. • The model determines the thermal and mechanical properties of complex patterns. • Results show a strong correlation between thermal and mechanical properties. • Dielectric bands asymmetrically affect thermal and mechanical behavior. • Machine learning can predict one property from another. As semiconductor performance improves through advanced package designs, it becomes important to consider both thermal and mechanical properties. Better understanding their relationship enhances system design and optimization. This study has introduced a numerical method that takes into account both thermal and mechanical fluxes to construct thermal and mechanical property maps for practical application in semiconductor engineering. Furthermore, this paper investigated the relationship between the two properties using the constructed property maps in commercially available semiconductor packages. Owing to the complexity of packaging design patterns, a coupled isoparametric mapping and machine learning (ML) method was introduced to investigate their effects. The model then determines and investigates the anisotropic equivalent thermal and mechanical properties of the commercialized semiconductor packages with complex pattern designs, according to the package materials, volume fractions of each material, and design patterns. This approach pursues more sophistication and practicality compared to simplified composite structure property evaluation models. Additionally, all processes in the algorithm are automated based on ML methods, making them practically applicable in the semiconductor industry. The study shows that there is a strong correlation between the thermal and mechanical characteristics in the complex package patterns. This relationship was able to form a fairly clear category based on the shape of the dielectric band because the dielectric band had significantly different effects on the thermal and mechanical fluxes. The study discussed the physical implications of the similarities and differences between the two behaviors and validated the results of the equivalent properties through finite element (FE) analysis. Finally, the study cross-validates the relationship by showing that information from one flux behavior can be used to predict the other based on the ML method. Based on the results, this paper can provide a better understanding of thermal and mechanical fluxes in semiconductor package patterns for package design and optimization. [ABSTRACT FROM AUTHOR]

Depositional setting characterization is one of the most important tasks in petroleum basin analysis. In this regard, artificial intelligence has emerged as a game-changer in the field of oil reservoir characterization, offering a myriad of benefits that significantly enhance the exploration and production processes within the oil and gas industry. Artificial Intelligence driven algorithms can efficiently process geological and geophysical data, well logs, and seismic information, allowing for a more comprehensive understanding of reservoir properties. To obtain more appropriate image of high reservoir quality zones, a case study was performed by integrating 3D seismic and well data related to on onshore oilfield, west of Iran. Supporting data were acquired from existing geochemical analyses of scanning electron microscope, thin-section investigation, and special core analysis laboratory measurements related to three wells of the studied oil field. The methodology developed in this study consists of three main phases, at the first step a complete thin section analysis is done to identify the main facies of the studied reservoir. Four mains microfacies and representative sedimentary environment were identified including: (a) Foraminifera bioclastic wackestone (Mid ramp-Distal), (b) Benthic foraminifera bioclast peloid wackestone to packstone (Mid ramp-Proximal), (c) Coated grains bioclast packstone to grainstone (Inner ramp-Shoal), (d) Bioclast Peloid wackestone (Inner ramp-Lagoon). To create a continuous pore throat size log, porosity and permeability logs were initially generated through petrophysical evaluation and artificial neural network analysis, achieving an accuracy of R2 = 0.95 for porosity and R2 = 0.84 for permeability. Subsequently, the pore throat size log was generated using the Winland equation, and the results were calibrated with pore throat sizes calculated from capillary pressure data analysis using the Washburn equation. Two different approaches including FZI and K-means clustering methods are also employed to recognize Hydraulic Flow Units. According to the Sum of Squared Errors (SSE) plot of the K-means algorithm, beyond three clusters, the reduction in SSE becomes marginal, suggesting that three clusters suit the dataset appropriately. In the next step, the sparse spike algorithm was used to generate a 3D acoustic impedance cube. Finally, post-stack seismic attributes, including the inverse of acoustic impedance, instantaneous frequency, a filter of 15/20–25/30, and amplitude-weighted phase, were selected to create a 3D pore throat size cube using a Probabilistic Neural Network, demonstrating a strong correlation of R2 = 91. The resulting pore throat size cube effectively illustrates that the Ilam-Upper and Ilam Main zones, which include HFU 3, exhibit high reservoir quality, with porosity, permeability, and mean pore throat size values of 16%, 20–67 mD, and 3–6 microns, respectively. In summary, the integration of acoustic impedance in oil reservoir characterization revolutionizes how we extract and manage hydrocarbon resources, driving efficiency, cost-effectiveness, and sustainability. [ABSTRACT FROM AUTHOR]

This study explores the relationship between natural fiber filling density (10%, 15%, 25%) and its impact on the bending properties of polymer compounds reinforced with Diss, Sisal and Luffa fibers. Using advanced techniques like fiber analysis and Fourier transform infrared spectrometry (FTIR), the research reveals that a 25% filling density results in the highest stress values (25.61 MPa, 22.21 MPa and 20.88 MPa) for Diss, Sisal and Luffa compounds, respectively, fostering robust bonds in Diss-reinforced polymers. The Artificial Neural Network (ANN) model demonstrates superior predictive capability with correlation coefficients exceeding 0.99 for stress and displacement, outperforming Response Surface Methodology (RSM). Analysis of Variance (ANOVA) underscores the impact of sample section parameters and fiber rate on stress, establishing the significance of type parameters and fiber rate on displacement. This integration of ANN and RSM represents a paradigm shift in predicting bending mechanical properties, advancing our understanding of composite materials for innovative applications. [ABSTRACT FROM AUTHOR]

