Humans use rich natural language to describe and communicate visual perceptions. In order to provide natural language descriptions for visual content, this paper combines two important ingredients. First, we generate a rich semantic representation of the visual content including e.g. object and activity labels. To predict the semantic representation we learn a CRF to model the relationships between different components of the visual input. And second, we propose to formulate the generation of natural language as a machine translation problem using the semantic representation as source language and the generated sentences as target language. For this we exploit the power of a parallel corpus of videos and textual descriptions and adapt statistical machine translation to translate between our two languages. We evaluate our video descriptions on the TACoS dataset [23], which contains video snippets aligned with sentence descriptions. Using automatic evaluation and human judgments we show significant improvements over several baseline approaches, motivated by prior work. Our translation approach also shows improvements over related work on an image description task.
Signal transducers and activators of transcription-3 (STAT3), a central cytoplasmic transcription factor, is frequently overexpressed and constitutively activated by tyrosine during malignant transformation. The overexpression and phosphorylation of STAT3 in pancreatic cancer has been described only recently, but the roles and mechanism still remain unclear. In this study, we elucidate the significance of the STAT3 signaling pathway in metastatic potentials of pancreatic cancer. We stably silence the expression of the STAT3 and p-STAT3 by using RNA interference (RNAi) in the pancreatic cancer cell line SW1990, and then reduce its invasion capacity in vitro and metastasis capacity in vivo compared to parental cells or cells tansfected with a control vector. Furthermore, silencing SW1990 cells with the STAT3 gene by RNAi also led to a decrease of matrix metalloproteinases-2 (MMP-2) and vascular endothelial growth factor (VEGF) at the mRNA and protein level. Collectively, these studies suggest that activation of the STAT3 signaling pathway plays an important role in the progression of pancreatic cancer, and that silence of the STAT3 gene with RNAi may be a useful anti-invasive therapeutic option in pancreatic cancer. (Cancer Sci 2007; 98: 1099-1106) P ancreatic cancer remains a widespread and difficult disease to treat with an overall 5-year survival of less than 5%. Thus, it represents one of the leading causes of cancer deaths in industrialized countries despite advances in medical therapy and surgical techniques.(1,2) Because of the insidious and aggressive natural history of the disease, most patients have local or metastatic spread at the time of presentation that precludes a resection. Therefore, less than 10% of cases constitute candidates for surgical resection at the time of diagnosis. Even among patients undergoing a potentially curative resection, the longterm outcome remains unsatisfactory due to early recurrence and metastatic disease.(3) Effective systemic therapy capable of reversing the aggressive biology of this disease is currently not available. Thus, understanding the molecular mechanisms underlying the aggressive biology of pancreatic cancer is one of the most important issues for this disease.STAT3, a member of the signal transduction and activation of transcription (STAT) family, is a central cytoplasmic transcription factor that is activated by phosphorylation of a conserved tyrosine residue in response to extracellular signals and oncogenes. Once tyrosine is phosphorylated, two STAT3 monomers form dimers through reciprocal phosphotyrosine-SH2 interactions. The dimers are phosphorylated STAT3 (p-STAT3), which translocate to the nucleus and bind to cognate DNA sequences, regulate the transcription of target genes and modulate fundamental cellular processes, such as proliferation and differentiation.(4,5) Lines of evidence demonstrate that elevated activity of STAT3 has been found frequently in a wide variety of human tumors including pancreatic cancer (6)(7)(8)(9) and STAT3 regulates a number...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.