This paper presents a new supervised method for vessel segmentation in retinal images. This method remolds the task of segmentation as a problem of cross-modality data transformation from retinal image to vessel map. A wide and deep neural network with strong induction ability is proposed to model the transformation, and an efficient training strategy is presented. Instead of a single label of the center pixel, the network can output the label map of all pixels for a given image patch. Our approach outperforms reported state-of-the-art methods in terms of sensitivity, specificity and accuracy. The result of cross-training evaluation indicates its robustness to the training set. The approach needs no artificially designed feature and no preprocessing step, reducing the impact of subjective factors. The proposed method has the potential for application in image diagnosis of ophthalmologic diseases, and it may provide a new, general, high-performance computing framework for image segmentation.

The means through which microbes and plants contribute to soil organic carbon (SOC) accumulation remain elusive due to challenges in disentangling the complex components of SOC. Here we use amino sugars and lignin phenols as tracers for microbial necromass and plant lignin components, respectively, and investigate their distribution in the surface soils across Mongolian grasslands in comparison with published data for other grassland soils of the world. While lignin phenols decrease, amino sugars increase with SOC contents in all examined grassland soils, providing continental-scale evidence for the key role of microbial necromass in SOC accumulation. Moreover, in contrast to clay’s control on amino sugar accumulation in fine-textured soils, aridity plays a central role in amino sugar accrual and lignin decomposition in the coarse-textured Mongolian soils. Hence, aridity shifts may have differential impacts on microbial-mediated SOC accumulation in grassland soils of varied textures.

No abstract