Liquid hydrogen (LH 2) is a future energy carrier compared to most conventional fuels. One of the practical bottlenecks in non-venting LH 2 tanks is to accurately predict the pressure evolution. Accordingly, many thermal models and correlations were developed; however significant deviations arose due to inherent complex thermal phenomena which have not been addressed yet. A general artificial neural network (ANN) model has been first developed based on 448 datasets amassed from different literature sources to predict the pressure evolution under different heat fluxes, initial filling ratios, as well as tank volumes and shapes. The model has been configured with a typical multi-layer feed-forward structure using a back-propagation learning algorithm and Levenberg-Marquardt training algorithm. For the ANN model, a preliminary optimization process has been conducted for reaching the optimal values for training-to-testing datasets ratio, learning parameters and optimum network structure. Besides, some criteria were set to evaluate the ANN model performance. The ANN model with the configuration of 5-11-11-1 was found to be adopted for such a prediction task. Compared to experimental results and literature models and correlations, the ANN model has been proven to predict the pressure evolution in non-venting self-pressurized LH 2 tanks with a mean square error of 1.1779 × 10−5 s2, root mean square error of 0.0034 s, mean absolute error of 6%, mean relative error of 0.2% as well as a satisfactory correlation coefficient of 0.99934 between the network outputs and corresponding targets. This research work provides an effective modeling approach based on artificial neural network to solve one complex problem related with LH 2 tanks with fast, accurate and consistent results. Also, the results of the present work can offer some practical guidance for the design of LH 2 tanks. Further, the present work gives some insights into new application areas of ANN which will be the emphasis of future research work. • Artificial neural network is applied for predicting pressure evolution in LH 2 tanks. • ANN model with a configuration of 5-11-11-1 is proposed for LH 2 tanks. • ANN prediction error of 1.95% was achieved lower than other thermal models. • Studied results reveal the ANN model can map the implicit relationship accurately. [ABSTRACT FROM AUTHOR]

Maximizing the recombinant protein yield necessitates optimizing the production medium. This can be done using a variety of methods, including the conventional “one‐factor‐at‐a‐time” approach and more recent statistical and mathematical methods such as artificial neural network (ANN), genetic algorithm, etc. Every approach has advantages and disadvantages of its own, yet even when a technique has flaws, it is nevertheless used to get the best results. Here, one categorical variable and four numerical parameters, including post‐induction time, inducer concentration, post‐induction temperature, and pre‐induction cell density, were optimized using the 232 experimental assays of the central composite design. The direct and indirect effects of factors on the yield of anti‐epithelial cell adhesion molecule extracellular domain fragment antibody were examined using statistical methods. The analysis of variance results indicate that the response surface methodology (RSM) model is effective in predicting the amount of produced single‐chain fragment variable (p‐value = 0.0001 and R2 = 0.905). For ANN modeling, the evaluation using normalized root mean square error (NRMSE) and R2 values shows a good fit (R2 = 0.942) and accurate predictions (NRMSE = 0.145). The analysis of error parameters and R2 of a dataset, which contained 30 data points randomly selected from the complete dataset, showed that the ANN model had a higher R2 value (0.968) compared to the RSM model (0.932). Furthermore, the ANN model demonstrated stronger predictive ability with a lower NRMSE (0.048 vs. 0.064). Induction at the cell density of 0.7 and an isopropyl β‐D‐1‐thiogalactopyranoside concentration of 0.6 mM for 32 h at 30°C in BW25113 was the ideal culture condition leading to the protein yield of 259.51 mg/L. Under the optimum conditions, the output values predicted by the ANN model (259.83 mg/L) were more in line with the experimental data (259.51 mg/L) than the RSM (276.13 mg/L) expected value. This outcome demonstrated that the ANN model outperforms the RSM in terms of prediction accuracy. [ABSTRACT FROM AUTHOR]

The growing popularity of artificial intelligence approaches has led to their application in a wide range of engineering fields. The most widely used artificial intelligence tool, artificial neural networks, can be used to predict data with high accuracy. An artificial neural network approach is being used to predict effective and accurate thermal conductivity and viscosity models for hybrid nanofluid systems. Here, new types of correlations relating to the thermophysical properties of Fly Ash–Cu nanoparticles with diameter sized 15.2 nm and which are temperature-dependent are developed. The highest thermal conductivity and viscosity values were obtained for hybrid nanofluids with a mixture ratio of 20:80, with maximum amplification exceeding 83.2% and 65%, respectively, over the base fluid. The Fly Ash–Cu/water hybrid nanofluid's viscosity and thermal conductivity are evaluated for a concentration range of 0–4%. The evaluation of the Fly Ash–Cu/water hybrid nanofluids system at concentrations ranging from 0 to 4% most likely entails a scientific or engineering study aimed at understanding the behavior and properties of this nanofluid mixture. Nanoparticles can agglomerate or settle in the base fluid over time, compromising the stability of the nanofluids. Researchers may be interested in determining how varied quantities of Fly Ash and Cu nanoparticles affect the nanofluid's stability and sedimentation behavior. The heat transfer potential is examined within the optimistic range of temperatures of 30–80∘C. Many fruitful results for turbulence and solar energy have been drawn. The Mouromtseff number achieved an optimal value for all concentration levels. The heat transfers of turbulent flow and thermal conductivity of hybrid nanofluids increase with the augmented values of concentrations and temperature. Researchers found an increase in thermal conductivity of hybrid nanofluids at 0–4% concentrations, potentially impacting heat transfer applications. The conclusion explores the potential integration of the developed correlations and neural network model into practical engineering or industrial applications involving solar energy and turbulence appliances. In this work, we extend the work of Kanti et al. [Sol. Energy Mater. Sol. Cells 234 (2022) 111423] which is on the properties of water-based fly ash-copper hybrid nanofluid for solar energy applications. [ABSTRACT FROM AUTHOR]

Patient-specific 3D models of the human mandible are finding increasing utility in medical fields such as oral and maxillofacial surgery, orthodontics, dentistry, and forensic sciences. The efficient creation of personalized 3D bone models poses a key challenge in these applications. Existing solutions often rely on 3D statistical models of human bone, offering advantages in rapid bone geometry adaptation and flexibility by capturing a range of anatomical variations, but also a disadvantage in terms of reduced precision in representing specific shapes. Considering this, the proposed parametric model allows for precise manipulation using morphometric parameters acquired from medical images. This paper highlights the significance of employing the parametric model in the creation of a personalized bone model, exemplified through a case study targeting mandibular prognathism reconstruction. A personalized model is described as 3D point cloud determined through the utilization of series of parametric functions, determined by the application of geometrical morphometrics, morphology properties, and artificial neural networks in the input dataset of human mandible samples. With 95.05% of the personalized model's surface area displaying deviations within −1.00–1.00 mm relative to the input polygonal model, and a maximum deviation of 2.52 mm, this research accentuates the benefits of the parametric approach, particularly in the preoperative planning of mandibular deformity surgeries. [ABSTRACT FROM AUTHOR]

