Time filter

Source Type

Ituiutaba, Brazil

Tambasco Bruno D.O.,Federal University of ABC | Do Nascimento M.Z.,Federal University of ABC | Do Nascimento M.Z.,Federal University of Uberlandia | Batista V.R.,Federal University of ABC | And 2 more authors.
Expert Systems with Applications | Year: 2016

In computer-aided diagnosis one of the crucial steps to classify suspicious lesions is the extraction of features. Texture analysis methods have been used in the analysis and interpretation of medical images. In this work we present a method based on the association among curvelet transform, local binary patterns, feature selection by statistical analysis and distinct classification methods, in order to support the development of computer aided diagnosis system. The similar features were removed by the statistical analysis of variance (ANOVA). The understanding of the features was evaluated by applying the decision tree, random forest, support vector machine and polynomial (PL) classifiers, considering the metrics accuracy (AC) and area under the ROC curve (AUC): the rates were calculated on images of breast tissues with different physical properties (commonly observed in clinical practice). The datasets were the Digital Database for Screening Mammography, Breast Cancer Digital Repository and UCSB biosegmentation benchmark. The investigated groups were normal-abnormal and benign-malignant. The association of curvelet transform, local binary pattern and ANOVA with the PL classifier achieved higher AUC and AC values for all cases: the obtained rates were among 91% and 100%. These results are relevant, specially when we consider the difficulties of clinical practice in distinguishing the studied groups. The proposed association is useful as an automated protocol for the diagnosis of breast tissues and may contribute to the diagnosis of breast tissues (mammographic and histopathological images). © 2016 Elsevier Ltd. All rights reserved. Source

Torres J.L.R.,Federal Rural University of Rio de Janeiro | Santana M.J.,IFTM | Neto A.P.,Brazilian National Council for Scientific and Technological Development | Pereira M.G.,Brazilian National Council for Scientific and Technological Development | Vieira D.M.S.,Brazilian National Council for Scientific and Technological Development
Bioscience Journal | Year: 2013

The use of irrigation in the bean crop in the winter, grown in no tillage, are alternatives used to increase crop productivity, reduce evaporation and increase water retention capacity of the soil. This study aimed to evaluate the productivity of irrigated beans and biomass production of cover crops in Uberaba-MG. The experimental design was a randomized block in a 4×5×3 factorial, which included four types of cover crops (sunn hemp, bean-to-pig, millet and fallow), five irrigation (40%, 70%, 100%, 130% and 160% of daily crop evapotranspiration) and three repetitions. Were evaluated the dry mass of the roof at the point of maximum flowering, productivity, number of pods per plant and grains of the bean crop grown on the crop residues left by these coverages. The precipitation decisively influenced the production of dry biomass of the tested cover.The bean yield was higher when grown on millet and beans on the lower in the bean-to-pig area. Source

Castoldi R.,Sao Paulo State University | Charlo H.C.O.,IFTM | Melo D.M.,Sao Paulo State University | Candido W.S.,Sao Paulo State University | And 3 more authors.
Horticultura Brasileira | Year: 2014

Lettuce is the most consumed leafy vegetable in Brazil. Its production, nonetheless, is increasing in difficulties mainly due to the infestation of producing areas with Bremia lactucae. Thus, the use of horizontal resistant cultivars is the most viable alternative in controlling progress of disease. So, the objective of this study was to obtain resistant lettuce progenies to the races of downy mildew: SPBl:01, SPBl:02, SPBl:03, SPBl:04, SPBl:05, SPBl:06 and SPBl:07. The breeding consisted of two stages: crossover of parentals to obtain the resistant lettuce progenies and endurance test of the progeny to B. lactucae races. The parentals used to obtain progenies with resistance factor R-18 and R-38 were: JAB 4-13-7 and JAB 4-13-7. The Pedigree method was used for obtaining the progeny, using as default selections the Hortência cultivar and JAB 4-13-7genotype. After selection and selfing of the plants in the field, the resistance or susceptibility test was realized, by inoculation of the progenies from the crosses, with mixture of distilled water + sporangia of B. lactucae races SPBl:01, SPBl:02, SPBl:03, SPBl:04, SPBl:05, SPBl:06 and SPBl:07 obtained from isolates collected in the years 2008-2010. Fifteen days after inoculation, the seedlings were selected as resistant or susceptible, discarding those presenting sporulation and necrotic spots caused by B. lactucae. Through the Pedigree method we found 69 progenies F3 with good agronomic characteristics. However, after resistance or susceptibility test, only 19 showed all the plants resistant to downy mildew. Source

Castoldi R.,Sao Paulo State University | Gomes R.F.,Sao Paulo State University | Charlo H.C.O.,IFTM | Melo D.M.,Sao Paulo State University | Braz L.T.,Sao Paulo State University
Horticultura Brasileira | Year: 2014

The objective of this work was to evaluate the performance of three Japanese group cucumber hybrids, which were cultivated in two cultivation systems. The experiment was designed in a split-plot randomized block design with two cultivation systems (coconut fiber and soil) for commercial Japanese cucumber hybrids ('Tsuyataro', 'Yoshinari' and 'Nankyoku'). After harvesting the cucumbers, we evaluated the number of marketable fruits per plant, the average fruit length, the mean fruit diameter, the bottom fruit diameter, the marketable production of fruits per plant and the marketable yield per hectare. There were significant interactions between the cultivation system and the hybrid, as indicated by the number of marketable fruits. The Nankyoku hybrid had the highest average (14.54 fruits pl-1), although it did not differ from the Yoshinari hybrid when grown in coconut fiber. In soil culture, the Yoshinari hybrid had the highest average number of fruits per plant (10.12 fruits pl-1) and did not differ from the Tsuyataro hybrid. Cultivation in coconut fiber provided better results for production traits and plant productivity. 'Yoshinari' and 'Nankyoku' were the most productive hybrids. Based on the cucumber cultivation results from a protected environment, the cultivation of hybrid Japanese cucumbers and Yoshinari and Nankyoku hybrids in coconut fiber is recommended. Source

Charlo H.C.O.,IFTM | de Oliveira S.F.,Sao Paulo State University | Vargas P.F.,Sao Paulo State University | Castoldi R.,Sao Paulo State University | And 2 more authors.
Horticultura Brasileira | Year: 2012

The production of sweet peppers in greenhouse has been severely affected by soil pathogens. With this, producers are adopting new cultivation techniques, such as cultivation in substrates, but information about this type of cultivation are still scarce. The aim of the present work was to evaluate nutrient accumulation in sweet peppers cultivated in coconut fiber with fertirrigation in greenhouse. Initially, 160 plants (Eppo cultivar) were divided into four blocks, where two plants per block were analyzed every 21 days after transplanting. The sweet peppers were cultivated in plastic pots of 13 L, containing coconut fiber, and placed in double rows with spacing of 0.5 × 0.8 m between single rows and 1.10 m between double rows. The commercial production of the mature fruits was estimated to be 97.3 t ha-1, where all the production was classified as Extra (maximum quality). The accumulation of the various determined nutrients were: 8.22 g plant-1 of N, 1.14 g plant-1 of P, 7.84 g plant-1 of K, 3.25 g plant-1 of Ca, 1.34 g plant-1 of Mg, 2.24 g plant-1 of S, 16.65 mg plant-1 of B, 3.36 mg plant-1 of Cu, 45.98 mg plant-1 of Fe, 34.78 mg plant-1 of Mn and 22.28 mg plant-1 of Zn. Source

Discover hidden collaborations