Hiroshima University

About: Hiroshima University is a education organization based out in Hiroshima, Japan. It is known for research contribution in the topics: Population & Cancer. The organization has 33602 authors who have published 69290 publications receiving 1495648 citations. The organization is also known as: Hiroshima Daigaku.

Defect Motion on an InP(110) Surface Observed with Noncontact Atomic Force Microscopy

Abstract: With an atomic force microscope operating in the noncontact mode in an ultrahigh vacuum, atomic-resolution imaging of the cleaved semi-insulating InP(110) surface has been achieved. By this method, atomic scale point defects and their motion were observed at room temperature, without the field-induced effects associated with scanning tunneling microscopy.

Deep Learning Models for Histopathological Classification of Gastric and Colonic Epithelial Tumours

Abstract: Histopathological classification of gastric and colonic epithelial tumours is one of the routine pathological diagnosis tasks for pathologists. Computational pathology techniques based on Artificial intelligence (AI) would be of high benefit in easing the ever increasing workloads on pathologists, especially in regions that have shortages in access to pathological diagnosis services. In this study, we trained convolutional neural networks (CNNs) and recurrent neural networks (RNNs) on biopsy histopathology whole-slide images (WSIs) of stomach and colon. The models were trained to classify WSI into adenocarcinoma, adenoma, and non-neoplastic. We evaluated our models on three independent test sets each, achieving area under the curves (AUCs) up to 0.97 and 0.99 for gastric adenocarcinoma and adenoma, respectively, and 0.96 and 0.99 for colonic adenocarcinoma and adenoma respectively. The results demonstrate the generalisation ability of our models and the high promising potential of deployment in a practical histopathological diagnostic workflow system.

The superoxide-producing NAD(P)H oxidase Nox4 in the nucleus of human vascular endothelial cells

Abstract: The superoxide-producing NAD(P)H oxidase Nox4 was initially identified as an enzyme that is highly expressed in the kidney and is possibly involved in oxygen sensing and cellular senescence. Although the oxidase is also abundant in vascular endothelial cells, its role remains to be elucidated. Here we show that Nox4 preferentially localizes to the nucleus of human umbilical vein endothelial cells (HUVECs), by immunocytochemistry and immunoelectron microscopy using three kinds of affinity-purified antibodies raised against distinct immunogens from human Nox4. Silencing of Nox4 by RNA interference (RNAi) abrogates nuclear signals given with the antibodies, confirming the nuclear localization of Nox4. The nuclear fraction of HUVECs exhibits an NAD(P)H-dependent superoxide-producing activity in a manner dependent on Nox4, which activity can be enhanced upon cell stimulation with phorbol 12-myristate 13-acetate. This stimulant also facilitates gene expression as estimated in the present transfection assay of HUVECs using a reporter regulated by the Maf-recognition element MARE, a DNA sequence that constitutes a part of oxidative stress response. Both basal and stimulated transcriptional activities are impaired by RNAi-mediated Nox4 silencing. Thus Nox4 appears to produce superoxide in the nucleus of HUVECs, thereby regulating gene expression via a mechanism for oxidative stress response.

Biological nitrogen fixation in mixed legume-cereal cropping systems

Abstract: Cereal/legume intercropping increases dry matter production and grain yield more than their monocultures. When fertilizer N is limited, biological nitrogen fixation (BNF) is the major source of N in legume-cereal mixed cropping systems. The soil N use patterns of component crops depend on the N source and legume species. Nitrogen transfer from legume to cereal increases the cropping system's yield and efficiency of N use. The use of nitrate-tolerant legumes, whose BNF is thought to be little affected by application of combined N, may increase the quantity of N available for the cereal component. The distance between the cereal and legume root systems is important because N is transferred through the intermingling of root systems. Consequently, the most effective planting distance varies with type of legume and cereal. Mutual shading by component crops, especially the taller cereals, reduces BNF and yield of the associated legume. Light interception by the legume can be improved by selecting a suitable plant type and architecture. Planting pattern and population at which maximum yield is achieved also vary among component species and environments. Crops can be mixed in different proportions from additive to replacement or substitution mixtures. At an ideal population ratio a semi-additive mixture may produce higher gross returns.