In the last decade, mesoporous hydroxyapatite (HA) materials have received much attention regarding environmental applications (including adsorption of soil and water contaminants, films for lithium batteries, catalysts, and support systems for catalysis) and open up ample opportunities regarding biomedical devices and structures. Molybdenum ions may strongly influence catalytic and sorption properties of materials. The present study shows an influence of hydrothermal-synthesis conditions (including temperature and duration of the treatment) on phase composition, textural properties, and morphology of molybdate-containing HA. Powders with high crystallinity manifested Brunauer–Emmett–Teller specific surface area up to 55 m2/g with a pore volume of 0.33 cm3/g. The molybdate-containing HA powders were investigated by chemical analysis, X-ray diffraction, Fourier-transform IR spectroscopy, and transmission electron microscopy. Their specific surface area, pore volume, and pore size distributions were determined by low-temperature nitrogen adsorption measurements, zeta potential was quantified, and electron paramagnetic resonance spectroscopy was performed. The concentration of MoO 4 2− preserving the one-phase state was found, and the impact of hydrothermal treatment mode was analyzed by machine learning using artificial neural networks. [Display omitted] [ABSTRACT FROM AUTHOR]

Epilepsy surgery is effective for patients with medication-resistant seizures, however 20–40% of them are not seizure free after surgery. Aim of this study is to evaluate the role of linear and non-linear EEG features to predict post-surgical outcome. We included 123 paediatric patients who underwent epilepsy surgery at Bambino Gesù Children Hospital (January 2009–April 2020). All patients had long term video-EEG monitoring. We analysed 1-min scalp interictal EEG (wakefulness and sleep) and extracted 13 linear and non-linear EEG features (power spectral density (PSD), Hjorth, approximate entropy, permutation entropy, Lyapunov and Hurst value). We used a logistic regression (LR) as feature selection process. To quantify the correlation between EEG features and surgical outcome we used an artificial neural network (ANN) model with 18 architectures. LR revealed a significant correlation between PSD of alpha band (sleep), Mobility index (sleep) and the Hurst value (sleep and awake) with outcome. The fifty-four ANN models gave a range of accuracy (46–65%) in predicting outcome. Within the fifty-four ANN models, we found a higher accuracy (64.8% ± 7.6%) in seizure outcome prediction, using features selected by LR. The combination of PSD of alpha band, mobility and the Hurst value positively correlate with good surgical outcome. [ABSTRACT FROM AUTHOR]

Artificial neural networks and supervised learning have become an essential part of science. Beyond using them for accurate input-output mapping, there is growing attention to a new feature-oriented approach. Under the assumption that networks optimised for a task may have learned to represent and utilise important features of the target system for that task, scientists examine how those networks manipulate inputs and employ the features networks capture for scientific discovery. We analyse this approach, show its hidden caveats, and suggest its legitimate use. We distinguish three things that scientists call a ‘feature’: parametric, diagnostic, and real-world features. The feature-oriented approach aims for real-world features by interpreting the former two, which also partially rely on the network. We argue that this approach faces a problem of non-uniqueness: there are numerous discordant parametric and diagnostic features and ways to interpret them. When the approach aims at novel discovery, scientists often need to choose between those options, but they lack the background knowledge to justify their choices. Consequentially, features thus identified are not promised to be real. We argue that they should not be used as evidence but only used instrumentally. We also suggest transparency in feature selection and the plurality of choices. [ABSTRACT FROM AUTHOR]

The aim of this investigation was to delve into the impact of abrasive water jet machining (AWJM) process variables on the surface roughness (Ra) and kerf angle (Ka) of hybrid fiber‐reinforced polyester composites. Utilizing both response surface methodology (RSM) and artificial neural network (ANN) prediction models, the study sought to optimize the input parameters for abrasive water jet machining, specifically in the context of paddy straw and PALF‐reinforced polyester hybrid composites. The process parameters targeted for optimization included the abrasive flow rate, traverse rate, and standoff distance during AWJM. The investigation identified an optimal combination of AWJM parameters that effectively meets the practical requirements for machining hybrid fiber‐reinforced polyester composites. According to the RSM, the suggested optimal values for the process parameters are an abrasive flow rate set at 300 g/min, traverse speed at 110 mm/min, and standoff distance at 1 mm. The ANN exhibited robust predictive capabilities, achieving high R2 scores of 0.932 and 0.962 for surface roughness and kerf angle, respectively. To enhance the performance of abrasive water jet machining and minimize surface roughness and kerf angle, the researchers conducted an optimization of the process parameters. Subsequently, confirmation experiments were executed to validate the predictive model and fine‐tune the set of process parameters for practical application. Highlights: Investigated AWJM impact on Ra value and kerf angle of hybrid composites.Used RSM and ANN models for parameter optimization in biocomposite.Optimal AWJM parameters: AFR (300 g/min), TS (110 mm/min), and SOD (1 mm).ANN showed strong predictions: R2 scores 0.932 (Ra) and 0.962 (Ka).Confirmation experiments validated the predictive model for applications. [ABSTRACT FROM AUTHOR]

Background: Survival prediction using high-dimensional molecular data is a hot topic in the field of genomics and precision medicine, especially for cancer studies. Considering that carcinogenesis has a pathway-based pathogenesis, developing models using such group structures is a closer mimic of disease progression and prognosis. Many approaches can be used to integrate group information; however, most of them are single-model methods, which may account for unstable prediction. Methods: We introduced a novel survival stacking method that modeled using group structure information to improve the robustness of cancer survival prediction in the context of high-dimensional omics data. With a super learner, survival stacking combines the prediction from multiple sub-models that are independently trained using the features in pre-grouped biological pathways. In addition to a non-negative linear combination of sub-models, we extended the super learner to non-negative Bayesian hierarchical generalized linear model and artificial neural network. We compared the proposed modeling strategy with the widely used survival penalized method Lasso Cox and several group penalized methods, e.g., group Lasso Cox, via simulation study and real-world data application. Results: The proposed survival stacking method showed superior and robust performance in terms of discrimination compared with single-model methods in case of high-noise simulated data and real-world data. The non-negative Bayesian stacking method can identify important biological signal pathways and genes that are associated with the prognosis of cancer. Conclusions: This study proposed a novel survival stacking strategy incorporating biological group information into the cancer prognosis models. Additionally, this study extended the super learner to non-negative Bayesian model and ANN, enriching the combination of sub-models. The proposed Bayesian stacking strategy exhibited favorable properties in the prediction and interpretation of complex survival data, which may aid in discovering cancer targets. [ABSTRACT FROM AUTHOR]

