Discoloration process modeling by neural network

The photo-oxidation of acid orange 52 dye was performed in the presence of H₂O₂, utilizing UV light, aiming the discoloration process modeling and the process variable influence characterization. The discoloration process was modeled by the use of feedforward neural network. Each sample was characterized by five independent variables (dye concentration, pH, hydrogen peroxide volume, temperature and time of operation) and a dependent variable (absorbance). The neural model has also provided, through Garson Partition coefficients and the Pertubation method, the independent variable influence order determination. The results indicated that the time of operation was the predominant variable and reaction mean temperature was the lesser influent variable. The neural model obtained presented coefficients of correlation on the order 0.98, for sets of trainability, validation and testing, indicating the power of prediction of the model and its character of generalization.     

Nanorobotics control design: a collective behavior approach for medicine

The authors present a new approach using genetic algorithms, neural networks, and nanorobotics concepts applied to the problem of control design for nanoassembly automation and its application in medicine. As a practical approach to validate the proposed design, we have elaborated and simulated a virtual environment focused on control automation for nanorobotics teams that exhibit collective behavior. This collective behavior is a suitable way to perform a large range of tasks and positional assembly manipulation in a complex three-dimensional workspace. We emphasize the application of such techniques as a feasible approach for the investigation of nanorobotics system design in nanomedicine. Theoretical and practical analyses of control modeling is one important aspect that will enable rapid development in the emerging field of nanotechnology.     

Effect of nanovesicle-encapsulated nisin on growth of Listeria monocytogenes in milk

Commercial nisin was encapsulated in nanovesicles (mean diameter 140 nm) prepared from partially purified soy lecithin. Nisin-loaded liposomes and unencapsulated (free) nisin were initially tested in BHI medium and skim milk inoculated with Listeria monocytogenes and incubated for 48 h at 30 °C. At such abuse temperature conditions, free nisin showed better inhibitory than the liposomal counterparts. Subsequently, the effect of encapsulated or free nisin was evaluated in combination with refrigeration (7 ± 1 °C) in both whole (3.25% fat) and skim (0% fat) milk for up to 14 day. A decrease of 3–4 log cycles in L. monocytogenes counts was observed for free and encapsulated nisin at 0.5 mg/ml concentration. Liposome encapsulation of antimicrobial peptides may be important to overcome stability issues and interaction with food components. The utilization of nanovesicle-encapsulated nisin in combination with low temperatures appeared to be effective to control L. monocytogenes in milk, emphasizing the importance of hurdle technology to assure food safety.     

Fate of beta blockers and psycho-active drugs in conventional wastewater treatment

The removal of beta blockers and psycho-active drugs was investigated in a representative conventional German WWTP by long-term measurement campaigns along different biological treatment processes. The activated sludge treatment with an elevated SRT of 18 d was the only process which led to a significant removal of certain beta blockers and psycho-active drugs. The removal efficiency was below 60% for all compounds except for the natural opium alkaloids codeine and morphine being removed by more than 80%. Primary biological transformation and sorption onto sludge as the main removal mechanisms were examined in lab-scale batch experiments. Sorption onto activated sludge was found to be negligible (<3%). The biological transformation could be described by pseudo-first order kinetics and the transformation constants k_biol were used to predict the removal of beta blockers and psycho-active drugs in an activated sludge unit with a model. For most compounds the removal efficiencies measured on the full-scale WWTP were within the 95% confidence intervals predicted by the model. The results from full-scale measurements and modeling indicate that biological transformation in the nitrification tank together with parameters such as the sludge retention time and the temperature is crucial regarding the biological transformation of beta blockers and psycho-active drugs in conventional WWTPs.     

Microbiological hazards involved in fresh‐cut lettuce processing

BACKGROUND: The increasing consumption of produce all over the world has resulted in increasing concern by the regulatory agencies with respect to the level of safety performed by the processors. The objective of the present study was to investigate the hazards involved in the various steps of fresh‐cut lettuce processing (reception/selection of raw material, washing, rinsing, sanitisation and final product) by means of microbiological analyses of microbial groups used as indicators of hygienic conditions and of pathogens. RESULTS: High microbial loads of mesophilic and psychrotrophic bacteria and Pseudomonas spp. were found in the ram reception (～6 log colony‐forming units (CFU) g^?1), which were reduced by a single logarithmic cycle for the last two microbial groups after the sanitisation step (P < 0.05), the latter being ineffective against the first microbial group (P > 0.05). Lower counts of yeasts and moulds, total coliforms (35 °C) and faecal coliforms (44 °C) were observed in the initial step (3.49–4.53 log CFU g^?1, 0.65–1.55 log most probable number (MPN) g^?1 and 0.50–0.90 log MPN g^?1 respectively), these values increasing significantly after the sanitisation step for yeasts and moulds (～5 log CFU g^?1) but remaining unaltered for coliforms (P > 0.05). Salmonella spp. were not found in any of the experiments carried out, while the presence of Escherichia coli was observed in the final product. CONCLUSIONS : Practices compromising the hygienic quality of the final product during commercial storage were observed and corrective measures suggested. To the best of the authors' knowledge, these are the first data on microbiological safety in Brazilian fresh‐cut processing plants. Copyright © 2008 Society of Chemical Industry     

Brazilian Artisanal Cheeses: An Overview of their Characteristics,Main Types and Regulatory Aspects

A large variety of cheeses exist in Brazil, reflecting historical and cultural aspects. Brazilian artisanal cheeses present differences in the processing, ripening time (when applied), type of milk used, texture, size, shape, color, moisture content, flavor, the use or not of starter cultures, among others. This review describes the main artisanal cheeses produced in Brazil, focusing on general and particular characteristics associated with their making process and geographical identity. Overall, the high variability of the physicochemical data and deficiency of information on sensorial properties of Brazilian artisanal cheeses were noticed. On the other hand, culture‐dependent methods were mostly used to expand the knowledge into the microbiology of these cheeses, whereas their microbial diversity has been recently discovered through the use of 16S rRNA gene sequencing‐based methods. The certification of a geographical indication for Brazilian artisanal cheeses may encompass an essential milestone for adding value to these products. Regardless of their significance in the diet, culture, and economy of producing regions, taken together, the reviewed literature discloses the need of insightful studies to generate scientific data to support the expansion of the market, while ensuring the protection of historic aspects related to the production of Brazilian artisanal cheeses.     

A Smart Deep Convolutional Neural Network for Real-Time Surface Inspection

A proper detection and classification of defects in steel sheets in real time have become a requirement for manufacturing these products, largely used in many industrial sectors. However, computers used in the production line of small to medium size companies, in general, lack performance to attend real-time inspection with high processing demands. In this paper, a smart deep convolutional neural network for using in real-time surface inspection of steel rolling sheets is proposed. The architecture is based on the state-of-the-art SqueezeNet approach, which was originally developed for usage with autonomous vehicles. The main features of the proposed model are: small size and low computational burden. The model is 10 to 20 times smaller when compared to other networks designed for the same task, and more than 700 times smaller than general networks. Also, the number of floating-point operations for a prediction is about 50 times lower than the ones used for similar tasks. Despite its small size, the proposed model achieved near-perfect accuracy on a public dataset of 1800 images of six types of steel rolling defects.