No abstract

Over the recent years, there has been an increasing interest in large-scale classification of remote sensing images. In this context, the Inria Aerial Image Labeling Benchmark has been released online in December 2016. In this paper, we discuss the outcomes of the first year of the benchmark contest, which consisted in dense labeling of aerial images into building / not building classes, covering areas of five cities not present in the training set. We present four methods with the highest numerical accuracies, all four being convolutional neural network approaches. It is remarkable that three of these methods use the U-net architecture, which has thus proven to become a new standard in image dense labeling.

Recently deep learningnamely convolutional neural networks (CNNs)have yielded impressive performance for the task of building segmentation on large overhead (e.g., satellite) imagery benchmarks. However, these benchmark datasets only capture a small fraction of the variability present in real-world overhead imagery, limiting the ability to properly train, or evaluate, models for real-world application. Unfortunately, developing a dataset that captures even a small fraction of real-world variability is typically infeasible due to the cost of imagery, and manual pixel-wise labeling of the imagery. In this work we develop an approach to rapidly and cheaply generate large and diverse virtual environments from which we can capture synthetic overhead imagery for training segmentation CNNs. Using this approach, generate and publicly-release a collection of synthetic overhead imagery termed Synthinel-1 with full pixel-wise building labels. We use several benchmark dataset to demonstrate that Synthinel-1 is consistently beneficial when used to augment real-world training imagery, especially when CNNs are tested on novel geographic locations or conditions.

Osteoporosis deteriorates bone mass and biomechanical strength, becoming a life-threatening cause to the elderly. MicroRNA is known to regulate tissue remodeling; however, its role in the development of osteoporosis remains elusive. In this study, we uncovered that silencing miR-29a expression decreased mineralized matrix production in osteogenic cells, whereas osteoclast differentiation and pit formation were upregulated in bone marrow macrophages as co-incubated with the osteogenic cells in transwell plates. In vivo, decreased miR-29a expression occurred in ovariectomy-mediated osteoporotic skeletons. Mice overexpressing miR-29a in osteoblasts driven by osteocalcin promoter (miR-29aTg/OCN) displayed higher bone mineral density, trabecular volume and mineral acquisition than wild-type mice. The estrogen deficiency-induced loss of bone mass, trabecular morphometry, mechanical properties, mineral accretion and osteogenesis of bone marrow mesenchymal cells were compromised in miR-29aTg/OCN mice. miR-29a overexpression also attenuated the estrogen loss-mediated excessive osteoclast surface histopathology, osteoclast formation of bone marrow macrophages, receptor activator nuclear factor-κ ligand (RANKL) and C–X–C motif chemokine ligand 12 (CXCL12) expression. Treatment with miR-29a precursor improved the ovariectomy-mediated skeletal deterioration and biomechanical property loss. Mechanistically, miR-29a inhibited RANKL secretion in osteoblasts through binding to 3′-UTR of RANKL. It also suppressed the histone acetyltransferase PCAF-mediated acetylation of lysine 27 in histone 3 (H3K27ac) and decreased the H3K27ac enrichment in CXCL12 promoters. Taken together, miR-29a signaling in osteogenic cells protects bone tissue from osteoporosis through repressing osteoclast regulators RANKL and CXCL12 to reduce osteoclastogenic differentiation. Arrays of analyses shed new light on the miR-29a regulation of crosstalk between osteogenic and osteoclastogenic cells. We also highlight that increasing miR-29a function in osteoblasts is beneficial for bone anabolism to fend off estrogen deficiency-induced excessive osteoclastic resorption and osteoporosis.

No abstract

No abstract

No abstract

No abstract