The interpretation of breast magnetic resonance imaging (MRI) in the healthcare field depends on the good knowledge and experience of radiologists. Recent developments in artificial intelligence (AI) have shown advances in the field of radiology. However, the desired levels have not been reached in the field of radiology yet. In this study, a novel model structure is proposed to characterize the diagnostic performance of AI technology for individual breast dynamic contrast material-enhanced (DCE) MRI sequences. In the proposed model structure, Inception-v3, EfficientNet-B3, and DenseNet-201 models were used as hybrids together with the Yolo-v3 algorithm to detect breast and cancer regions. In the proposed model, DCE-MRI sequences (T2, ADC, Diffusion, Non-Contrast Fat Non-Suppressed T1, Non-Contrast Fat Suppressed T1, Contrast Fat Suppressed T1, and Subtraction T1) were evaluated separately and validation was made, thus providing a unique perspective. According to the validation results, the model structure with the best performance was determined as Yolo-v3 + DenseNet-201. With this model structure, 92.41% accuracy, 0.5936 loss, 92.44% sensitivity, and 92.44% specificity rates were obtained. In addition, it was determined that the results obtained without using contrast material in the best model were 91.53% accuracy, 0.9646 loss, 92.19% sensitivity, and 92.19% specificity. Therefore, it is predicted that the need for contrast material use can be reduced with the help of this model structure. [ABSTRACT FROM AUTHOR]

Purpose: Fused Filament Fabrication (FFF) is one of the growing technologies in additive manufacturing, that can be used in a number of applications. In this method, process parameters can be customized and their simultaneous variation has conflicting impacts on various properties of printed parts such as dimensional accuracy (DA) and surface finish. These properties could be improved by optimizing the values of these parameters. Design/methodology/approach: In this paper, four process parameters, namely, print speed, build orientation, raster width, and layer height which are referred to as "input variables" were investigated. The conflicting influence of their simultaneous variations on the DA of printed parts was investigated and predicated. To achieve this goal, a hybrid Genetic Algorithm – Artificial Neural Network (GA-ANN) model, was developed in C#.net, and three geometries, namely, U-shape, cube and cylinder were selected. To investigate the DA of printed parts, samples were printed with a central through hole. Design of Experiments (DoE), specifically the Rotational Central Composite Design method was adopted to establish the number of parts to be printed (30 for each selected geometry) and also the value of each input process parameter. The dimensions of printed parts were accurately measured by a shadowgraph and were used as an input data set for the training phase of the developed ANN to predict the behavior of process parameters. Then the predicted values were used as input to the Desirability Function tool which resulted in a mathematical model that optimizes the input process variables for selected geometries. The mean square error of 0.0528 was achieved, which is indicative of the accuracy of the developed model. Findings: The results showed that print speed is the most dominant input variable compared to others, and by increasing its value, considerable variations resulted in DA. The inaccuracy increased, especially with parts of circular cross section. In addition, if there is no need to print parts in vertical position, the build orientation should be set at 0° to achieve the highest DA. Finally, optimized values of raster width and layer height improved the DA especially when the print speed was set at a high value. Originality/value: By using ANN, it is possible to investigate the impact of simultaneous variations of FFF machines' input process parameters on the DA of printed parts. By their optimization, parts of highly accurate dimensions could be printed. These findings will be of significant value to those industries that need to produce parts of high DA on FFF machines. [ABSTRACT FROM AUTHOR]

The present research introduces the best architecture model for predicting the unsoaked California bearing ratio (CBRu) of soil by comparing the models based on the least square support vector machine (LSSVM), long- short-term memory (LSTM), and artificial neural network (ANN) approach. The two kernel functions, linear and polynomial, have been selected to create LSSVM models. The developed LSTM models have been optimized by the Adam algorithm. In the employed ANN models, the Levenberg–Marquardt (LM), BFGS Quasi-Newton (BFG), scaled conjugate gradient (SCG), gradient descent with momentum (GDM), gradient descent (GD), and gradient descent with adaptive learning (GDA) algorithms have been used in the backpropagation process. For this purpose, three databases, such as training, testing and validation, have been compiled from the published research. A laboratory database has been developed by performing laboratory experiments for soil samples collected from and around Kota, Rajasthan, used for cross-validation of the best architecture model. The statistical tools, such as root means square error (RMSE), mean absolute error (MAE), correlation coefficient (R), mean absolute percentage error (MAPE), variance accounted for (VAF), weighted mean absolute percentage error (WMAPE), Nash–Sutcliffe efficiency (NS), normalized mean bias error (NMBE), Legate and McCabe's index (LMI), root mean square error to observation's standard deviation ratio (RSR), a20-index, index of agreement (IOA) and index of scatter (IOS) have been used to measure the performance of the models. The LSTM model MD 14 has achieved higher performance and accuracy (RMSE = 0.9127%, MAE = 0.8114%, R = 0.9863%, MAPE = 9.0772%, VAF = 97.26, WMAPE = 0.0669%, NS = 0.9708, NMBE = 0.0687%, LMI = 0.1926, RSR = 0.1708, a20-index = 93.88, IOA = 0.9037 and IOS = 0.0752) in testing phase. For the performance validation, model (MD) 14 has predicted the CBRu of the validation database. Also, model MD 14 has attained higher performance (RMSE = 1.2671%, MAE = 1.0161%, R = 0.9909) in the validation phase. By comparing the performances and performing score analysis, the LSTM model MD 14 has been recognized as the best architecture model for predicting the unsoaked CBR of soil. Moreover, model MD 14 has gained over 96% (R = 0.9689) accuracy in predicting the CBRu of laboratory-tested soil samples. The present research also represents that the nonlinear approach has achieved higher performance with a high overfitting ratio. In addition, the artificial neural network requires a large database to predict the unsoaked CBR with higher performance and the least overfitting ratio. The present research also rejects the null hypothesis of normality. Sensitivity analysis illustrates that gravel content and maximum dry density of soil affect the prediction of unsoaked CBR. [ABSTRACT FROM AUTHOR]

