Learn More
BACKGROUND Although warfarin and other anticoagulants can prevent ischemic events, they can cause hemorrhage. Quantifying the rate of hemorrhage is crucial for determining the risks and net benefits of prescribing antithrombotic therapy. Our objective was to find a bleeding classification scheme that could quantify the risk of hemorrhage in elderly patients(More)
We present a novel unsupervised deep learning framework for anomalous event detection in complex video scenes. While most existing works merely use hand-crafted appearance and motion features, we propose Appearance and Motion DeepNet (AMDN) which utilizes deep neural networks to automatically learn feature representations. To exploit the complementary(More)
Obtaining labels can be expensive or timeconsuming, but unlabeled data is often abundant and easier to obtain. Most learning tasks can be made more efficient, in terms of labeling cost, by intelligently choosing specific unlabeled instances to be labeled by an oracle. The general problem of optimally choosing these instances is known as active learning. As(More)
KChIPs coassemble with pore-forming Kv4 alpha subunits to form a native complex in the brain and heart and regulate the expression and gating properties of Kv4 K(+) channels, but the mechanisms underlying these processes are unknown. Here we report a co-crystal structure of the complex of human Kv4.3 N-terminus and KChIP1 at a 3.2-A resolution. The(More)
Bcl-xL plays a critical role in maintaining cell survival. However, the relationship between the potential interaction of Bcl-xL with other cytosolic proteins and the regulation of cell survival remains incompletely defined. We have identified translationally controlled tumor protein (TCTP), a multifunctional protein, as a novel antiapoptotic(More)
Dimethylated histone H3 lysine 9 (H3K9me2) is a critical epigenetic mark for gene repression and silencing and plays an essential role in embryogenesis and carcinogenesis. Here, we investigated the effects of hypoxic stress on H3K9me2 at both global and gene-specific level. We found that hypoxia increased global H3K9me2 in several mammalian cell lines. This(More)
Supervised learning from multiple labeling sources is an increasingly important problem in machine learning and data mining. This paper develops a probabilistic approach to this problem when annotators may be unreliable (labels are noisy), but also their expertise varies depending on the data they observe (annotators may have knowledge about different parts(More)
Most previous heterogeneous transfer learning methods learn a cross-domain feature mapping between heterogeneous feature spaces based on a few cross-domain instance-correspondences, and these corresponding instances are assumed to be representative in the source and target domains respectively. However, in many realworld scenarios, this assumption may not(More)
BACKGROUND Major depressive disorder (MDD) is one of the important causes of disease burden in the general population. Given the experiencing rapid economic and social changes since the early 1990s and the internationally recognized diagnostic criteria and interview instruments across the surveys during 2001-2010 in china, the epidemiological studies on MDD(More)
MicroRNAs (miRNAs) have been believed to associate with malignant progression including cancer cell proliferation, apoptosis, differentiation, angiogenesis, invasion and metastasis. However, the functions of miRNAs are intricate, one miRNA can directly or indirectly target multiple genes and function as oncogene or tumor suppressor gene. In this study, we(More)