Application of neural networks to simulate the growth profile of lactic acid bacteria in green olive fermentation

The growth profile of five strains of lactic acid bacteria (Lactobacillus plantarum ACA-DC 287, L. plantarum ACA-DC 146, Lactobacillus paracasei ACA-DC 4037, Lactobacillus sakei LQC 1378, and Leuconostoc mesenteroides LQC 1398) was investigated in controlled fermentation of cv. Conservolea green olives with a multilayer perceptron network, a combined logistic-Fermi function, and a two-term Gompertz function. Neural network training was based on the steepest-descent gradient learning algorithm. Model performance was compared with the experimental data with five statistical indices, namely coefficient of determination (R2), root mean square error (RMSE), mean relative percentage error (MRPE), mean absolute percentage error (MAPE), and standard error of prediction (SEP). The experimental data set consisted of 125 counts (CFU per milliliter) of lactic acid bacteria during the green olive fermentation process for up to 38 days (5 strains x 25 sampling days). For model development, a standard methodology was followed, dividing the data set into training (120 data) and validation (25 data) subsets. Our results demonstrated that the developed network was able to model the growth and survival profile of all the strains of lactic acid bacteria during fermentation equally well with the statistical models. The performance indices for the training subset of the multilayer perceptron network were R2 = 0.987, RMSE = 0.097, MRPE = 0.069, MAPE = 0.933, and SEP = 1.316. The relevant mean values for the logistic-Fermi and two-term Gompertz functions were R2 = 0.981 and 0.989, RMSE = 0.109 and 0.083, MRPE = 0.026 and 0.030, MAPE = 1.430 and 1.076, and SEP = 1.490 and 1.127, respectively. For the validation subset, the network also gave good predictions (R2 = 0.968, RMSE = 0.149, MRPE = 0.100, MAPE = 1.411, and SEP = 2.009).     

Survival of Bacillus cereus vegetative cells during Spanish-style fermentation of conservolea green olives

The aim of this research was to investigate the survival of artificially inoculated Bacillus cereus during Spanish-style green olive fermentation. Olives were initially treated with lye and subjected to different fermentation procedures including (i) heat shock (85 degrees C for 10 min) and inoculation with Lactobacillus plantarum ACA-DC 287, (ii) heat shock and inoculation with L. plantarum ACA-DC 146, (iii) heat shock and inoculation with Lactobacillus pentosus isolated during previous studies, and (iv) fermentation by indigenous flora (control process). Microbial growth and survival, pH, titratable acidity, and organic acid evolution were monitored. Inactivation of B. cereus was observed during all processes. The pathogen population declined during all fermentations, but a tailing effect was observed in the brines when the population reached 2 log CFU/ml, at which point the pathogen does not pose a risk to human health. The rate of inactivation was higher in heat-shocked inoculated olives (mean of -2.21 log CFU/day) compared with control olives (-1.26 log/day), indicating an advantage of heat shock and inoculation over spontaneous fermentation. The production of organic acids (primarily lactic acid) during fermentation seemed to be the main factor that determined the behavior of the pathogen under stress conditions prevailing in the brine. Principal components analysis was useful for distinguishing among the different fermentation processes on the basis of the relevant organic acid profile.     

Multiresolution Hierarchical Path-Planning for Small UAVs Using Wavelet Decompositions

We present an algorithm for solving the shortest (collision-free) path planning problem for an agent (e.g., a small UAV) with limited on-board computational resources. The agent has detailed knowledge of the environment and the obstacles only in the vicinity of its current position. Far away obstacles are only partially known and may even change dynamically. The algorithm makes use of the wavelet transform to construct an approximation of the environment at different levels of resolution. We associate with this multiresolution representation of the environment a graph, whose dimension can be made commensurate to the on-board computational resources of the agent. The adjacency list of the graph can be efficiently constructed directly from the approximation and detail wavelet coefficients, thus further speeding up the whole process. Simulations are presented to test the efficiency of the algorithm using non-trivial scenarios.     

