Learn More
This paper introduces a novel rotation-based framework for arbitrary-oriented text detection in natural scene images. We present the Rotation Region Proposal Networks (RRPN), which is designed to generate inclined proposals with text orientation angle information. The angle information is then adapted for bounding box regression to make the proposals more(More)
miR-205 is an epithelial-specific miRNA and has been shown to orchestrate some cellular processes such as epithelial mesenchymal transition (EMT) and differentiation fate of stem cells in mammary gland. miR-205 play a part of a tumor suppressor in human cancers. However, the role of miR-205 in lung cancer is unclear. In this study, we detected the(More)
In this paper, we introduce an active annotation and learning framework for the face recognition task. Starting with an initial label deficient face image training set, we iteratively train a deep neural network and use this model to choose the examples for further manual annotation. We follow the active learning strategy and derive the Value of Information(More)
OBJECTIVE This study aims to observe the protective effects of heparin on endothelial cells in sepsis and explore the involved signal pathway regulated by heparin. Methods Human vascular endothelial cells were treated by TNFα in vitro to simulate the inflammatory environment when sepsis occurred. They were intervened by heparin and the expression levels of(More)
We compared the outcomes in patients who were <1 year old and had hydronephrosis with SFU grade 3–4 PUJ obstruction to observe the potential recovery of renal morphology and DRF after successful pyeloplasty. All children younger than 1 year old with SFU grade 3–4 PUJ obstruction from January 2013 to June 2015 were retrospectively analyzed. A total of 92(More)
  • 1