Fernando Mateo

Changes in ochratoxin A and type B trichothecenes contained in wheat flour during dough fermentation and bread-baking.

Ochratoxin A (OTA) and type B trichothecenes are mycotoxins that occur frequently in cereals and thus can be found in cereal by-products such as bread. The aim of this work was to study the variation of the levels of OTA, deoxynivalenol (DON), 3-acetyldeoxynivalenol (3-ADON) and nivalenol (NIV) during the bread-making process. This was done by using wheat flour spiked with different levels of toxins. Mycotoxin levels were controlled after fermentation of the dough with yeasts (Saccharomyces cerevisiae) and after further baking at different temperature–time combinations. Analysis of variance (ANOVA) of the results showed a significant reduction in OTA level (p < 0.05) during fermentation of the dough. The reduction ranged between 29.8% and 33.5%, depending on the initial concentration of toxin in the flour. During this period, the level of the other mycotoxins studied was not modified. By contrast, in the baking phase there were significant changes in the levels of the four mycotoxins, although the reduction was similar under all the baking conditions. Considering all the temperature–time conditions tested, it can be concluded that during the baking period the average reduction of OTA, NIV, 3-ADON, and DON was 32.9%, 76.9%, 65.6%, and 47.9%, respectively.

Influence of nitrogen and carbon sources on the production of ochratoxin A by ochratoxigenic strains of Aspergillus spp. isolated from grapes.

This work studies the influence of nitrogen and carbon source on ochratoxin A production by three Aspergillus isolates A. ochraceus (Aso2), A. carbonarius (Ac25) and A. tubingensis (Bo66), all isolated from grapes. A basal medium (0.01 g/l FeSO4.7H2O, 0.5 g/l MgSO4.7H2O, 0.5 g/l Na2HPO4.2H2O, 1.0 g/l KCl) was prepared. This medium was supplemented with different nitrogen sources, both inorganic [(NH4)3PO(4), 0.3 g/l plus NH4NO3, 0.2 g/l] and organic (histidine, proline, arginine, phenylalanine, tryptophan or tyrosine) at two concentrations (0.05 g/l or 0.3 g/l), and different carbon sources (sucrose, glucose, maltose, arabinose or fructose) at three concentrations (10 g/l, 50 g/l or 150 g/l). A medium with sucrose (18 g/l) and glucose (1 g/l) was also tested. After a 10-day incubation period at 25 degrees C the highest levels of OTA (44.0 ng/ml, 13.5 ng/ml and 0.49 ng/ml for A. ochraceus, A. carbonarius and A. tubingensis, respectively) were obtained in the cultures containing 150 g/l of arabinose and 0.05 g/l of phenylalanine. Analysis of variance of the data showed that there were significant differences (p-value 0.05) among the OTA levels in the cultures with regard to carbon source and isolate. No significant differences were detected in OTA production regarding nitrogen source, although 0.05 g/l of phenylalanine generally favoured OTA production in the cultures of the three isolates. The dynamics of toxin production in the cultures of each isolate using the optimized basal medium supplemented with 0.05 g/l of phenylalanine and 150 g/l of arabinose for a period of 42 days at 25 degrees C was also studied. The maximum level of OTA was detected on the 3rd day of incubation in A. tubingensis cultures and on the 35th and 43(rd) days of incubation in A. ochraceus and A. carbonarius, respectively. This is the first study in which defined media have been used to assess the influence of carbon and nitrogen sources on OTA production by isolates of OTA-producing species isolated from grapes and to analyse the dynamics of toxin production in these species in a defined culture medium. This optimized medium for OTA production is being used in current studies aimed at elucidating its biosynthetic pathway.

Minimising the delta test for variable selection in regression problems

The problem of selecting an adequate set of variables from a given data set of a sampled function becomes crucial by the time of designing the model that will approximate it. Several approaches have been presented in the literature although recent studies showed how the delta test is a powerful tool to determine if a subset of variables is correct. This paper presents new methodologies based on the delta test such as tabu search, genetic algorithms and the hybridisation of them, to determine a subset of variables which is representative of a function. The paper considers as well the scaling problem where a relevance value is assigned to each variable. The new algorithms were adapted to be run in parallel architectures so better performances could be obtained in a small amount of time, presenting great robustness and scalability.

Machine learning methods to forecast temperature in buildings

Efficient management of energy in buildings saves a very important amount of resources (both economic and technological). As a consequence, there is a very active research in this field. One of the keys of energy management is the prediction of the variables that directly affect building energy consumption and personal comfort. Among these variables, one can highlight the temperature in each room of a building. In this work we apply different machine learning techniques along with other classical ones for predicting the temperatures in different rooms. The obtained results demonstrate the validity of these techniques for predicting temperatures and, therefore, for the establishment of optimal policies of energy consumption.

