Simple approximate MAP inference for Dirichlet processes mixtures

The Dirichlet process mixture model (DPMM) is a ubiquitous, flexible Bayesian nonparametric statistical model. However, full probabilistic inference in this model is analytically intractable, so that computationally intensive techniques such as Gibbs sampling are required. As a result, DPMM-based methods, which have considerable potential, are restricted to applications in which computational resources and time for inference is plentiful. For example, they would not be practical for digital signal processing on embedded hardware, where computational resources are at a serious premium. Here, we develop a simplified yet statistically rigorous approximate maximum a-posteriori (MAP) inference algorithm for DPMMs. This algorithm is as simple as DP-means clustering, solves the MAP problem as well as Gibbs sampling, while requiring only a fraction of the computational effort. (For freely available code that implements the MAP-DP algorithm for Gaussian mixtures see http://www.maxlittle.net/.) Unlike related small variance asymptotics (SVA), our method is non-degenerate and so inherits the “rich get richer” property of the Dirichlet process. It also retains a non-degenerate closed-form likelihood which enables out-of-sample calculations and the use of standard tools such as cross-validation. We illustrate the benefits of our algorithm on a range of examples and contrast it to variational, SVA and sampling approaches from both a computational complexity perspective as well as in terms of clustering performance. We demonstrate the wide applicabiity of our approach by presenting an approximate MAP inference method for the infinite hidden Markov model whose performance contrasts favorably with a recently proposed hybrid SVA approach. Similarly, we show how our algorithm can applied to a semiparametric mixed-effects regression model where the random effects distribution is modelled using an infinite mixture model, as used in longitudinal progression modelling in population health science. Finally, we propose directions for future research on approximate MAP inference in Bayesian nonparametrics.

Effects of fibrosis morphology on reentrant ventricular tachycardia inducibility and simulation fidelity in patient-derived models

Myocardial fibrosis detected via delayed-enhanced magnetic resonance imaging (MRI) has been shown to be a strong indicator for ventricular tachycardia (VT) inducibility. However, little is known regarding how inducibility is affected by the details of the fibrosis extent, morphology, and border zone configuration. The objective of this article is to systematically study the arrhythmogenic effects of fibrosis geometry and extent, specifically on VT inducibility and maintenance. We present a set of methods for constructing patient-specific computational models of human ventricles using in vivo MRI data for patients suffering from hypertension, hypercholesterolemia, and chronic myocardial infarction. Additional synthesized models with morphologically varied extents of fibrosis and gray zone (GZ) distribution were derived to study the alterations in the arrhythmia induction and reentry patterns. Detailed electrophysiological simulations demonstrated that (1) VT morphology was highly dependent on the extent of fibrosis, which acts as a structural substrate, (2) reentry tended to be anchored to the fibrosis edges and showed transmural conduction of activations through narrow channels formed within fibrosis, and (3) increasing the extent of GZ within fibrosis tended to destabilize the structural reentry sites and aggravate the VT as compared to fibrotic regions of the same size and shape but with lower or no GZ. The approach and findings represent a significant step toward patient-specific cardiac modeling as a reliable tool for VT prediction and management of the patient. Sensitivities to approximation nuances in the modeling of structural pathology by image-based reconstruction techniques are also implicated.

Variational and learning models for image and time series inverse problems

Inverse problems are at the core of many challenging applications. Variational and learning models provide estimated solutions of inverse problems as the outcome of specific reconstruction maps. In the variational approach, the result of the reconstruction map is the solution of a regularized minimization problem encoding information on the acquisition process and prior knowledge on the solution. In the learning approach, the reconstruction map is a parametric function whose parameters are identified by solving a minimization problem depending on a large set of data. In this thesis, we go beyond this apparent dichotomy between variational and learning models and we show they can be harmoniously merged in unified hybrid frameworks preserving their main advantages. We develop several highly efficient methods based on both these model-driven and data-driven strategies, for which we provide a detailed convergence analysis. The arising algorithms are applied to solve inverse problems involving images and time series. For each task, we show the proposed schemes improve the performances of many other existing methods in terms of both computational burden and quality of the solution. In the first part, we focus on gradient-based regularized variational models which are shown to be effective for segmentation purposes and thermal and medical image enhancement. We consider gradient sparsity-promoting regularized models for which we develop different strategies to estimate the regularization strength. Furthermore, we introduce a novel gradient-based Plug-and-Play convergent scheme considering a deep learning based denoiser trained on the gradient domain. In the second part, we address the tasks of natural image deblurring, image and video super resolution microscopy and positioning time series prediction, through deep learning based methods. We boost the performances of supervised, such as trained convolutional and recurrent networks, and unsupervised deep learning strategies, such as Deep Image Prior, by penalizing the losses with handcrafted regularization terms.