This study proposes a novel approach for detecting and recognizing asphalt pavement cracks. Image processing techniques including discrete Fourier transform (DFT), steerable filter (SF), and integral projection (IP) are combined to extract features from pavement images. The new method employs DFT as a low-pass filter to enhance the original digital image. Accordingly, a crack salient map is constructed by the utilization of SF. Based on such salient map, IP of the image is established and its statistical properties are computed to characterize the condition of a pavement image. A soft computing model that integrates the artificial neural network, Levenberg–Marquardt (LM) backpropagation training, and Differential Flower Pollination (DFP) algorithm is put forward for the task of pavement recognition. The integration of LM and DFP aims at combining the global search capability of metaheuristic and the local search efficiency of the steepest descent-based backpropagation approach. During the optimization process of DFP, LM algorithm is performed with a randomly selected member of the population to boost the convergence rate and enhance the searching effectiveness. This new training framework of ANN is, therefore, named as ANN with Random Boosted Differential Flower Pollination (RBDFP-ANN). A data set of pavement images including four class labels (no crack, longitudinal crack, transverse crack, and alligator crack) has been collected to train and verify the RBDFP-ANN. Experimental results supported by statistical test confirm that the proposed model achieves a superior prediction accuracy rate (89.25%) over other benchmark methods. Accordingly, the RBDFP-ANN model is very potential to be used by transportation agencies in the task of asphalt pavement inspection. [ABSTRACT FROM AUTHOR]

This paper presents a new accurate method to compute the mechanical response of pavement structures using an Artificial Neural Network (ANN) model coupled with Multi-Layer Elastic Analysis (MLEA). The ANN model is used to improve the numerical integration of the response function used in the MLEA method. It requires four inputs: total pavement thickness, the diameter of the contact area, radial distance, and depth of the response point; and it was trained on one million hypothetical pavement structures. The developed method has been validated by a comparative analysis against boundary conditions, finite element analysis, and available MLEA solutions using various hypothetical pavement structures. The results demonstrate that the developed solution gives excellent response in the vicinity of the pavement surface together with a significant improvement in computational efficiency. [ABSTRACT FROM AUTHOR]

Purpose: This study aims to elucidate the machining properties of low-cost expanded clay-reinforced syntactic foams by using different neural network models for the first time in the literature. The main goal of this endeavor is to create a casting machining-neural network modeling flow-line for real-time foam manufacturing in the industry. Design/methodology/approach: Samples were manufactured via an industry-based die-casting technology. For the slot milling tests performed with different cutting speeds, depth of cut and lubrication conditions, a 3-axis computer numerical control (CNC) machine was used and the force data were collected through a digital dynamometer. These signals were used as input parameters in neural network modelings. Findings: Among the algorithms, the scaled-conjugated-gradient (SCG) methodology was the weakest average results, whereas the Levenberg–Marquard (LM) approach was highly successful in foreseeing the cutting forces. As for the input variables, an increase in the depth of cut entailed the cutting forces, and this circumstance was more obvious at the higher cutting speeds. Research limitations/implications: The effect of milling parameters on the cutting forces of low-cost clay-filled metallic syntactics was examined, and the correct detection of these impacts is considerably prominent in this paper. On the other side, tool life and wear analyses can be studied in future investigations. Practical implications: It was indicated that the milling forces of the clay-added AA7075 syntactic foams, depending on the cutting parameters, can be anticipated through artificial neural network modeling. Social implications: It is hoped that analyzing the influence of the cutting parameters using neural network models on the slot milling forces of metallic syntactic foams (MSFs) will be notably useful for research and development (R&D) researchers and design engineers. Originality/value: This work is the first investigation that focuses on the estimation of slot milling forces of the expanded clay-added AA7075 syntactic foams by using different artificial neural network modeling approaches. [ABSTRACT FROM AUTHOR]

This paper presents a new approach for the online estimation of stator and rotor resistance of induction motors for speed sensorless vector-controlled drives, using feed-forward artificial neural networks with advanced adaptive learning rates. For the rotor resistance estimation, a neural network model based on rotor speed and stator currents is developed. The rotor flux linkages acquired from the voltage model are compared with the neural network model. The feed-forward neural network employs an adaptive learning rate as the function of the obtained error during training for quick convergence with minimal estimation error. A two-layered neural network model based on the stator voltage and current equations is developed for the stator resistance estimation. The d-q axes stator currents obtained from the developed model are compared with the acquired d-q axes stator currents. For the fast convergence with minimal estimation error, an adaptive learning rate as the function of error is adopted during training. Furthermore, the neural network estimates the induction motor's speed. The simulation and experimental results justify that the developed algorithms track variation in the resistances quickly and precisely along with the speed as compared with the conventional constant learning rate algorithm, leading to reliable operation of the drive. [ABSTRACT FROM AUTHOR]

The automobile industry relies primarily on spot welding operations, particularly resistance spot welding (RSW). The performance and durability of the resistance spot-welded joints are significantly impacted by the welding quality outputs, such as the shear force, nugget diameter, failure mode, and the hardness of the welded joints. In light of this, the present study sought to determine how the aforementioned welding quality outputs of 0.5 and 1 mm thick austenitic stainless steel AISI 304 were affected by RSW parameters, such as welding current, welding time, pressure, holding time, squeezing time, and pulse welding. In order to guarantee precise evaluation and experimental analysis, it is essential that they are supported by a numerical model using an intelligent model. The primary objective of this research is to develop and enhance an intelligent model employing artificial neural network (ANN) models. This model aims to provide deeper knowledge of how the RSW parameters affect the quality of optimum joint behavior. The proposed neural network (NN) models were executed using different ANN structures with various training and transfer functions based on the feedforward backpropagation approach to find the optimal model. The performance of the ANN models was evaluated in accordance with validation metrics, like the mean squared error (MSE) and correlation coefficient (R2). Assessing the experimental findings revealed the maximum shear force and nugget diameter emerged to be 8.6 kN and 5.4 mm for the case of 1–1 mm, 3.298 kN and 4.1 mm for the case of 0.5–0.5 mm, and 4.031 kN and 4.9 mm for the case of 0.5–1 mm. Based on the results of the Pareto charts generated by the Minitab program, the most important parameter for the 1–1 mm case was the welding current; for the 0.5–0.5 mm case, it was pulse welding; and for the 0.5–1 mm case, it was holding time. When looking at the hardness results, it is clear that the nugget zone is much higher than the heat-affected zone (HZ) and base metal (BM) in all three cases. The ANN models showed that the one-output shear force model gave the best prediction, relating to the highest R and the lowest MSE compared to the one-output nugget diameter model and two-output structure. However, the Levenberg–Marquardt backpropagation (Trainlm) training function with the log sigmoid transfer function recorded the best prediction results of both ANN structures. [ABSTRACT FROM AUTHOR]