Predictive assessment of ochratoxin A accumulation in grape juice based-medium by Aspergillus carbonarius using neural networks.

Aims:  To study the ability of multi‐layer perceptron artificial neural networks (MLP‐ANN) and radial‐basis function networks (RBFNs) to predict ochratoxin A (OTA) concentration over time in grape‐based cultures of Aspergillus carbonarius under different conditions of temperature, water activity (aw) and sub‐inhibitory doses of the fungicide carbendazim.

Approximate k-NN delta test minimization method using genetic algorithms: Application to time series

In many real world problems, the existence of irrelevant input variables (features) hinders the predictive quality of the models used to estimate the output variables. In particular, time series prediction often involves building large regressors of artificial variables that can contain irrelevant or misleading information. Many techniques have arisen to confront the problem of accurate variable selection, including both local and global search strategies. This paper presents a method based on genetic algorithms that intends to find a global optimum set of input variables that minimize the Delta Test criterion. The execution speed has been enhanced by substituting the exact nearest neighbor computation by its approximate version. The problems of scaling and projection of variables have been addressed. The developed method works in conjunction with MATLABs Genetic Algorithm and Direct Search Toolbox. The goodness of the proposed methodology has been evaluated on several popular time series examples, and also generalized to other non-time-series datasets.

RCGA-S/RCGA-SP Methods to Minimize the Delta Test for Regression Tasks

Frequently, the number of input variables (features) involved in a problem becomes too large to be easily handled by conventional machine-learning models. This paper introduces a combined strategy that uses a real-coded genetic algorithm to find the optimal scaling (RCGA-S) or scaling + projection (RCGA-SP) factors that minimize the Delta Test criterion for variable selection when being applied to the input variables. These two methods are evaluated on five different regression datasets and their results are compared. The results confirm the goodness of both methods although RCGA-SP performs clearly better than RCGA-S because it adds the possibility of projecting the input variables onto a lower dimensional space.

Optimal Pruned K-Nearest Neighbors: OP-KNN Application to Financial Modeling

The paper proposes a methodology called OP-KNN, which builds a one hidden-layer feed forward neural network, using nearest neighbors neurons with extremely small computational time. The main strategy is to select the most relevant variables beforehand, then to build the model using KNN kernels. Multi-response sparse regression (MRSR) is used as the second step in order to rank each k-th nearest neighbor and finally as a third step leave-one-out estimation is used to select the number of neighbors and to estimate the generalization performances. This new methodology is tested on a toy example and is applied to financial modeling.

Accurate Simulation Testbench for Nuclear Imaging Systems

Current testbenches for nuclear imaging devices aim to simulate only a single stage of the system at a time. This approach is useful in early design stages where accuracy is not necessary. However, it would be desirable that different tools could be combined to achieve more detailed simulations. In this work, we present a high precision testbench that has been developed to test nuclear imaging systems. Its accuracy lies in the possibility of linking different simulation tools using the right one for each part of the system. High energy events are simulated using Geant4 (High Energy Simulator). Analog and digital electronics are verified using Cadence Spectre and ModelSim. This testbench structure allows testing any physical topology, scintillation crystals, photomultiplier tubes (PMTs), avalanche photodiodes (APDs), with any kind of ASIC, discrete analog and digital electronics, thus reducing the prototyping and design time. New system developments can be easily verified using behavioral and circuital description models for analog and digital electronics. Finally, a dual-head continuous LSO scintillation crystal positron emission tomography (PET) system has been used as an example for evaluation of the testbench.

Incidence position estimation in a PET detector using a discretized positioning circuit and neural networks

The correct determination of the position of incident photons is a crucial issue in PET imaging. In this paper we study the use of Neural Networks (NNs) for position estimation of photons impinging on gamma-ray detector modules for PET cameras based on continuous scintillators and Multi-Anode Photomultiplier Tubes (MA-PMTs). We have performed a thorough analysis of the NN architecture and training procedures, using realistic simulated inputs, in order to achieve the best results in terms of spatial resolution and bias correction. The results confirm that NNs can partially model and correct the non-uniform detector response using only the position-weighted signals from a simple 2D Discretized Positioning Circuit (DPC). Linearity degradation for oblique incidence is also investigated. Finally, the NN can be implemented in hardware for parallel real time corrected Line-of-Response (LOR) estimation.