Applications of network-based approaches in drug research

Recent research trends in computer-aided drug design have shown an increasing interest towards the implementation of advanced approaches able to deal with large amount of data. This demand arose from the awareness of the complexity of biological systems and from the availability of data provided by high-throughput technologies. As a consequence, drug research has embraced this paradigm shift exploiting approaches such as that based on networks. Indeed, the process of drug discovery can benefit from the implementation of network-based methods at different steps from target identification to drug repurposing. From this broad range of opportunities, this thesis is focused on three main topics: (i) chemical space networks (CSNs), which are designed to represent and characterize bioactive compound data sets; (ii) drug-target interactions (DTIs) prediction through a network-based algorithm that predicts missing links; (iii) COVID-19 drug research which was explored implementing COVIDrugNet, a network-based tool for COVID-19 related drugs. The main highlight emerged from this thesis is that network-based approaches can be considered useful methodologies to tackle different issues in drug research. In detail, CSNs are valuable coordinate-free, graphically accessible representations of structure-activity relationships of bioactive compounds data sets especially for medium-large libraries of molecules. DTIs prediction through the random walk with restart algorithm on heterogeneous networks can be a helpful method for target identification. COVIDrugNet is an example of the usefulness of network-based approaches for studying drugs related to a specific condition, i.e., COVID-19, and the same ‘systems-based’ approaches can be used for other diseases. To conclude, network-based tools are proving to be suitable in many applications in drug research and provide the opportunity to model and analyze diverse drug-related data sets, even large ones, also integrating different multi-domain information.

Data-stream driven Fuzzy-granular approaches for system maintenance

Intelligent systems are currently inherent to the society, supporting a synergistic human-machine collaboration. Beyond economical and climate factors, energy consumption is strongly affected by the performance of computing systems. The quality of software functioning may invalidate any improvement attempt. In addition, data-driven machine learning algorithms are the basis for human-centered applications, being their interpretability one of the most important features of computational systems. Software maintenance is a critical discipline to support automatic and life-long system operation. As most software registers its inner events by means of logs, log analysis is an approach to keep system operation. Logs are characterized as Big data assembled in large-flow streams, being unstructured, heterogeneous, imprecise, and uncertain. This thesis addresses fuzzy and neuro-granular methods to provide maintenance solutions applied to anomaly detection (AD) and log parsing (LP), dealing with data uncertainty, identifying ideal time periods for detailed software analyses. LP provides deeper semantics interpretation of the anomalous occurrences. The solutions evolve over time and are general-purpose, being highly applicable, scalable, and maintainable. Granular classification models, namely, Fuzzy set-Based evolving Model (FBeM), evolving Granular Neural Network (eGNN), and evolving Gaussian Fuzzy Classifier (eGFC), are compared considering the AD problem. The evolving Log Parsing (eLP) method is proposed to approach the automatic parsing applied to system logs. All the methods perform recursive mechanisms to create, update, merge, and delete information granules according with the data behavior. For the first time in the evolving intelligent systems literature, the proposed method, eLP, is able to process streams of words and sentences. Essentially, regarding to AD accuracy, FBeM achieved (85.64+-3.69)%; eGNN reached (96.17+-0.78)%; eGFC obtained (92.48+-1.21)%; and eLP reached (96.05+-1.04)%. Besides being competitive, eLP particularly generates a log grammar, and presents a higher level of model interpretability.

Applied genomics to monitor the genetic diversity and integrity of honey bee (Apis mellifera) populations and define sustainable strategies to benefit the beekeeping sector

Honey bees are considered keystone species in ecosystem, the effect of harmful pesticides for the honey bees, the action of extreme climatic waves and their consequence on honey bees health can cause the loss of many colonies which could contribute to the reduction of the effective population size and incentive the use of non-autochthonous queens to replace dead colonies. Over the last decades, the use of non-ligustica bee subspecies in Italy has increased and together with the mentioned phenomena exposed native honey bees to hybridization, laeding to a dramatic loss of genetic erosion and admixture. Healthy genetic diversity within honey bee populations is critical to provide tolerance and resistance to current and future threatening. Nowadays it is urgent to design strategies for the conservation of local subspecies and their valorisation on a productive scale. In this Thesis we applied genomics tool for the analysis of the genetic diversity and the genomic integrity of honey bee populations in Italy are described. In this work mtDNA based methods are presented using honey bee DNA or honey eDNA as source of information of the genetic diversity of A. mellifera at different level. Taken together, the results derived from these studies should enlarge the knowledge of the genetic diversity and integrity of the honey bee populations in Italy, filling the gap of information necessary to design efficient conservation programmes. Furthermore, the methods presented in these works will provide a tool for the honey authentication to sustain and valorise beekeeping products and sector against frauds.

