Classification of masses in mammographic image using wavelet domain features and polynomial classifier

Breast cancer is the most common cancer among women. In CAD systems, several studies have investigated the use of wavelet transform as a multiresolution analysis tool for texture analysis and could be interpreted as inputs to a classifier. In classification, polynomial classifier has been used due to the advantages of providing only one model for optimal separation of classes and to consider this as the solution of the problem. In this paper, a system is proposed for texture analysis and classification of lesions in mammographic images. Multiresolution analysis features were extracted from the region of interest of a given image. These features were computed based on three different wavelet functions, Daubechies 8, Symlet 8 and bi-orthogonal 3.7. For classification, we used the polynomial classification algorithm to define the mammogram images as normal or abnormal. We also made a comparison with other artificial intelligence algorithms (Decision Tree, SVM, K-NN). A Receiver Operating Characteristics (ROC) curve is used to evaluate the performance of the proposed system. Our system is evaluated using 360 digitized mammograms from DDSM database and the result shows that the algorithm has an area under the ROC curve Az of 0.98 ± 0.03. The performance of the polynomial classifier has proved to be better in comparison to other classification algorithms. © 2013 Elsevier Ltd. All rights reserved.

Desenvolvimento de uma ferramenta computacional de apoio ao ensino de sistemas eletrônicos digitais

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Retificador entrelaçado boost, no modo de condução descontínua, com técnica de correção da corrente de entrada e elevado fator de potência, para aplicação em sistema trólebus

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Optimum-Path Forest Applied for Breast Masses Classification

In Computer-Aided Diagnosis-based schemes in mammography analysis each module is interconnected, which directly affects the system operation as a whole. The identification of mammograms with and without masses is highly needed to reduce the false positive rates regarding the automatic selection of regions of interest for further image segmentation. This study aims to evaluate the performance of three techniques in classifying regions of interest as containing masses or without masses (without clinical findings), as well as the main contribution of this work is to introduce the Optimum-Path Forest (OPF) classifier in this context, which has never been done so far. Thus, we have compared OPF against with two sorts of neural networks in a private dataset composed by 120 images: Radial Basis Function and Multilayer Perceptron (MLP). Texture features have been used for such purpose, and the experiments have demonstrated that MLP networks have been slightly better than OPF, but the former is much faster, which can be a suitable tool for real-time recognition systems.

Análise dinâmica e estrutural de virabrequim

This paper deals with static analysis and dynamic stress of an ensemble crank and crankshaft, contained within a combustion engine of 4 cylinders. Aimed to illustrate procedures for computer-aided analysis having as goal the optimization of components according to the need for the market demand. Thus, to work in static and dynamic analyzes were summarized the basic theory involved in the calculations and analyzes necessary actuation forces and held a brief introduction about the manufacturing process and forging. Subsequently, it was applied with the use of software in a case the crankshaft, to obtain the performance and structural dynamic thereof. There was a conservative result and critical points in the fillet of the crankshaft bearing, as well as for lubrication hole. It was concluded that there are possibilities for improvements in the manufacturing process and design optimization in order to provide lower criticality and a more robust part

Análise e aferição de uma fresadora CNC através da utilização das tecnologias assistidas por computador, no escaneamento 3D de amostras e processamento de imagens

The use of computer-assisted technologies such as CAD - Computed Aided Design, CAM - Computed Aided Manufacturing, CAE - Computed Aided Engineering and CNC - Computed Numerical Control, are priorities in engineering and product designers. However, the dimensional measurement between the virtual and the real product design requires research, and dissemination procedures among its users. This work aims to use these technologies, through analysis and measurement of a CNC milling machine, designed and assembled in the university. Through the use of 3D scanning, and analyzing images of the machined samples, and its original virtual files, it was possible to compare the sizes of these samples in counterposition to the original virtual dimensions, we can state that the distortions between the real and virtual, are within acceptable limits for this type of equipment. As a secondary objective, this work seeks to disseminate and make more accessible the use of these technologies.

Three-Dimensional Characterization of Pores in Ti-6Al-4V Alloy

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Simulation Tools of DC-DC Converters for Power Electronics Education

This paper presents interactive simulation tools to support the enhancement of power electronics education considering DC to DC converters. These interactive tools make use of the benefits of Java language to provide a dynamic and interactive visualization of simulations in steady-state for idealized non-isolated DC to DC converters. Additionally, this paper discusses the features and the usage of the interactive educational tools to serve as a first design tool for the laboratory experiments in the power electronics courses. In this way, some results were confronted with a well-known simulator package and with experimental results in order to validate the developed interactive simulation tools.

