A Markov random field model for combining optimum-path forest classifiers using decision graphs and game strategy approach

The research on multiple classifiers systems includes the creation of an ensemble of classifiers and the proper combination of the decisions. In order to combine the decisions given by classifiers, methods related to fixed rules and decision templates are often used. Therefore, the influence and relationship between classifier decisions are often not considered in the combination schemes. In this paper we propose a framework to combine classifiers using a decision graph under a random field model and a game strategy approach to obtain the final decision. The results of combining Optimum-Path Forest (OPF) classifiers using the proposed model are reported, obtaining good performance in experiments using simulated and real data sets. The results encourage the combination of OPF ensembles and the framework to design multiple classifier systems. © 2011 Springer-Verlag.

Uterine and luteal hemodynamic evaluation of the non pregnant mare

Background: Studies with Doppler ultrassonography started at the end of the 90s for the determination of physiological and pathological alterations in the reproductive tract of the mare. Uterine alterations caused by inflammation, response from seminal plasma infusion, hormonal variations during estrous and diestrus, pregnancy and action of various vasoactive factors influence on the vascular perfusion detected by Doppler ultrasound. The development of efficient methods for uterine quality evaluation is of big importance for field equine reproduction veterinarians, once uterine environment is responsible for pregnancy maintenance. Review: Nowadays, the most used methods of uterine evaluation are the mode B ultrassonography, cytology, culture and biopsy. Hemodynamic evaluation of the uterus can be done by spectral data collected from large vessels, as A. uterine and its ramifications, or from subjective or objective evaluations from endometrium, miometrium and mesometrium attachment, which provide data referent to local and specific alterations of the evaluated area. Alterations in uterine vascular perfusion has been detected during estrous cycle, during pregnancy and in cases of infusion of inflammatory substances. These alterations happen because of vasoactive substances that act in the uterus during these events, however, most of these vasoactive substances are probably not even known. Also, important hemodynamic alterations in old mares, as an increase in vascular resistance, have been described. This increase might result from fibrosis of the uterus and in women it is considered to be a cause of infertility. In mares, periglandular fibrosis of the endometrium is considered to be the major diagnosable cause of embryonic and fetal loss in older mares. For the CL, ovarian artery of the mare supplies the ovary as well as the oviduct and therefore can be used for evaluation of these areas. The CL evaluation can also be done by the percentage of luteum area with colored signals as an indicator of the extent of blood flow. The percentage of the CL area with colored signals is determined subjectively by images observations in real time and/or by a freezing Power Doppler cross-section image with the maximum number of color pixels taped and the total number of color pixels is assessed by a computer analysis system. Therefore, a high correlation between plasmatic progesterone and CL vascularization also allows the CL evalution by this technique. In a first report, CL circulation reached its maximum on D5, the progesterone concentration in peripheral blood increased until D7 and in a posterior report, maximum perfusion was achieved two days after the maximum progesterone concentration (D8). Blood flow reduced between D10-D14 some days before the plasma progesterone decrease and, during the luteolytic period (D15-D17), the decline in CL blood-flow area was greater than blood flow decrease. Conclusion: Doppler ultrassonography add knowledge about uterine viability and CL functionality can be easily used by veterinarians in the field. It is a noninvasive method that provides real time results. However, because of the short time studies in this area have been done, many other answers still need to be found until normal and pathological patterns will be established.

Mutual occlusion between real and virtual elements in Augmented Reality based on fiducial markers

Augmented Reality (AR) systems which use optical tracking with fiducial marker for registration have had an important role in popularizing this technology, since only a personal computer with a conventional webcam is required. However, in most these applications, the virtual elements are shown only in the foreground a real element does not occlude a virtual one. The method presented enables AR environments based on fiducial markers to support mutual occlusion between a real element and many virtual ones, according to the elements position (depth) in the environment. © 2012 IEEE.

High Content Screening of a Kinase-Focused Library Reveals Compounds Broadly-Active against Dengue Viruses

Dengue virus is a mosquito-borne flavivirus that has a large impact in global health. It is considered as one of the medically important arboviruses, and developing a preventive or therapeutic solution remains a top priority in the medical and scientific community. Drug discovery programs for potential dengue antivirals have increased dramatically over the last decade, largely in part to the introduction of high-throughput assays. In this study, we have developed an image-based dengue high-throughput/high-content assay (HT/HCA) using an innovative computer vision approach to screen a kinase-focused library for anti-dengue compounds. Using this dengue HT/HCA, we identified a group of compounds with a 4-(1-aminoethyl)-N-methylthiazol-2-amine as a common core structure that inhibits dengue viral infection in a human liver-derived cell line (Huh-7.5 cells). Compounds CND1201, CND1203 and CND1243 exhibited strong antiviral activities against all four dengue serotypes. Plaque reduction and time-of-addition assays suggests that these compounds interfere with the late stage of viral infection cycle. These findings demonstrate that our image-based dengue HT/HCA is a reliable tool that can be used to screen various chemical libraries for potential dengue antiviral candidates. © 2013 Cruz et al.