Wavelets And Adaptive Grids For The Discontinuous Galerkin Method

In this paper, space adaptivity is introduced to control the error in the numerical solution of hyperbolic systems of conservation laws. The reference numerical scheme is a new version of the discontinuous Galerkin method, which uses an implicit diffusive term in the direction of the streamlines, for stability purposes. The decision whether to refine or to unrefine the grid in a certain location is taken according to the magnitude of wavelet coefficients, which are indicators of local smoothness of the numerical solution. Numerical solutions of the nonlinear Euler equations illustrate the efficiency of the method. © Springer 2005.

Migração lateral da desembocadura do Rio Itapocú, SC, Brasil: evolução morfológica e condicionantes físicas

Desembocaduras são ambientes bastante dinâmicos e sujeitos à complexa interação entre fatores estabilizadores e desestabilizadores. Dependendo dessa interação, desembocaduras podem apresentar a tendência de migração ao longo de barreiras arenosas. Um dos mecanismos mais eficientes de transporte de sedimento paralelo à costa, e consequentemente migração de canais, são as correntes longitudinais geradas pelas ondas se aproximando obliquamente à costa. A motivação do presente trabalho é entender o comportamento morfodinâmico do sistema de desembocadura do rio Itapocú, localizado no centro-norte de Santa Catarina (SC), frente aos processos forçantes que atuam na sua migração ao longo da linha de costa. A morfologia dos pontais arenosos foi obtida a partir de levantamentos morfológicos com o uso de DGPS. Para analisar a refração de ondas foi utilizado o modelo numérico MIKE 21 SW, sendo considerados como condições de contorno os dados de ondas referentes ao ano de 2002 e os dados de ondas previstos referentes ao período de coleta. Os dados de saída do modelo foram utilizados para estimar a deriva litorânea potencial na região. Os resultados morfológicos obtidos demonstraram uma migração da desembocadura para o norte durante o período analisado, sendo mais intenso durante o inverno e o verão. Ondas incidentes do quadrante sul sofreram mais o fenômeno da refração e as ondas de leste apresentaram menor variação angular ao se aproximarem à costa. A deriva litorânea potencial anual para os dados de ondas de 2002 apresentou sentido norte-sul, com inversão de sentido durante o outono. Utilizando os dados de ondas previstas para o período dos levantamentos, a deriva litorânea potencial estimada apresentou sentido sul-norte, concordando com a migração observada. Na região próxima a desembocadura, nos pontais arenosos, a deriva potencial apresentou direção para o norte durante todas as estações. Os dados de descarga fluvial não apresentaram influência na migração do canal, porém apresentaram uma relação com a largura do mesmo sazonalmente.Os dados de morfologia juntamente com os dados de deriva litorânea referentes às ondas de 2004/2005 mostraram claramente a migração do canal para o norte sendo a deriva a principal contribuinte para a migração da desembocadura.

Increased Occurrence of Dental Anomalies Associated with Second-Premolar Agenesis

Objective: To evaluate the prevalence of dental anomalies in patients with agenesis of second premolars and compare the findings with the prevalence of these anomalies in the general population. Materials and Methods: A Brazilian sample of 203 patients aged 8 to 22 years was selected. All patients presented agenesis of at least one second premolar. Panoramic and periapical radiographs and dental casts were used to analyze the presence of other associated dental anomalies, including agenesis of other permanent teeth, ectopia of unerupted permanent teeth, infraocclusion of deciduous molars, microdontia of maxillary lateral incisors, and supernumerary teeth. The occurrence of these anomalies was compared with occurrence data previously reported for the general population. Statistical testing was performed using the chi-square test (P < .05) and the odds ratio. Results: The sample with agenesis of at least one second premolar presented a significantly increased prevalence rate of permanent tooth agenesis (21%), excluding third molars. Among the sample segment aged 14 years or greater (N = 77), occurrence of third-molar agenesis (48%) exceeded twice its normal frequency. Significant increases in occurrence of microdontia of maxillary lateral incisors (20.6%), infraocclusion of deciduous molars (24.6%), and distoangulation of mandibular second premolars (7.8%) were observed. Palatally displaced canine anomaly was also significantly elevated (8.1%). Conclusion: The results provide evidence that agenesis of other permanent teeth, microdontia, deciduous molar infraocclusion, and certain dental ectopias are the products of the same genetic mechanisms that cause second-premolar agenesis. (Angle Orthod. 2009;79:436-441.)