Simulation Tools for Power Electronics Courses Based on Java Technologies

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Ensino e orientação para projeto de conversores CC-CC através de aplicativos em java

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Evaluation of fertilizing potential of frozen-thawed dog spermatozoa diluted in ACP-106 (R) using an in vitro sperm-oocyte interaction assay

The aim of present study was to evaluate frozen canine semen with ACP-106 (R) (Powder Coconut Water) using an in vitro sperm-oocyte interaction assay (SOIA). Ten ejaculates from five stud dogs were diluted in ACP-106 (R) containing 20% egg yolk, submitted to cooling in a thermal box for 40 min and in a refrigerator for 30 min. After this period, a second dilution was performed using ACP-106 (R) containing 20% egg yolk and 12% glycerol. Samples were thawed at 38 degrees C for 1 min. Post-thaw motility was evaluated by light microscopy and by using a computer aided semen analysis (CASA). Plasma membrane integrity and sperm morphology/acrosomal status were evaluated by fluorescent probes (C-FDA/PI) and Bengal Rose respectively. Moreover, frozen-thawed semen was analysed by a SOIA. Subjective post-thaw motility was 52.0 +/- 14.8% and it was significant higher than the total motility estimated by CASA (23.0 +/- 14.8%) because this system considered the egg yolk debris as immotile spermatozoa. Although normal sperm rate and acrosomal integrity evaluated by Bengal Rose stain was 89.6 +/- 3.1 % and 94.3 +/- 3.1 %, respectively, post-thaw percentage of intact plasma membrane was only 35.1 +/- 14.3%. Regarding SOIA, the percentage of interacted oocytes (bound, penetrated and bound and/or penetrated) was 75.3%. Using regression analysis, it was found significant relations between some CASA patterns and data for SOIA. In conclusion, the freezing-thawing procedure using ACP-106 (R) was efficient for maintain the in vitro fertility potential of dog spermatozoa.

Prognostic value of canine frozen-thawed semen parameters on in vitro sperm-oocyte interactions

Combining the data from conventional semen analysis with oocyte penetration assays should improve the assessment of the fertilizing ability of a semen sample. Thus, the objective of the present study was to evaluate the prognostic value of various semen parameters on the in vitro interactions between frozen-thawed canine sperm and homologous oocytes. Ten ejaculates from five stud dogs (two ejaculates/dog) were collected by digital manipulation. Semen samples were evaluated, extended in Tris-egg yolk-glycerol, frozen and stored in liquid nitrogen, and thawed several weeks later. Samples were evaluated for motility and sperm populations by computer-aided semen analysis (CASA), plasma membrane integrity (carboxy-fluorescein diacetate and propidium iodide), and sperm morphology (Bengal Rose). Thawed spermatozoa were also incubated with homologous oocytes for 18 h in an atmosphere of 5% CO2 and 95% air at 38 degrees C and sperm-oocyte interactions were evaluated. Simple linear regression models were calculated, with sperm parameters as independent variables and sperm-oocyte interactions as the dependent variable. There were significant associations between: percentage of oocytes bound to spermatozoa and beat cross frequency (BCF; R-2 = 63%); percentage of oocytes that interacted with spermatozoa and BCF (R-2 = 73%); and number of penetrated spermatozoa and velocity average pathway (VAP; R-2 = 64%) and velocity straight line (VSL; R-2 = 64%). Although plasma membrane integrity and sperm morphology had little prognostic value for in vitro interactions between canine frozen-thawed sperm and homologous oocytes, some motility patterns (evaluated by CASA) were predictive of in vitro sperm-oocyte interactions. (c) 2005 Elsevier B.V. All rights reserved.

Three-Dimensional Ultrasonographic Assessment of Fetal Total Lung Volume as a Prognostic Factor in Primary Pleural Effusion

Objectives-The purpose of this study was to predict perinatal outcomes using fetal total lung volumes assessed by 3-dimensional ultrasonography (3DUS) in primary pleural effusion.Methods-Between July 2005 and July 2010, total lung volumes were prospectively estimated in fetuses with primary pleural effusion by 3DUS using virtual organ computer-aided analysis software. The first and last US examinations were considered in the analysis. The observed/expected total lung volumes were calculated. Main outcomes were perinatal death (up to 28 days of life) and respiratory morbidity (orotracheal intubation with mechanical respiratory support >48 hours).Results-Twelve of 19 fetuses (63.2%) survived. Among the survivors, 7 (58.3%) had severe respiratory morbidity. The observed/expected total lung volume at the last US examination before birth was significantly associated with perinatal death (P < .01) and respiratory morbidity (P < .01) as well as fetal hydrops (P < .01) and bilateral effusion (P = .01).Conclusions-Fetal total lung volumes may be useful for the prediction of perinatal outcomes in primary pleural effusion.

Development of a micro-heat exchanger with stacked plates using LTCC technology

A green ceramic tape micro-heat exchanger was developed using Low Temperature Co-fired Ceramics technology (LTCC). The device was designed by using Computational Aided Design software and simulations were made using a Computational Fluid Dynamics package (COMSOL Multiphysics) to evaluate the homogeneity of fluid distribution in the microchannels. Four geometries were proposed and simulated in two and three dimensions to show that geometric details directly affect the distribution of velocity in the micro-heat exchanger channels. The simulation results were quite useful for the design of the microfluidic device. The micro-heat exchanger was then constructed using the LTCC technology and is composed of five thermal exchange plates in cross-flow arrangement and two connecting plates, with all plates stacked to form a device with external dimensions of 26 x 26 x 6 mm(3).

Project and Development of ErgoCoIn Version 2.0

Usability evaluation (UE) methods may have several purposes: identify and diagnose usability problems; evaluate the implementation, comparing the one under evaluation with other systems, and also perform acceptance test. These methods are usually expensive, time-consuming and complex. In this context, the ErgoCoIn 1.0 has being used as an accessible system that can be used by several people all over the world, even if they are not usability experts or evaluators. This paper aims to present a new version of this environment considering that its basic objective is to propose a higher quality for the questions. Additionally, a new presentation questionnaire window small enough to be used both in web browsers and mobile devices is proposed.