Regularizing the effect of input noise injection in feedforward neural networks training

Injecting input noise during feedforward neural network (NN) training can improve generalization performance markedly. Reported works justify this fact arguing that noise injection is equivalent to a smoothing regularization with the input noise variance playing the role of the regularization parameter. The success of this approach depends on the appropriate choice of the input noise variance. However, it is often not known a priori if the degree of smoothness imposed on the FNN mapping is consistent with the unknown function to be approximated. In order to have a better control over this smoothing effect, a loss function putting in balance the smoothed fitting induced by the noise injection and the precision of approximation, is proposed. The second term, which aims at penalizing the undesirable effect of input noise injection or controlling the deviation of the random perturbed loss, was obtained by expressing a certain distance between the original loss function and its random perturbed version. In fact, this term can be derived in general for parametrical models that satisfy the Lipschitz property. An example is included to illustrate the effectiveness of learning with this proposed loss function when noise injection is used.     

Antioxidant and antithrombotic effects of green mussels (Perna canaliculus) in rats

In the past decade, the use of marine mussels as seafood is being more popular. They considered being a rich source of various nutritional bioactive compounds that aroused the scientific community's interest. This study investigated the antioxidant and the antithrombotic consequences on Sprague-Dawley male rats after adding dried New Zealand mussel Perna canaliculus in their diet. The biochemical, hematological and histopathological changes were also observed. Forty rats were divided into four groups according to the amount of dried mussels in their diet, in addition to a control group that consumed a basal diet only. Group 1 consumed 25% dried mussels in its basal diet; Group 2 consumed 35% dried mussels in its basal diet, and Group 3 was consumed 45% dried mussels in its basal diet. The biochemical parameters showed improvements in liver function. Interestingly, the lipid profile decreased, especially the low-density lipoprotein cholesterol (LDL-C) levels which were reduced significantly in Group 3 (p < .01). These observations were accompanied by a decrease in iron levels significantly as the amount of dried mussels increased (p < .01). Furthermore, the noticed changes in the hematological profile prove that there is an increase in antithrombotic activity. Dried mussels had potent antioxidant effects in the liver as shown by increased lipid peroxide (p < .05), reduced glutathione (p < .05), and glutathione peroxidase (GSH-Px; p < .05). Additionally, antioxidant activity in the kidney was shown to increase through GSH-Px activity (p < .01). In conclusion, these results indicate that consuming dried mussels resulted in improved biochemical and antioxidants activities and could be used as an antithrombotic agent.     

Development and Characterization of Gentamicin-Loaded Arabinoxylan-Sodium Alginate Films as Antibacterial Wound Dressing

Biopolymer-based antibacterial films are attractive materials for wound dressing application because they possess chemical, mechanical, exudate absorption, drug delivery, antibacterial, and biocompatible properties required to support wound healing. Herein, we fabricated and characterized films composed of arabinoxylan (AX) and sodium alginate (SA) loaded with gentamicin sulfate (GS) for application as a wound dressing. The FTIR, XRD, and thermal analyses show that AX, SA, and GS interacted through hydrogen bonding and were thermally stable. The AXSA film displays desirable wound dressing characteristics: transparency, uniform thickness, smooth surface morphology, tensile strength similar to human skin, mild water/exudate uptake capacity, water transmission rate suitable for wound dressing, and excellent cytocompatibility. In Franz diffusion release studies, >80% GS was released from AXSA films in two phases in 24 h following the Fickian diffusion mechanism. In disk diffusion assay, the AXSA films demonstrated excellent antibacterial effect against E.coli, S. aureus, and P. aeruginosa. Overall, the findings suggest that GS-loaded AXSA films hold potential for further development as antibacterial wound dressing material.     

Upper bound for algebraic immunity on a subclass of Maiorana McFarland class of bent functions

Studying algebraic immunity of Boolean functions is recently a very important research topic in cryptography. It is recently proved by Courtois and Meier that for any Boolean function of n-variable the maximum algebraic immunity is . We found a large subclass of Maiorana McFarland bent functions on n-variable with a proven low level of algebraic immunity . To the best of our knowledge we provide for the first time a new upper bound for algebraic immunity for a nontrivial class of Boolean functions. We also discuss that this result has some fascinating implications.     

Green synthesis of silver nanoparticles using bovine skin gelatin and its antibacterial effect on clinical bacterial isolates