Comparison of univariate and multivariate calibration for the determination of micronutrients in pellets of plant materials by laser induced breakdown spectrometry

The application of laser induced breakdown spectrometry (LIBS) aiming the direct analysis of plant materials is a great challenge that still needs efforts for its development and validation. In this way, a series of experimental approaches has been carried out in order to show that LIBS can be used as an alternative method to wet acid digestions based methods for analysis of agricultural and environmental samples. The large amount of information provided by LIBS spectra for these complex samples increases the difficulties for selecting the most appropriated wavelengths for each analyte. Some applications have suggested that improvements in both accuracy and precision can be achieved by the application of multivariate calibration in LIBS data when compared to the univariate regression developed with line emission intensities. In the present work, the performance of univariate and multivariate calibration, based on partial least squares regression (PLSR), was compared for analysis of pellets of plant materials made from an appropriate mixture of cryogenically ground samples with cellulose as the binding agent. The development of a specific PLSR model for each analyte and the selection of spectral regions containing only lines of the analyte of interest were the best conditions for the analysis. In this particular application, these models showed a similar performance. but PLSR seemed to be more robust due to a lower occurrence of outliers in comparison to the univariate method. Data suggests that efforts dealing with sample presentation and fitness of standards for LIBS analysis must be done in order to fulfill the boundary conditions for matrix independent development and validation. (C) 2009 Elsevier B.V. All rights reserved.

Prediction of Crude Oil Properties and Chemical Composition by Means of Steady-State and Time-Resolved Fluorescence

Steady-state and time-resolved fluorescence measurements are reported for several crude oils and their saturates, aromatics, resins, and asphaltenes (SARA) fractions (saturates, aromatics and resins), isolated from maltene after pentane precipitation of the asphaltenes. There is a clear relationship between the American Petroleum Institute (API) grade of the crude oils and their fluorescence emission intensity and maxima. Dilution of the crude oil samples with cyclohexane results in a significant increase of emission intensity and a blue shift, which is a clear indication of the presence of energy-transfer processes between the emissive chromophores present in the crude oil. Both the fluorescence spectra and the mean fluorescence lifetimes of the three SARA fractions and their mixtures indicate that the aromatics and resins are the major contributors to the emission of crude oils. Total synchronous fluorescence scan (TSFS) spectral maps are preferable to steady-state fluorescence spectra for discriminating between the fractions, making TSFS maps a particularly interesting choice for the development of fluorescence-based methods for the characterization and classification of crude oils. More detailed studies, using a much wider range of excitation and emission wavelengths, are necessary to determine the utility of time-resolved fluorescence (TRF) data for this purpose. Preliminary models constructed using TSFS spectra from 21 crude oil samples show a very good correlation (R(2) > 0.88) between the calculated and measured values of API and the SARA fraction concentrations. The use of models based on a fast fluorescence measurement may thus be an alternative to tedious and time-consuming chemical analysis in refineries.

Leaf area index estimation with MODIS reflectance time series and model inversion during full rotations of Eucalyptus plantations

The leaf area index (LAI) of fast-growing Eucalyptus plantations is highly dynamic both seasonally and interannually, and is spatially variable depending on pedo-climatic conditions. LAI is very important in determining the carbon and water balance of a stand, but is difficult to measure during a complete stand rotation and at large scales. Remote-sensing methods allowing the retrieval of LAI time series with accuracy and precision are therefore necessary. Here, we tested two methods for LAI estimation from MODIS 250m resolution red and near-infrared (NIR) reflectance time series. The first method involved the inversion of a coupled model of leaf reflectance and transmittance (PROSPECT4), soil reflectance (SOILSPECT) and canopy radiative transfer (4SAIL2). Model parameters other than the LAI were either fixed to measured constant values, or allowed to vary seasonally and/or with stand age according to trends observed in field measurements. The LAI was assumed to vary throughout the rotation following a series of alternately increasing and decreasing sigmoid curves. The parameters of each sigmoid curve that allowed the best fit of simulated canopy reflectance to MODIS red and NIR reflectance data were obtained by minimization techniques. The second method was based on a linear relationship between the LAI and values of the GEneralized Soil Adjusted Vegetation Index (GESAVI), which was calibrated using destructive LAI measurements made at two seasons, on Eucalyptus stands of different ages and productivity levels. The ability of each approach to reproduce field-measured LAI values was assessed, and uncertainty on results and parameter sensitivities were examined. Both methods offered a good fit between measured and estimated LAI (R(2) = 0.80 and R(2) = 0.62 for model inversion and GESAVI-based methods, respectively), but the GESAVI-based method overestimated the LAI at young ages. (C) 2010 Elsevier Inc. All rights reserved.