The timely and precise prediction of cardiovascular disease (CVD) risk is essential for effective prevention and intervention. This study proposes a novel framework that integrates the two-phase Taguchi method (TPTM), the hyperparameter artificial neural network (HANN), and a genetic algorithm (GA) called TPTM-HANN-GA. This framework efficiently optimizes hyperparameters for an artificial neural network (ANN) model during the training stage, significantly enhancing prediction accuracy for cardiovascular disease (CVD) risk. The proposed TPTM-HANN-GA framework requires far fewer experiments than a traditional grid search, making it highly suitable for application in resource-constrained, low-power computers, and edge artificial intelligence (edge AI) devices. Furthermore, the proposed TPTM-HANN-GA framework successfully identified the optimal configurations for the ANN model's hyperparameters, resulting in a hidden layer of 4 nodes, a tanh activation function, an SGD optimizer, a learning rate of 0.23425849, a momentum rate of 0.75462782, and seven hidden nodes. This optimized ANN model achieves 74.25% accuracy in predicting the risk of cardiovascular disease, which exceeds the existing state-of-the-art GA-ANN and TSTO-ANN models. The proposed TPTM-HANN-GA framework enables personalized CVD prediction to be efficiently conducted on low-power computers and edge-AI devices, achieving the goal of point-of-care testing (POCT) and empowering individuals to manage their heart health effectively. [ABSTRACT FROM AUTHOR]

In this paper, an improved modular multilevel converter (MMC) current controller is proposed for grid current harmonic mitigation in solid-state transformers (SSTs), regardless of the non-linear or unbalanced load positions at the SST stages. The proposed MMC current controller is achieved using three control strategies based on the harmonic order and the harmonic sequence: a low-order harmonic compensator, a negative-sequence harmonic compensator for unbalanced control, and a high-order harmonic compensator based on an artificial neural network (ANN) controller, trained offline using a fuzzy logic (FL) controller. The use of such non-linear controllers for both training and control ensures an active filtering feature for the MMC controller. This makes the proposed solution a good alternative to solutions based on extra filters or an increased switching frequency, which inevitably increases the system costs and losses. The proposed control strategy is implemented and evaluated in MATLAB/Simulink software under various load conditions and parameter changes, and results from multiple simulations are presented. [ABSTRACT FROM AUTHOR]

This paper proposes an encoder measurement error compensation method using an adaptive linear neuron, a type of artificial neural network, for use in the vector control of two-phase stepping motors. Stepping motors can have an asymmetric structure due to an eccentricity that occurs in the manufacturing and assembly process. When performing vector control with a flux angle measured by an incremental encoder attached to stepping motors with eccentricity, the following two kinds of encoder measurement errors occur. First, an offset position error occurs during the forced excitation process for the initial rotor position alignment. Second, sinusoidal position and speed errors with a frequency equal to the mechanical speed occur. In this study, how the eccentricity causes offset and sinusoidal measurement errors in addition to the phenomena caused by these encoder measurement errors are analyzed. From these analyses, an encoder error compensation method based on an adaptive linear neuron is proposed. The validity of the proposed encoder errors compensation methods is verified through experiments on a two-phase stepping motor drive system. [ABSTRACT FROM AUTHOR]

The aim of this study is to predict the total ergodic capacity of near users in a cognitive radio (CR)-based non-orthogonal multiple access (NOMA) system model using the proposed artificial neural network (ANN) architecture. The input dataset used in this study was collected from the CR-NOMA system model and consists of the path loss coefficient, power allocation coefficient, signal-to-noise ratio, the distance between the source-relay-destination, and the ratio of the power of the secondary user to that of the primary user. Using a supervised learning method, the output data are trained and input into the ANN to estimate the ergodic capacity of nearby users using test data. The trained system model demonstrates an accuracy of 96.43% for training data, 96.34% for validation data, and 95.66% for test data when estimating the total ergodic capacity. [ABSTRACT FROM AUTHOR]

Entropy generation analysis combined with hybrid nanofluid flow principles contribute to the development of more efficient drug delivery systems for cancer treatment. By using nanofluids, it is possible to improve the transport and targeted release of therapeutic agents to specific sites in the body, allowing for better control and efficiency in treating cancerous cells. Keeping these important applications in view, in the current analysis, the production of irreversibility and stagnant point hybrid nanofluid flow has been considered on a Riga plate. The impacts of nonlinear convection and solar radiations have also been used in this study. Glycol (C3H8O2) is taken as base fluid, while nanoparticles of copper (Cu) and aluminum oxide (Al2O3) have been mixed in it to obtain a hybrid nanofluid. The leading equations for the study have converted to dimensionless form by employing a set of suitable variables and then have been solved by using an artificial neural network (ANN). In order to evaluate the effectiveness of the least mean square neural network algorithm (LMS-NNA), statistical neural network techniques are employed, encompassing error analysis and curve-fitting graphs. It has been revealed in this work that an upsurge in EMHD Riga plate factor and the Grashof number escalates the velocity distribution for both nanoparticles as well as hybrid nanoparticles and is opposed by augmentation in width factor for electrode/magnet. The increase in the nanoparticle volume fraction from 0.01 to 0.05 escalates the heat transfer rate up to 7.6% in the case of nanofluid with Cu-nanoparticles while this increase is 9.3% using hybrid nanofluid Cu + Al2O3. These results show that HNF are more efficient in improving the HT rate. [ABSTRACT FROM AUTHOR]

