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.

Quantificação de fibrose e enfisema em pacientes com paracoccidioidomicose (PCM) pulmonar, através de imagens de tomografia computadorizada de alta resolução (TCAR)

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Mudanças morfológicas na evolução do megaleque do Taquari: uma análise com base em dados orbitais

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

Avaliação da influência de diferentes materiais oclusais na confecção de próteses fixas implantossuportadas unitárias com diferentes diâmetros dos implantes osseointegrados: estudo pelo método dos elementos finitos tridimensionais

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

Influência de diferentes pilares e materiais de recobrimento estético na manutenção da pré-carga do parafuso de retenção de coroas implantosuportadas

Pós-graduação em Odontologia - FOA

Avaliação da adaptação marginal de restaurações cerâmicas inlay após a cimentação

Pós-graduação em Odontologia Restauradora - ICT

Método para extração e caracterização de lesões de pele usando difusão anisotrópica, crescimento de regiões, watersheds e contornos ativos

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Geotecnologias digitais no ensino médio: avaliação prática de seu potencial

Pós-graduação em Geografia - IGCE

Estudo das características morfométricas de diferentes regiões do intestino delgado e índices zootécnicos em quatro linhagens de frangos de corte

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

Desenvolvimento e avaliação de um sistema informatizado de instrução para a formação de conceitos

Ensino Programado, Máquinas de Ensinar e o Sistema Personalizado de Instrução (PSI - Personalized System of Instruction) são abordagens comportamentais para o ensino que orientaram a criação de um software instrucional (SII) para o ensino de conceitos de esquemas de reforçamento baseado em tarefas de escolha, com o qual se avaliou duas condições: escolhas entre definições e entre exemplos, com estudantes experientes e ingênuos. A estruturação das tarefas considerou os desempenhos como treino de intraverbais representativos de conceitos. Não ocorreram diferenças marcantes no desempenho entre as condições; estudantes ingênuos e não ingênuos se beneficiaram igualmente nas duas condições (conceitos e exemplos). Ocorreram indícios de transferência de aprendizagem entre a tarefa de escolha e uma tarefa classificatória complementar. A comparabilidade entre as condições complexas usadas, questões de múltipla escolha entre exemplos ou definições de conceitos deve ser abordada cautelosamente. Os erros concentrados no primeiro bloco de questões de cada conceito indicou que as relações modelo-comparação se transferiam para as questões apresentadas nos blocos seguintes. Variações paramétricas em estudos futuros, entretanto, poderão gerar mais evidências de variáveis favorecedoras da aprendizagem de conceitos em tarefas de escolha em computador.

Development of a CAD system for automatic classification of microcalcifications based on FPGA

This paper presents a Computer Aided Diagnosis (CAD) system that automatically classifies microcalcifications detected on digital mammograms into one of the five types proposed by Michele Le Gal, a classification scheme that allows radiologists to determine whether a breast tumor is malignant or not without the need for surgeries. The developed system uses a combination of wavelets and Artificial Neural Networks (ANN) and is executed on an Altera DE2-115 Development Kit, a kit containing a Field-Programmable Gate Array (FPGA) that allows the system to be smaller, cheaper and more energy efficient. Results have shown that the system was able to correctly classify 96.67% of test samples, which can be used as a second opinion by radiologists in breast cancer early diagnosis. (C) 2013 The Authors. Published by Elsevier B.V.

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.

Modelagem computacional das túnicas cardíacas para processamento e controle de qualidade

This paper considers a study of the anatomical features of the cardiac system and a three-dimensional model of the different tunics that comprise the heart wall, for processing and quality control of radiological images. The structures are built by the layer overlapping method, where a layer can be understood as a slice of the three-dimensional object. The pericardium, myocardium and endocardium were represented with three-dimensional cylinders and hexagons. The spatial arrangement of the cardiac system is determined by an background image of a real model, which values are defined according to the shape of the region and on the anatomical patients characteristics. The results are significant, considering the anatomical structures details, as well as the representation of the thicknesses of the regions of the heart wall. The validation of the anatomical model was accomplished through comparisons with dimensions obtained from a real model and allows verifying that the model is appropriate. The degree of representation will allow the verification of the influence of radiological parameters, morphometric peculiarities and stage of the diseases on the quality of the images, as well as on the performance of the Computer-Aided Diagnosis (CAD).

Influência de tratamentos de superfície e agentes cimentantes na sobrevivência de coroas de Y-TZP recobertas por porcelana

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

Inlays Made From a Hybrid Material: Adaptation and Bond Strengths

The aim of this study was to evaluate the internal fit, marginal adaptation, and bond strengths of inlays made of computer-aided design/computer-aided manufacturing feldspathic ceramic and polymer-infiltrated ceramic. Twenty molars were randomly selected and prepared to receive inlays that were milled from both materials. Before cementation, internal fit was achieved using the replica technique by molding the internal surface with addition silicone and measuring the cement thicknesses of the pulpal and axial walls. Marginal adaptation was measured on the occlusal and proximal margins of the replica. The inlays were then cemented using resin cement (Panavia F2.0) and subjected to two million thermomechanical cycles in water (200 N load and 3.8-Hz frequency). The restored teeth were then cut into beams, using a lathe, for microtensile testing. The contact angles, marginal integrity, and surface patterns after etching were also observed. Statistical analysis was performed using two-way repeated measures analysis of variance (p<0.05), the Tukey test for internal fit and marginal adaptation, and the Student t-test for bond strength. The failure types (adhesive or cohesive) were classified on each fractured beam. The results showed that the misfit of the pulpal walls (p=0.0002) and the marginal adaptation (p=0.0001) of the feldspathic ceramic were significantly higher when compared to those of the polymer-infiltrated ceramic, while the bond strength values of the former were higher when compared to those of the latter. The contact angle of the polymer-infiltrated ceramic was also higher. In the present study, the hybrid ceramic presented improved internal and marginal adaptation, but the bond strengths were higher for the feldspathic ceramic.