Rationale: Pericytes are key regulators of vascular maturation, but their value for cardiac repair remains unknown. Objective: We investigated the therapeutic activity and mechanistic targets of saphenous vein-derived pericyte progenitor cells (SVPs) in a mouse myocardial infarction (MI) model. Methods and Results: SVPs have a low immunogenic profile and are resistant to hypoxia/starvation (H/S).Transplantation of SVPs into the peri-infarct zone of immunodeficient CD1/Foxn-1 nu/nu or immunocompetent CD1 mice attenuated left ventricular dilatation and improved ejection fraction compared to vehicle. Moreover, SVPs reduced myocardial scar, cardiomyocyte apoptosis and interstitial fibrosis, improved myocardial blood flow and neovascularization, and attenuated vascular permeability. SVPs secrete vascular endothelial growth factor A, angiopoietin-1, and chemokines and induce an endogenous angiocrine response by the host, through recruitment of vascular endothelial growth factor B expressing monocytes. The association of donor-and recipient-derived stimuli activates the proangiogenic and prosurvival Akt/eNOS/Bcl-2 signaling pathway. Moreover, microRNA-132 (miR-132) was constitutively expressed and secreted by SVPs and remarkably upregulated, together with its transcriptional activator cyclic AMP response element-binding protein, on stimulation by H/S or vascular endothelial growth factor B. We next investigated if SVP-secreted miR-132 acts as a paracrine activator of cardiac healing. In vitro studies showed that SVP conditioned medium stimulates endothelial tube formation and reduces myofibroblast differentiation, through inhibition of Ras-GTPase activating protein and methyl-CpG-binding protein 2, which are validated miR-132 targets. Furthermore, miR-132 inhibition by antimiR-132 decreased SVP capacity to improve contractility, reparative angiogenesis, and interstitial fibrosis in infarcted hearts. Key Words: pericytes-based cell therapy Ⅲ myocardial infarction Ⅲ angiogenesis Ⅲ VEGF-B Ⅲ microRNA-132 W ith myocardial infarction (MI) remaining a major cause of morbidity and mortality worldwide, cell therapy now aims to offer a novel option for cardiac repair. 1 Clinical trials showed that administration of bone marrowderived progenitor cells (PCs) improves left ventricular (LV) function in patients with coronary artery disease. 2-4 However, more specialized cells are warranted to fulfill specific regenerative needs of the ischemic myocardium. ConclusionPericytes provide the physical strength and nurturing signals that instruct neovessels to organize in a stable and efficient tubular network. 5 On the other hand, ischemic disease and associated risk factors may impair pericyte recruitment. 6 -8 Therefore, a supply-side approach with fresh pericytes from exogenous sources could be helpful therapeutically. However, difficulties in isolating and expanding bona-fide pericytes from accessible human tissues have so far precluded clinical applications.Two main mural cell populations, probably originating from a common emb...

This paper focuses on bearing fault diagnosis with limited training data. A major challenge in fault diagnosis is the infeasibility of obtaining sufficient training samples for every fault type under all working conditions. Recently deep learning based fault diagnosis methods have achieved promising results. However, most of these methods require large amount of training data. In this study, we propose a deep neural network based few-shot learning approach for rolling bearing fault diagnosis with limited data. Our model is based on the siamese neural network, which learns by exploiting sample pairs of the same or different categories. Experimental results over the standard Case Western Reserve University (CWRU) bearing fault diagnosis benchmark dataset showed that our few-shot learning approach is more effective in fault diagnosis with limited data availability. When tested over different noise environments with minimal amount of training data, the performance of our few-shot learning model surpasses the one of the baseline with reasonable noise level. When evaluated over test sets with new fault types or new working conditions, few-shot models work better than the baseline trained with all fault types. All our models and datasets in this study are open sourced and can be downloaded from https://mekhub.cn/as/fault_diagnosis_with_few-shot_learning/. INDEX TERMS Deep learning, few-shot learning, bearing fault diagnosis, limited data.

No abstract

Abstract. Human osteopontin (OPN) is a glycosylated phosphoprotein which is expressed in a variety of tissues in the body. In recent years, accumulating evidence has indicated that the aberrant expression of OPN is closely associated with tumourigensis, progression and most prominently with metastasis in several tumour types. In this review, we present the current knowledge on the expression profiles of OPN and its main splice variants in human cancers, as well as the potential implications in patient outcome. We also discuss its putative clinical application as a cancer biomarker and as a therapeutic target. IntroductionOsteopontin (OPN) is a bone associated, extracellular matrix glycosylated phosphoprotein which is produced by several cell types, including osteoblasts, osteoclasts, immune cells, endothelial cells, epithelial cells and extra-osseous cells (skin, kidney and lung) (1-3). Due to differences in post-translational modification (PTM) (phosphorylation, glycosylation, sulfation and proteolysis) from different cellular sources, OPN has a molecular weight ranging from 41 to 75 kDa, which may have a cell type-specific structure and function (4-7). OPN plays a major role in various normal physiological processes, including bone remodelling, immune-regulation, inflammation and vascularisation (8,9). In addition, OPN has also been shown to be involved in carcinogenesis with multi-functional activities (10)(11)(12).OPN is involved in a series of biological functions through interactions with different integrins and CD44. Therefore, OPN is classified as a member of the ʻsmall integrin-binding ligand N-linked glycoproteins' (SIBLINGs) together with other molecules, including bone sialoprotein (BSP), dentin matrix protein 1 (DMP1), dentin sialophosphoprotein (DSPP) and matrix extracellular phosphoglycoprotein (MEPE) (13). Two critical integrin binding sequences of OPN have been identified: arginine-glycine-aspartic acid (RGD) and serine-valine-valinetyrosine-glutamate-leucine-arginine (SVVYGLR). OPN interacts mainly with various αv (particularly αvβ1, αvβ3, αvβ5) integrin receptors via the classical RGD sequence, and interacts with α9β1, α4β1, α4β7 via SVVYGLR (14-16). In addition, it also interacts with the CD44 splice variants, CD44v3, CD44v6 and CD44v7, via the C-terminal fragment calcium binding site (17)(18)(19)(20). These properties of OPN induce the activation of signal transduction pathways, leading to cell proliferation, adhesion, invasion and migration, which have been demonstrated by both in vitro and in vivo models (21-23). The binding of OPN to integrins and CD44 initiates a downstream signalling cascade via the PI3K/AKT signalling pathway leading to NF-κB mediated cell proliferation and survival (24)(25)(26). In additon, through the Ras/Raf/MEK/ERK signalling pathway, an OPN-integrin complex and subsequent induction of AP-1-dependent gene expression, urokinase-type plasminogen activator (uPA) and matrix metalloproteinases (MMPs) confer a metastatic phenotype on some cancer cell types (27)(28)(29...

This paper addresses the task of nuclei segmentation in high-resolution histopathological images. We propose an automatic end-to-end deep neural network algorithm for segmentation of individual nuclei. A nucleus-boundary model is introduced to predict nuclei and their boundaries simultaneously using a fully convolutional neural network. Given a color normalized image, the model directly outputs an estimated nuclei map and a boundary map. A simple, fast and parameter-free post-processing procedure is performed on the estimated nuclei map to produce the final segmented nuclei. An overlapped patch extraction and assembling method is also designed for seamless prediction of nuclei in large whole-slide images. We also show the effectiveness of data augmentation methods for nuclei segmentation task. Our experiments showed our method outperforms prior state-of-theart methods. Moreover, it is efficient that one 1000X1000 image can be segmented in less than 5 seconds. This makes it possible to precisely segment the whole-slide image in acceptable time.