This study addresses COVID-19 testing as a nonlinear sampling problem, aiming to uncover the dependence of the true infection count in the population on COVID-19 testing metrics such as testing volume and positivity rates. Employing an artificial neural network, we explore the relationship among daily confirmed case counts, testing data, population statistics, and the actual daily case count. The trained artificial neural network undergoes testing in in-sample, out-of-sample, and several hypothetical scenarios. A substantial focus of this paper lies in the estimation of the daily true case count, which serves as the output set of our training process. To achieve this, we implement a regularized backcasting technique that utilize death counts and the infection fatality ratio (IFR), as the death statistics and serological surveys (providing the IFR) as more reliable COVID-19 data sources. Addressing the impact of factors such as age distribution, vaccination, and emerging variants on the IFR time series is a pivotal aspect of our analysis. We expect our study to enhance our understanding of the genuine implications of the COVID-19 pandemic, subsequently benefiting mitigation strategies. [ABSTRACT FROM AUTHOR]

In this COVID-19 pandemic era, where people are losing their lives, there are several species and plants available on our Mother Earth that are beneficial to boosting the human immune system and sustaining their life. These plant leaves and trunks can also be used to effectively treat a variety of diseases in humans. The images of plants, as well as the identification of the leaf through artificial intelligence, are critical for obtaining such benefits. The proposed system automatically grades the various species and classify images of plant leaves into different families. The leaf image parameter is used to extract these properties such as colour, texture, shape, and so on. The proposed system makes use of colour and texture to form characteristics. The colour pattern for the texture uses GLCM and Shape extraction forms to extract colour information such as HSV (hue, saturation, value). The ANN (artificial neural network) algorithm is used to classify leaf images. Colour extraction, texture, and shape features, both alone and in combination, are used for classification. Using combined features yields better results than using single features. In the proposed system data set, 285 images are captured, 210 images with ANN are trained, and 75 images are used for the test set. With ANN, the system achieves an approximate 93.33% accuracy. The results were also validated with SVM (support vector machine), which provides an approximate 48% accuracy, indicating that ANN outperforms SVM. [ABSTRACT FROM AUTHOR]

• Aging of FeS nanoparticles seriously affects arsenic removal but slight on copper. • Controllable FeS are the targeted sulfurizing agents in different acidity wastewater. • BP-GA model are developed to precisely predict copper and arsenic removal with R of 0.9923. • Temperature and FeS aging time are necessary to gain high separation accuracy. • The model provides an essential reference in selecting sulfurizing reagent types. Iron sulfide (FeS) is a promising material for separating copper and arsenic from strongly acidic wastewater due to its S2− slow-release effect. However, uncertainties arise because of the constant changes in wastewater composition, affecting the selection of operating parameters and FeS types. In this study, the aging method was first used to prepare various controllable FeS nanoparticles to weaken the arsenic removal ability without affecting the copper removal. Orthogonal experiments were conducted, and the results identified the Cu/As ratio, H 2 SO 4 concentration, and FeS dosage as the three main factors influencing the separation efficiency. The backpropagation artificial neural network (BP-ANN) model was established to determine the relationship between the influencing factors and the separation efficiency. The correlation coefficient (R) of overall model was 0.9923 after optimizing using genetic algorithm (GA). The BP-GA model was also solved using GA under specific constraints, predicting the best solution for the separation process in real-time. The predicted results show that the high temperature and long aging time of FeS were necessary to gain high separation efficiency, and the maximum separation factor can reached 1,400. This study provides a suitable sulfurizing material and a set of methods and models with robust flexibility that can successfully predict the separation efficiency of copper and arsenic from highly acidic environments. [Display omitted] [ABSTRACT FROM AUTHOR]

Traditional rule-based and statistical methods have limitations when dealing with complex language structures and semantics. In neural network machine translation algorithms, the objective function is usually to improve the accuracy of n-ary words. However, this does not guarantee a more natural and accurate translation. To overcome these challenges, this paper proposes an optimization algorithm for English natural translation processing based on neural networks, which combines Generative Adversarial Network (GAN) and Transformer models. In GAN, the generative model uses the Transformer model to generate false samples, while the discriminative model uses a binary classifier based on convolutional neural networks and attention mechanisms to distinguish between true and false samples. During the training process, reinforcement learning algorithms are added to evaluate and adjust the generated sentences, and the parameters of the generated model are updated. The classification results of the discriminative model are used together with the Bilingual Evaluation Basis Value (BLEU) objective function to evaluate false samples, and the results are fed back to the generating model to guide parameter updates and optimization. Extensive experiments were conducted on a standard English-Chinese machine translation dataset to evaluate our method. Compared with the benchmark model that only uses supervised learning methods, our neural network-based optimization algorithm for English natural translation processing has achieved significant improvements in translation quality. According to statistical comparison, compared with the Transformer model (BLUE = 33.63 and AP = 90%) and the deep learning model based on long-term and short-term memory (BLUE = 30.26 and AP = 83%), the GAN and Transformer models proposed as the best framework exhibit better performance in bilingual evaluation deficiency (BLEU) (34.35) and accuracy (AP = 95%). [ABSTRACT FROM AUTHOR]

AbstractAppointment scheduling for hospital patients is becoming increasingly important because it can reduce patient’s waiting time and enhance hospitals’ medical quality. In the case of ultrasound examinations, multiple radiological technologists use multiple rooms to conduct examinations, which involves the sequencing of examination work, time uncertainty, and operational characteristics. To enable hospital managers to effectively evaluate radiological technologists’ effort when carrying out ultrasound examinations, this study uses classification methods in data exploration, including decision tree (DT), logistic regression (LR), and artificial neural network (ANN), to establish a time prediction model for outpatient ultrasound examination. This study found that the DT model (i.e. C4.5) had the highest classification accuracy. After reviewing and removing rules generated from the model with an accuracy below 70%, the overall accuracy of the model was 76.31%. The classification results of the model are a reference for outpatient ultrasound appointment scheduling. The results of the study can effectively predict the examination time needed for ultrasound outpatients requiring different examination positions, thereby fairly allocating the patients to each room for the examination and reducing patient waiting time and radiological technologists’ idle time. Therefore, hospital managers can improve the hospital’s overall medical quality. [ABSTRACT FROM AUTHOR]

In the wake of the contemporary competitive business landscape, the retention of employees has become one of the most important yet difficult tasks for any corporate. Retaining top-performing employees not only improves organizational performance but also reduces recruitment costs. In this study, the authors investigate the major drivers leading to employee attrition and using machine learning algorithms implemented on a well proven and validated IBM HR data set. Although the data set tags the samples for a target variable (attrited and non-attrited), the work presented in this paper comes up with another labelling (1. likely to leave, 2. On the verge of leaving, 3. will stay). The data set is evaluated over top 10 Machine learning algorithms and a competitive analysis is made between them based on various factors. The best model has shown a prediction accuracy of over 85% +. Managers are provided with insights and recommendations at the end that will help companies to proactively identify at-risk employees and implement effective retention strategies. [ABSTRACT FROM AUTHOR]