Sistema microcontrolado para medição de forças na região plantar e nos membros superiores de pacientes

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Análise de multirresolução baseada em polinômio potência de Sigmóide - Wavelet

Pós-graduação em Ciência da Computação - IBILCE

Estudo cinesiográfico da deglutição em indivíduos com dentição natural e com próteses totais

Pós-graduação em Reabilitação Oral - FOAR

Improving face recognition with multispectral fusion and support vector machines

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Composição de dicionários visuais utilizando agrupamento de dados por Florestas de Caminhos Ótimos

Pós-graduação em Ciência da Computação - IBILCE

Desenvolvimento e teste de um sistema para separação automática das fases glicerol-biodiesel utilizando sensor de condutividade elétrica

Pós-graduação em Engenharia Mecânica - FEIS

Static Video Summarization through Optimum-Path Forest Clustering

This paper introduces the Optimum-Path Forest (OPF) classifier for static video summarization, being its results comparable to the ones obtained by some state-of-the-art video summarization techniques. The experimental section has been conducted using several image descriptors in two public datasets, followed by an analysis of OPF robustness regarding one ad-hoc parameter. Future works are guided to improve OPF effectiveness on each distinct video category.

HTS and HTSn: New shape descriptors based on Hough transform statistics

With the widespread proliferation of computers, many human activities entail the use of automatic image analysis. The basic features used for image analysis include color, texture, and shape. In this paper, we propose a new shape description method, called Hough Transform Statistics (HTS), which uses statistics from the Hough space to characterize the shape of objects or regions in digital images. A modified version of this method, called Hough Transform Statistics neighborhood (HTSn), is also presented. Experiments carried out on three popular public image databases showed that the HTS and HTSn descriptors are robust, since they presented precision-recall results much better than several other well-known shape description methods. When compared to Beam Angle Statistics (BAS) method, a shape description method that inspired their development, both the HTS and the HTSn methods presented inferior results regarding the precision-recall criterion, but superior results in the processing time and multiscale separability criteria. The linear complexity of the HTS and the HTSn algorithms, in contrast to BAS, make them more appropriate for shape analysis in high-resolution image retrieval tasks when very large databases are used, which are very common nowadays. (C) 2014 Elsevier Inc. All rights reserved.

Expression of the classical and nonclassical HLA molecules in breast cancer

Considering that downregulation of HLA expression could represent a potential mechanism for breast carcinogenesis and metastasis, the aim of the present study was to use immunohistochemical methods to analyze the expression of HLA-Ia, HLA-DR, HLA-DQ, HLA-E, and HLA-G in invasive ductal carcinoma (IDC) of the breast and to relate this HLA profile to anatomopathological parameters. Fifty-two IDC from breast biopsies were stratified according to histological differentiation (well, moderately, and poorly differentiated) and to the presence of metastases in axillary lymph nodes. The expression of HLA molecules was assessed by immunohistochemistry, using a computer-assisted system. Overall, 31 (59.6%) out of the 52 IDC breast biopsies exhibited high expression of HLA-G, but only 14 (26.9%) showed high expression of HLA-E. A large number (41, 78.8%) of the biopsies showed low expression of HLA-Ia, while 45 (86.5%) showed high expression of HLA-DQ and 36 (69.2%) underexpressed HLA-DR. Moreover, 24 (41.2%) of 52 biopsies had both low HLA-Ia expression and high HLA-G expression, while 11 (21.2%) had low HLA-Ia expression and high HLA-E expression. These results suggest that, by different mechanisms, the downregulation of HLA-Ia, HLA-E, and HLA-DR and the upregulation of HLA-G and HLA-DQ are associated with immune response evasion and breast cancer aggressiveness.

A simulation tool for transient stability analysis on vector computers

This paper deals with transient stability analysis based on time domain simulation on vector processing. This approach requires the solution of a set of differential equations in conjunction of another set of algebraic equations. The solution of the algebraic equations has presented a scalar as sequential set of tasks, and the solution of these equations, on vector computers, has required much more investigations to speedup the simulations. Therefore, the main objective of this paper has been to present methods to solve the algebraic equations using vector processing. The results, using a GRAY computer, have shown that on-line transient stability assessment is feasible.

Swarm-based descriptor combination and its application for image classification

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)