Modelling the effect of temperature and water activity on the growth rate and growth/no growth interface of Byssochlamys fulva and Byssochlamys nivea

The purpose of the present study was to apply a modelling approach to define the growth rate and growth/no growth interface of Byssochlamys fulva and Byssochlamys nivea on a synthetic medium as a function of temperature and water activity. Both fungal species were grown on malt extract agar at different temperatures (10, 15, 20, 25, 30, 35, 40 and 45 °C) and a_w levels (0.88, 0.90, 0.92, 0.94, 0.96 and 0.99) for a period of 30 days. Growth responses were evaluated over time in terms of colony diameter changes. Growth data were fitted to the primary model of Baranyi and the resulting growth rates were further modeled as a function of temperature and water activity using the cardinal model with inflection (CMI) ( Rosso et al., 1993). A logistic regression quadratic polynomial model was also employed to predict the probability of growth over storage time. Estimated parameters for minimum, maximum and optimum temperatures for growth were 9.1 °C, 46.4 °C and 32.1 °C for B. fulva and 10.5 °C, 43.2 °C and 32.1 °C for B. nivea. The respective values for a_w were 0.893, 0.993 and 0.985 for B. fulva and 0.892, 0.992 and 0.984 for B. nivea. No growth was observed at 0.88 a_w regardless of temperature for both species, whereas B. nivea ascospores could not grow at 10 and 45 °C irrespective of a_w. Regarding growth boundaries, the degree of agreement between predictions and observations was >98% concordant for both species. The erroneously predicted growth cases were 1.4–4.2% false positive and 2.1–3.5% false negative for B. nivea and B. fulva, respectively. The developed logistic model was validated with two literature data sets as well as with data from independent experiments carried out on fruit juices. Validation results showed that agreement with literature data for growth was 25 out of 36 (69.4%) cases, whereas validation on fruit juice data failed in only 6 cases (5 false positives and 1 false negative) out of 128 cases.     

LES and RANS comparison of flow and pollutant dispersion in urban environment

This study investigates air pollution dispersion in urban areas by means of Computational Fluid Dynamics (CFD). The commercial CFD software FLUENT was used to implement two different turbulence simulation methods (RANS and LES), in domains similar to complex urban environments. Particularly, different combinations of roof shapes were studied and simulation results of pollutant (ethane) concentrations were compared against experimental data. The building height (H) to the neighbour building distance (B) ratio was also taken into consideration. Previous studies showed that both RANS and LES models are accurate enough to predict pollutant concentrations fields in B/H = 1. In the present study the incapability of RANS models to predict accurately pollutant concentration in B/H = 0.5 for all roof shapes configurations is revealed.     

Improved weight management using genetic information to personalize a calorie controlled diet

Background  

Gene-environment studies demonstrate variability in nutrient requirements depending upon individual variations in genes affecting nutrient metabolism and transport. This study investigated whether the inclusion of genetic information to personalize a patient's diet (nutrigenetics) could improve long term weight management.     

Microstructural Phenomena Controlling Losses in NiCuZn-Ferrites as Studied by Transmission Electron Microscopy

The magnetic losses of NiCuZn-ferrites with the chemical formula (Ni_0.37−δZn_0.63Cu_δ)Fe_1.93O₄ (0<δ<0.22) are found to exhibit a minimum as a function of the copper content. This minimum corresponds to a certain sintering temperature, which is also a function of the copper content. At sintering temperatures lower than the optimum, insufficient densification is considered to be the reason for increased losses. At sintering temperatures higher than the optimum, the increased losses are caused by the presence of copper segregation to the grain boundaries and by some identified microstructural defects such as subgrain boundaries and compositional modulations. Internal stress relaxation is believed to be the reason for those chemical and morphological imperfections. In contrast to the power losses, the magnetic permeability does not show the previous dependency and increases monotonically with increased density or grain size.