Advancing Bag-of-visual-words Representations For Lesion Classification In Retinal Images.

Diabetic Retinopathy (DR) is a complication of diabetes that can lead to blindness if not readily discovered. Automated screening algorithms have the potential to improve identification of patients who need further medical attention. However, the identification of lesions must be accurate to be useful for clinical application. The bag-of-visual-words (BoVW) algorithm employs a maximum-margin classifier in a flexible framework that is able to detect the most common DR-related lesions such as microaneurysms, cotton-wool spots and hard exudates. BoVW allows to bypass the need for pre- and post-processing of the retinographic images, as well as the need of specific ad hoc techniques for identification of each type of lesion. An extensive evaluation of the BoVW model, using three large retinograph datasets (DR1, DR2 and Messidor) with different resolution and collected by different healthcare personnel, was performed. The results demonstrate that the BoVW classification approach can identify different lesions within an image without having to utilize different algorithms for each lesion reducing processing time and providing a more flexible diagnostic system. Our BoVW scheme is based on sparse low-level feature detection with a Speeded-Up Robust Features (SURF) local descriptor, and mid-level features based on semi-soft coding with max pooling. The best BoVW representation for retinal image classification was an area under the receiver operating characteristic curve (AUC-ROC) of 97.8% (exudates) and 93.5% (red lesions), applying a cross-dataset validation protocol. To assess the accuracy for detecting cases that require referral within one year, the sparse extraction technique associated with semi-soft coding and max pooling obtained an AUC of 94.2 ± 2.0%, outperforming current methods. Those results indicate that, for retinal image classification tasks in clinical practice, BoVW is equal and, in some instances, surpasses results obtained using dense detection (widely believed to be the best choice in many vision problems) for the low-level descriptors.

Teores de nutrientes na alface irrigada com água residuária aplicada por sistemas de irrigação

A trial was carried out to evaluate the chemical composition in the aerial part of lettuce, cv. 'Elisa', irrigated with wastewater treated with constructed wetland and source deposit water, grown on a Rhodic Hapludox Soil, using the irrigation systems sprinkle, subsurface drip and surface drip irrigation. The experiment was carried out from August 17th to October 3rd of 2001 and the chemical analyses of the lettuce were accomplished to 47 days after transplanting of the seedling. The aerial part of the lettuce was analyzed as for the levels of total nitrogen, nitrate, phosphorus, potassium, calcium, magnesium, sulfur, iron, manganese, copper, zinc, sodium, boron, cobalt and molybdenum. The sodium and the sulfur presented higher levels than the maximum suitable in the aerial part of the lettuce and the smallest level of magnesium, while other chemical elements analyzed were normal and appropriate considering the standard for well-nourished plants, not being influenced by the water type. The sodium was the chemical element that presented the highest levels in the aerial part of the lettuce in the treatments irrigated with wastewater, presenting significant difference in relationship to the treatments irrigated with source deposit water in the three irrigation systems. The use of the different irrigation systems by the application of wastewater treated with constructed wetland did not interfere in the levels of nutrients in the aerial part of the lettuce.

Physiological limitations in two sugarcane varieties under water suppression and after recovering

Increasing water scarcity and depleted water productivity in irrigated soils are inducing farmers to adopt improved varieties, such as those with high-capacity tolerance. The use of tolerant varieties of sugarcane might substantially avoid the decline of productivity under water deficit. This research aimed to evaluate the harmful effects of drought on the physiology of two sugarcane varieties (RB867515 and RB962962) during the initial development. Young plants were subjected to irrigation suspension until total stomata closure, and then rewatered. Significant reduction on stomatal conductance, transpiration, and net photosynthesis were observed. RB867515 showed a faster stomatal closure while RB962962 slowed the effects of drought on the gas exchanges parameters with a faster recovering after rewatering. Accumulation of carbohydrates, amino acids, proline, and protein in the leaves and roots of the stressed plants occurred in both varieties, substantially linked to reduction of the leaf water potential. Due to the severity of stress, this accumulation was not enough to maintain the cell turgor pressure, so relative water content was diminished. Water stress affected the contents of chlorophyll (a, b, and total) in both varieties, but not the levels of carotenoids. There was a significant reduction in dry matter under stress. In conclusion, RB962962 variety endured stressed conditions more than RB867515, since it slowed down the damaging effects of drought on the gas exchanges. In addition, RB962962 presented a faster recovery than RB867515, a feature that qualifies it as a variety capable of enduring short periods of drought without major losses in the initial stage of its development.