Nanoparticles are being increasingly used in day‐to‐day life. Therefore, concerns have been raised regarding their interactions with the surrounding environment. This study focused on a simple green method for synthesizing silver nanoparticles (Ag‐NPs) in an autoclave at 15 psi (103 kPa) and 121°C. An aqueous solution of AgNO₃ as a precursor of Ag‐NPs and gelatin (type B) reducing and/or stabilizing (capping) agent were used. The effect of various AgNO₃ concentrations of certain gelatin concentration and various gelatin concentrations at constant AgNO₃ concentration, and autoclaving time, was studied. UV‐Vis spectra ascribed that the presence of localized surface plasmon resonance (SPR) of the synthesized Ag‐NPs. TEM images and the selected area of electron diffraction confirmed, the formation of Ag‐NPs with a diameter of approximately 5 ±0.35 nm. Furthermore, FT‐IR revealed that a gelatin polymer matrix stabilized the synthesized Ag‐NPs. The Well diffusion assay was used to test the effect of Ag‐NPs on six clinical bacterial isolates, where Gram positive bacteria were more susceptible to Ag‐NPs than Gram negative bacteria. Therefore, Ag‐NPs capped by gelatin have remarkable potential effect as an antibacterial agent, and they not only have various medical applications but can also be used in biological, pharmaceutical and industrial fields.Inspec keywords: silver, nanoparticles, nanomedicine, antibacterial activity, microorganisms, nanofabrication, skin, gelatin, ultraviolet spectra, visible spectra, surface plasmon resonance, transmission electron microscopy, electron diffraction, Fourier transform infrared spectra, polymers, biomedical materialsOther keywords: green synthesis, silver nanoparticles, bovine skin gelatin, antibacterial effect, clinical bacterial isolates, autoclave, reducing agent, stabilising agent, ultraviolet‐visible spectra, localised surface plasmon resonance, transmissions electron microscope images, electron diffraction, Fourier transform infrared spectroscopy, gelatin polymer matrix, well diffusion assay, gram negative bacteria, Escherichia coli, Pseudomonas aeruginosa, Klebsiella pneumoniae, gram positive bacteria, Bacillus megaterium, Streptococcus pyogenes, Staphylococcus aureus, temperature 121 degC, Ag     

Adsorption of Methyl Chloride on Molecular Sieves,Silica Gels,and Activated Carbon

Adsorption of methyl chloride (CH₃Cl or MeCl) on five different types of adsorbents was investigated experimentally at increasing pressures and room temperature. Prior to adsorption, all adsorbents were analyzed to assess their physical and chemical characteristics. The experimental data was then used to determine the adsorption isotherms, heats of adsorption, adsorption rates, and their respective theoretical models. The MeCl adsorption capacity was found to reasonably correlate with the adsorbent's surface area. The MeCl adsorption isotherm and adsorption rates were fitted for the first time to a Freundlich isotherm model and pseudo first-/second-order kinetic models, respectively. The range of heat of adsorption indicated a physisorption type of bonding; hence, the investigated adsorbents can potentially be regenerated for cyclic adsorption.     

Thermal radiation and variable electrical conductivity effects on MHD peristaltic motion of Carreau nanofluids: Radiotherapy and thermotherapy of oncology treatment

This study intends to investigate the influences of thermal radiation and variable electrical conductivity on the MHD peristaltic flow of Carreau nanofluids as the radiotherapy and thermotherapy are required for cancer treatment. Formulation of temperature‐dependent electrical conductivity is introduced for the first time in the peristaltic literature. The related equations of momentum, mass, and concentration are reformulated using lubrication approximations (ie, tiny or zero Reynolds number and long wavelength). These simplified equations are solved numerically with the aid of Parameteric‐NDSolve. Results for velocity, temperature, and concentration distributions are obtained in three‐dimensional analytical forms. The streamline graphs are offered in the terminus, which elucidate the trapping bolus phenomenon. A “special case” of our results offered to get the solutions over certain contours for the behaviors of velocity, temperature, and nanoparticle concentration. It is found that the magnetic nanoparticles acquire more energy at high temperature, enabling them to destroy and damage tumors tissues (thermotherapy of oncology). Radiation is the reason for spreading the energy, that is, the energy of all the fluid nanoparticles does not continue with the same value. Therefore, in cancer treatment, doctors use high doses of radiation to cure cancer cells and prevent it from returning (radiotherapy of oncology).     

Intelligent Model for Constructed Facilities Surface Assessment

Computerized intelligent systems can simulate human expertise as well as analyze and process vast amounts of data instantaneously. This paper presents a hybrid intelligent computerized model for constructed facilities surface quality assessment. The model uses computers to analyze digital images of the areas to be assessed to identify and measure defects. Moreover, neural networks are used to train the system to automate the process and replicate the experts' knowledge in identifying the defects. Most techniques, currently used in construction quality assessment, rely mostly on subjective criteria. The model applies digital image processing and neural network techniques for constructed facilities surface quality assessment to make the process objective, quantitative, consistent, and reliable. Highway steel bridge coating assessment was used to exemplify the generic model.     

Time scale analysis for fluidized bed melt granulation III: Binder solidification rate

In series I and II of this study ( [Chua et al., 2010a] and [Chua et al., 2010b] ), we discussed the time scale of granule–granule collision, droplet–granule collision and droplet spreading in Fluidized Bed Melt Granulation (FBMG). In this third one, we consider the rate at which binder solidifies. Simple analytical solution, based on classical formulation for conduction across a semi-infinite slab, was used to obtain a generalized equation for binder solidification time. A multi-physics simulation package (Comsol) was used to predict the binder solidification time for various operating conditions usually considered in FBMG. The simulation results were validated with experimental temperature data obtained with a high speed infrared camera during solidification of ‘macroscopic’ (mm scale) droplets. For the range of microscopic droplet size and operating conditions considered for a FBMG process, the binder solidification time was found to fall approximately between 10⁻³ and 10⁻¹ s. This is the slowest compared to the other three major FBMG microscopic events discussed in this series (granule–granule collision, granule–droplet collision and droplet spreading).