Hereditary hemochromatosis: Mutations in genes involved in iron homeostasis in Brazilian patients

Background: p.C282Y mutation and rare variants in the HFE gene have been associated with hereditary hemochromatosis (HH). HH is also caused by mutations in other genes, such as the hemojuvelin (HJV), hepcidin (HAMP), transferrin receptor 2 (TFR2) and ferroportin (SLC40A1). The low rate homozygous p.C282Y mutation in Brazil is suggestive that mutations in non-HFE genes may be linked to HH phenotype. Aim: To screen exon-by-exon DNA sequences of HFE, HJV, HAMP, TFR2 and SLC40A1 genes to characterize the molecular basis of HH in a sample of the Brazilian population. Materials and methods: Fifty-one patients with primary iron overload (transferrin saturation >= 50% in females and >= 60% in males) were selected. Subsequent bidirectional DNA sequencing of HFE, HJV, HAMP, TFR2 and SLC40A1 exons was performed. Results: Thirty-seven (72.5%) out of the 51 patients presented at least one HFE mutation. The most frequent genotype associated with HH was the homozygous p.C282Y mutation (n = 11, 21.6%). In addition, heterozygous HFE p.S65C mutation was found in combination with p.H63D in two patients and homozygous HFE p.H63D was found in two patients as well. Sequencing in the HJV and HAMP genes revealed HJV p.E302K, HJV p.A310G, HJV p.G320V and HAMP p.R59G alterations. Molecular and clinical diagnosis of juvenile hemochromatosis (homozygous form for the HJV p.G320V) was described for the first time in Brazil. Three TFR2 polymorphisms (p.A75V, p.A617A and p.R752H) and six SLC40A1 polymorphisms (rs13008848, rs11568351, rs11568345, rs11568344, rs2304704, rs11568346) and the novel mutation SLC40A1 p.G204S were also found. Conclusions: The HE p.C282Y in homozygosity or in heterozygosity with p.H63D was the most frequent mutation associated with HH in this sample. The HJV p.E302K and HAMP p.R59G variants, and the novel SLC40A1 p.G2045 mutation may also be linked to primary iron overload but their role in the pathophysiology of HH remain to be elucidated. (C) 2011 Elsevier Inc. All rights reserved.

Embedded Voxel Colouring

The reconstruction of a complex scene from multiple images is a fundamental problem in the field of computer vision. Volumetric methods have proven to be a strong alternative to traditional correspondence-based methods due to their flexible visibility models. In this paper we analyse existing methods for volumetric reconstruction and identify three key properties of voxel colouring algorithms: a water-tight surface model, a monotonic carving order, and causality. We present a new Voxel Colouring algorithm which embeds all reconstructions of a scene into a single output. While modelling exact visibility for arbitrary camera locations, Embedded Voxel Colouring removes the need for a priori threshold selection present in previous work. An efficient implementation is given along with results demonstrating the advantages of posteriori threshold selection.

Special issue on advances in data mining and its applications

Data mining is the process to identify valid, implicit, previously unknown, potentially useful and understandable information from large databases. It is an important step in the process of knowledge discovery in databases, (Olaru & Wehenkel, 1999). In a data mining process, input data can be structured, seme-structured, or unstructured. Data can be in text, categorical or numerical values. One of the important characteristics of data mining is its ability to deal data with large volume, distributed, time variant, noisy, and high dimensionality. A large number of data mining algorithms have been developed for different applications. For example, association rules mining can be useful for market basket problems, clustering algorithms can be used to discover trends in unsupervised learning problems, classification algorithms can be applied in decision-making problems, and sequential and time series mining algorithms can be used in predicting events, fault detection, and other supervised learning problems (Vapnik, 1999). Classification is among the most important tasks in the data mining, particularly for data mining applications into engineering fields. Together with regression, classification is mainly for predictive modelling. So far, there have been a number of classification algorithms in practice. According to (Sebastiani, 2002), the main classification algorithms can be categorized as: decision tree and rule based approach such as C4.5 (Quinlan, 1996); probability methods such as Bayesian classifier (Lewis, 1998); on-line methods such as Winnow (Littlestone, 1988) and CVFDT (Hulten 2001), neural networks methods (Rumelhart, Hinton & Wiliams, 1986); example-based methods such as k-nearest neighbors (Duda & Hart, 1973), and SVM (Cortes & Vapnik, 1995). Other important techniques for classification tasks include Associative Classification (Liu et al, 1998) and Ensemble Classification (Tumer, 1996).