Classification of Multiple H&E Images via an Ensemble Computational Scheme.

In this work, a computational scheme is proposed to identify the main combinations of handcrafted descriptors and deep-learned features capable of classifying histological images stained with hematoxylin and eosin. The handcrafted descriptors were those representatives of multiscale and multidimensional fractal techniques (fractal dimension, lacunarity and percolation) applied to quantify the histological images with the corresponding representations via explainable artificial intelligence (xAI) approaches. The deep-learned features were obtained from different convolutional neural networks (DenseNet-121, EfficientNet-b2, Inception-V3, ResNet-50 and VGG-19). The descriptors were investigated through different associations. The most relevant combinations, defined through a ranking algorithm, were analyzed via a heterogeneous ensemble of classifiers with the support vector machine, naive Bayes, random forest and K-nearest neighbors algorithms. The proposed scheme was applied to histological samples representative of breast cancer, colorectal cancer, oral dysplasia and liver tissue. The best results were accuracy rates of 94.83% to 100%, with the identification of pattern ensembles for classifying multiple histological images. The computational scheme indicated solutions exploring a reduced number of features (a maximum of 25 descriptors) and with better performance values than those observed in the literature. The presented information in this study is useful to complement and improve the development of computer-aided diagnosis focused on histological images.

Diazotrophic bacteria improve Hymenaea courbaril seedlings growth and survival in iron mine tailings.

One of the largest accidents with mine tailings happened in Brazil in 2015, with the rupture of the Fundão dam, and the physical characteristics of these tailings make it difficult to recover degraded areas. Hymenaea courbaril is a tree species native to Brazil that has low nutritional and water requirements, besides its capacity for survival in contaminated environments. In this study we hypothesized that inoculation with diazotrophs would improve the growth and physiology of H. courbaril in tailings, favoring the reforestation process aiming the recovery of the accident site. Every 20 days for 60 days, we investigated the morphophysiology of H. courbaril grown in iron mine tailings or soil, with the addition of nitrate (N-positive control), non-inoculation (negative control) or inoculation with native diazotrophic bacteria previously isolated from the tailings (UNIFENAS100-569; UNIFENAS100-654 and UNIFENAS100-638). We found that H. courbaril has survival capacity under mine tailings, with no growth alteration in the tailings, although there were signs of reduced ability for photoprotective responses. Inoculation with diazotrophic bacteria improved physiological aspects of H. courbaril and strain UNIFENAS100-638 was the most effective in favoring total growth of plants, net photosynthetic rate and root morphology under mine tailings. The survival capacity and growth of H. courbaril indicates the possibility of its use for reforestation in areas degraded by mine tailings. Further studies are necessary in field conditions and with a larger experimental period to more thoroughly understand H. courbaril tolerance.