Design deep neural network architecture using a genetic algorithm for estimation of pile bearing capacity.

Determination of pile bearing capacity is essential in pile foundation design. This study focused on the use of evolutionary algorithms to optimize Deep Learning Neural Network (DLNN) algorithm to predict the bearing capacity of driven pile. For this purpose, a Genetic Algorithm (GA) was developed to select the most significant features in the raw dataset. After that, a GA-DLNN hybrid model was developed to select optimal parameters for the DLNN model, including: network algorithm, activation function for hidden neurons, number of hidden layers, and the number of neurons in each hidden layer. A database containing 472 driven pile static load test reports was used. The dataset was divided into three parts, namely the training set (60%), validation (20%) and testing set (20%) for the construction, validation and testing phases of the proposed model, respectively. Various quality assessment criteria, namely the coefficient of determination (R2), Index of Agreement (IA), mean absolute error (MAE) and root mean squared error (RMSE), were used to evaluate the performance of the machine learning (ML) algorithms. The GA-DLNN hybrid model was shown to exhibit the ability to find the most optimal set of parameters for the prediction process.The results showed that the performance of the hybrid model using only the most critical features gave the highest accuracy, compared with those obtained by the hybrid model using all input variables.

Investigation and Optimization of the C-ANN Structure in Predicting the Compressive Strength of Foamed Concrete.

Development of Foamed Concrete (FC) and incessant increases in fabrication technology have paved the way for many promising civil engineering applications. Nevertheless, the design of FC requires a large number of experiments to determine the appropriate Compressive Strength (CS). Employment of machine learning algorithms to take advantage of the existing experiments database has been attempted, but model performance can still be improved. In this study, the performance of an Artificial Neural Network (ANN) was fully analyzed to predict the 28 days CS of FC. Monte Carlo simulations (MCS) were used to statistically analyze the convergence of the modeled results under the effect of random sampling strategies and the network structures selected. Various statistical measures such as Coefficient of Determination (R2), Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE) were used for validation of model performance. The results show that ANN is a highly efficient predictor of the CS of FC, achieving a maximum R2 value of 0.976 on the training part and an R2 of 0.972 on the testing part, using the optimized C-ANN-[3,4,5,1] structure, which compares with previous published studies. In addition, a sensitivity analysis using Partial Dependence Plots (PDP) over 1000 MCS was also performed to interpret the relationship between the input parameters and 28 days CS of FC. Dry density was found as the variable with the highest impact to predict the CS of FC. The results presented could facilitate and enhance the use of C-ANN in other civil engineering-related problems.

Characterization of serum antibodies from women immunized with Gardasil: A study of HPV-18 infection of primary human keratinocytes.

The prevalent human papillomaviruses (HPVs) infect human epithelial tissues. Infections by the mucosotropic HPV genotypes cause hyperproliferative ano-genital lesions. Persistent infections by high-risk (HR) HPVs such as HPV-16, HPV-18 and related types can progress to high grade intraepithelial neoplasias and cancers. Prophylactic HPV vaccines are based on DNA-free virus-like particles (VLPs) composed of the major capsid protein L1 of HPV-16, -18, -6 and -11 (Gardasil) or HPV-16 and -18 (Cervarix). Sera from vaccinated animals effectively prevent HPV pseudovirions to infect cell lines and mouse cervical epithelia. Both vaccines have proven to be highly protective in people. HPV pseudovirions are assembled in HEK293TT cells from matched L1 and L2 capsid proteins to encapsidate a reporter gene. Pseudovirions and genuine virions have structural differences and they infect cell lines or primary human keratinocytes (PHKs) with different efficiencies. In this study, we show that sera and isolated IgG from women immunized with Gardasil prevent authentic HPV-18 virions from infecting PHKs, whereas non-immune sera and purified IgG thereof are uniformly ineffective. Using early passage PHKs, neutralization is achieved only if immune sera are added within 2-4h of infection. We attribute the timing effect to a conformational change in HPV virions, thought to occur upon initial binding to heparan sulfate proteoglycans (HSPG) on the cell surface. This interpretation is consistent with the inability of immune IgG bound to or taken up by PHKs to neutralize the virus. Interestingly, the window of neutralization increases to 12-16h in slow growing, late passage PHKs, suggestive of altered cell surface molecules. In vivo, this window might be further lengthened by the time required to activate the normally quiescent basal cells to become susceptible to infection. Our observations help explain the high efficacy of HPV vaccines.

Klebsiella pneumoniae oropharyngeal carriage in rural and urban Vietnam and the effect of alcohol consumption.

INTRODUCTION: Community acquired K. pneumoniae pneumonia is still common in Asia and is reportedly associated with alcohol use. Oropharyngeal carriage of K. pneumoniae could potentially play a role in the pathogenesis of K. pneumoniae pneumonia. However, little is known regarding K. pneumoniae oropharyngeal carriage rates and risk factors. This population-based cross-sectional study explores the association of a variety of demographic and socioeconomic factors, as well as alcohol consumption with oropharyngeal carriage of K. pneumoniae in Vietnam. METHODS AND FINDINGS: 1029 subjects were selected randomly from age, sex, and urban and rural strata. An additional 613 adult men from a rural environment were recruited and analyzed separately to determine the effects of alcohol consumption. Demographic, socioeconomic, and oropharyngeal carriage data was acquired for each subject. The overall carriage rate of K. pneumoniae was 14.1% (145/1029, 95% CI 12.0%-16.2%). By stepwise logistic regression, K. pneumoniae carriage was found to be independently associated with age (OR 1.03, 95% CI 1.02-1.04), smoking (OR 1.9, 95% CI 1.3-2.9), rural living location (OR 1.6, 95% CI 1.1-2.4), and level of weekly alcohol consumption (OR 1.7, 95% CI 1.04-2.8). CONCLUSION: Moderate to heavy weekly alcohol consumption, old age, smoking, and living in a rural location are all found to be associated with an increased risk of K. pneumoniae carriage in Vietnamese communities. Whether K. pneumoniae carriage is a risk factor for pneumonia needs to be elucidated.

Oral and nasal sensitization promote distinct immune responses and lung reactivity in a mouse model of peanut allergy.

Despite structural and functional differences between the initial sites of contact with allergens in the gastrointestinal and nasal tracts, few animal models have examined the influence of the mucosal routes of sensitization on host reactivity to food or environmental antigens. We compared the oral and nasal routes of peanut sensitization for the development of a mouse model of allergy. Mice were sensitized by administration of peanut proteins in the presence of cholera toxin as adjuvant. Antibody and cytokine responses were characterized, as well as airway reactivity to nasal challenge with peanut or unrelated antigens. Oral sensitization promoted higher levels of IgE, but lower IgG responses, than nasal sensitization. Both orally and nasally sensitized mice experienced airway hyperreactivity on nasal peanut challenge. The peanut challenge also induced lung eosinophilia and type 2 helper T-cell-type cytokines in orally sensitized mice. In contrast, peanut challenge in nasally sensitized mice promoted neutrophilia and higher levels of lung MAC-1(+) I-A(b low) cells and inflammatory cytokines. In addition, nasal but not oral, sensitization promoted lung inflammatory responses to unrelated antigens. In summary, both oral and nasal peanut sensitization prime mice for airway hyperreactivity, but the initial mucosal route of sensitization influences the nature of lung inflammatory responses to peanut and unrelated allergens.