84 resultados para Classifier Generalization Ability


Relevância:

20.00% 20.00%

Publicador:

Resumo:

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.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

We consider in this work the electromagnetic current for a system composed of two charged bosons and show that it has a structure of many bodies even in the impulse approximation, when described in the light-front time x+. In terms of the two-body component for the bound state, the current contains two-body operators. We consider the photon interacting with two bosons and the process of pair creation connected to this interaction, interpreting it as a zero mode contribution to the current and discuss the consequences of this pair creation to the components of currents in the light front. © 2013 American Physical Society.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

Pós-graduação em Engenharia Mecânica - FEG

Relevância:

20.00% 20.00%

Publicador:

Resumo:

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

Relevância:

20.00% 20.00%

Publicador:

Resumo:

Three culture media (Brucella agar, Farrell medium, and CITA) were compared for their effectiveness in inhibiting contamination and for isolating Brucella spp. One hundred lymph nodes from pigs (n = 50) and wild boars (n = 50) with lymphadenitis were collected in slaughterhouses in the State of Sao Paulo and were assessed on these three selective media for Brucella spp. All of the samples were negative for Brucella spp. on the three culture media. On the agar medium, fungal (70 plates) and Gram-positive bacterial (59 plates) contaminants were observed; in the CITA medium, the absence of fungal and Gram-positive bacteria on 15 plates was observed; no bacterial or fungal growth was observed on the Farrell media. The results demonstrated that the CITA and Farrell media inhibited the growth of contaminants better than the Brucella agar.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

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

Relevância:

20.00% 20.00%

Publicador:

Resumo:

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Relevância:

20.00% 20.00%

Publicador:

Resumo:

In general, pattern recognition techniques require a high computational burden for learning the discriminating functions that are responsible to separate samples from distinct classes. As such, there are several studies that make effort to employ machine learning algorithms in the context of big data classification problems. The research on this area ranges from Graphics Processing Units-based implementations to mathematical optimizations, being the main drawback of the former approaches to be dependent on the graphic video card. Here, we propose an architecture-independent optimization approach for the optimum-path forest (OPF) classifier, that is designed using a theoretical formulation that relates the minimum spanning tree with the minimum spanning forest generated by the OPF over the training dataset. The experiments have shown that the approach proposed can be faster than the traditional one in five public datasets, being also as accurate as the original OPF. (C) 2014 Elsevier B. V. All rights reserved.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

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

Relevância:

20.00% 20.00%

Publicador:

Resumo:

In this letter, a speech recognition algorithm based on the least-squares method is presented. Particularly, the intention is to exemplify how such a traditional numerical technique can be applied to solve a signal processing problem that is usually treated by using more elaborated formulations.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

This article presents some results of a study investigating the use of the simulator VLAB, developed by Carnegie Mellon University, to increase the capacity of graduate students in Chemistry in the preparation of experimental procedures. The activity was conducted in two stages, first the students were asked a manuscript describing in detail the procedure and equipment used in the technique of dissolution and delivering it to the teacher. Later in the computer lab, each student performed the same procedure using the simulator. The activity has been proposed to determine if there could be accuracy gains of the description of the process when performed by the simulator. One of the main positive aspects obtained using the simulator students was the recognition of error, important in that the great majority recognized omitted or confused any process step. This reflection showed the students that in addition to testing a procedure using the simulator facilitates the organization of thought and helps arrange it logically. With the use of simulators, students are encouraged to take risks, to explore, to experience a new way, and a real lab usually offers much less room for risk, exploitation and failure.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

Nickel-titanium rotatory systems should remove dentin during the root canal preparation to promote the enlargement and tapered shape with continuously narrowing towards the apex. Objective: The aim of this study was to analyze the cutting ability of three NiTi rotatory systems (n = 12): ProTaper (group 1), Mtwo (group 2), and K3 (group 3). Material and methods: Thirty six maxillary molar teeth were weighted on an analytical balance before and after the rotatory preparation of the mesio-buccal root canal. Data was statistically analyzed by ANOVA and Tukey s test with a significance level of 5%. Results: The results revealed the following mass differences (g) before and after the root canal preparation: ProTaper (group 1 – 0.0159 ± 0.004), Mtwo (group 2 – 0.0125 ± 0.002), and K3 (group 3 – 0.007 ± 0.003). Conclusion: ProTaper showed the highest cutting ability among the three tested nickel-titanium rotatory systems followed by Mtwo and K3.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

The Mineral Trioxide Aggregate (MTA) has excellent biological property. However, its consistency makes it difficult to be inserted into retrograde cavities. Objective:To evaluate the ability of different methods to fill retrograde cavities with MTA. Material and methods: Root canals of thirty single-rooted resin teeth were prepared and filled. After the cut of 3 mm short of apical third, retrograde cavities with 3 mm deep were prepared using an ultrasound device and retrotips (CVD, São José dos Campos, SP, Brazil). The retrograde preparation was evaluate by using an operative microscope (D.F. Vasconcellos, São Paulo, SP, Brazil). The teeth were randomly divided into three groups (n = 10), according to the method: 1) condenser (Trinity, São Paulo, SP, Brazil); 2) MTA applicator (Angelus, Londrina, Brazil) + condenser; 3) condenser associated with ultrasound (CVD, São José dos Campos, SP, Brazil). After the filling of retrograde cavities with white MTA (Angelus, Londrina, Brazil), teeth were radiographed using a digital system (Kodak RVG 6000, Rochester, NY, USA). The images were analyzed by UTHSCSA Image Tool 3.0 software. The percentage of filling was calculated by the proportion between the total area of retrograde cavity and the filled area. The radiographic density mean of each third of retrograde cavity filled with MTA was measured by using the histogram tool of the software. The results were submitted to ANOVA and Tukey tests, with 5% of significance. Results: There was no difference in percentage of filling among the groups (p > 0.05) (approximately 85%). By comparing the thirds, the condenser and MTA applicator groups showed higher density for apical and middle third than cervical third (p < 0.05). The ultrasound group presented similar density among the thirds. Conclusion: The filling ability was similar for the studied methods. Ultrasound promoted better distribution of MTA in retrograde cavity, but did not increase the density of material.