SQL injection is one of the important security issues in web applications because it allows an attacker to interact with the application's database. SQL injection attacks can be detected using machine learning algorithms. The effective features should be employed in the training stage to develop an optimal classifier with optimal accuracy. Identifying the most effective features is an NP-complete combinatorial optimization problem. Feature selection is the process of selecting the training dataset's smallest and most effective features. The main objective of this study is to enhance the accuracy, precision, and sensitivity of the SQLi detection method. In this study, an effective method to detect SQL injection attacks has been proposed. In the first stage, a specific training dataset consisting of 13 features was prepared. In the second stage, two different binary versions of the Gray-Wolf algorithm were developed to select the most effective features of the dataset. The created optimal datasets were used by different machine learning algorithms. Creating a new SQLi training dataset with 13 numeric features, developing two different binary versions of the gray wolf optimizer to optimally select the features of the dataset, and creating an effective and efficient classifier to detect SQLi attacks are the main contributions of this study. The results of the conducted tests indicate that the proposed SQL injection detector obtain 99.68% accuracy, 99.40% precision, and 98.72% sensitivity. The proposed method increases the efficiency of attack detection methods by selecting 20% of the most effective features. [ABSTRACT FROM AUTHOR]

Previous research on online health knowledge payment has mainly focused on the health consultation context and knowledge contributors-related characteristics. However, less attention has been given to the roles of health knowledge itself and the underlying platforms. Based on the perceived value theory and S-O-R framework, this study investigated how health knowledge and platform characteristics influence consumers’ purchase intention. Through a survey of 432 Chinese respondents, we employed a multi-analytical approach, combining structural equation modelling (SEM) with artificial neural network (ANN) techniques. SEM results indicate that knowledge personalisation, platform information quality, and platform service quality positively influence both utilitarian and hedonic values, subsequently affecting purchase intention. Furthermore, our findings support the partial mediation role of perceived value. ANN results highlight platform information quality as the most influential stimulus factor. Our study broadens the research scope of online health knowledge payment and offers theoretical and practical insights for its further advancement. [ABSTRACT FROM AUTHOR]

Background: Paclitaxel is commonly used as a second-line therapy for advanced gastric cancer (AGC). The decision to proceed with second-line chemotherapy and select an appropriate regimen is critical for vulnerable patients with AGC progressing after first-line chemotherapy. However, no predictive biomarkers exist to identify patients with AGC who would benefit from paclitaxel-based chemotherapy. Methods: This study included 288 patients with AGC receiving second-line paclitaxel-based chemotherapy between 2017 and 2022 as part of the K-MASTER project, a nationwide government-funded precision medicine initiative. The data included clinical (age [young-onset vs. others], sex, histology [intestinal vs. diffuse type], prior trastuzumab use, duration of first-line chemotherapy), and genomic factors (pathogenic or likely pathogenic variants). Data were randomly divided into training and validation sets (0.8:0.2). Four machine learning (ML) methods, namely random forest (RF), logistic regression (LR), artificial neural network (ANN), and ANN with genetic embedding (ANN with GE), were used to develop the prediction model and validated in the validation sets. Results: The median patient age was 64 years (range 25–91), and 65.6% of those were male. A total of 288 patients were divided into the training (n = 230) and validation (n = 58) sets. No significant differences existed in baseline characteristics between the training and validation sets. In the training set, the areas under the ROC curves (AUROC) for predicting better progression-free survival (PFS) with paclitaxel-based chemotherapy were 0.499, 0.679, 0.618, and 0.732 in the RF, LR, ANN, and ANN with GE models, respectively. The ANN with the GE model that achieved the highest AUROC recorded accuracy, sensitivity, specificity, and F1-score performance of 0.458, 0.912, 0.724, and 0.579, respectively. In the validation set, the ANN with GE model predicted that paclitaxel-sensitive patients had significantly longer PFS (median PFS 7.59 vs. 2.07 months, P = 0.020) and overall survival (OS) (median OS 14.70 vs. 7.50 months, P = 0.008). The LR model predicted that paclitaxel-sensitive patients showed a trend for longer PFS (median PFS 6.48 vs. 2.33 months, P = 0.078) and OS (median OS 12.20 vs. 8.61 months, P = 0.099). Conclusions: These ML models, integrated with clinical and genomic factors, offer the possibility to help identify patients with AGC who may benefit from paclitaxel chemotherapy. [ABSTRACT FROM AUTHOR]

Forecasting using artificial neural networks (ANNs) has previously been demonstrated to provide skilful monthly and seasonal rainfall forecasts for many parts of the world, including Australia. However, few rainfall forecasting studies have been reported for lead times beyond 12 months, which can be very useful for flood and drought management, management of dam infrastructure for hydroelectricity generation and for agriculture This investigation examines monthly rainfall forecasts using ANNs for Gympie, Lismore and Brisbane in eastern Australia, locations that have experienced severe flooding associated with very heavy rainfall during the past two decades, as well as periods of drought. These locations have long continuous rainfall records extending back more than 100 years. Monthly rainfall forecasts were generated for these locations for each month using ANNs provided by Neurosolutions Infinity. Input variables comprised both local values of rainfall and temperatures and large-scale climate indices including the Southern Oscillation Index and the Inter-Decadal Pacific Oscillation. The study shows that lead times using ANNs can be extended without significant loss in forecast skill. Average monthly rainfall forecast skill with lead times of 12, 36, 60 and 84 months for the three sites lay in the range 34.8–54.1% relative to climatology. This compares very favourably with reported skill using methods based on physical climate models where skill is often about equivalent to climatology (0%). At longer lead times between 108 and 156 months the skill scores using ANNs decline into the range 10.8–26.6%. The application of ANN forecasting can be very valuable in providing the public with much better warnings of flood and drought conditions than currently given by the Australian Bureau of Meteorology, allowing sufficient lead time to take appropriate action. [ABSTRACT FROM AUTHOR]