Recent generative adversarial network based methods have shown promising results for the charming but challenging task of synthesizing images from text descriptions. These approaches can generate images with general shape and color but often produce distorted global structures with unnatural local semantic details. It is due to ineffectiveness of convolutional neural networks in capturing the high-level semantic information for pixel-level image synthesis. In this paper, we propose a Dual Attentional Generative Adversarial Network (DualAttn-GAN) in which the dual attention modules are introduced to enhance local details and global structures by attending to related features from relevant words and different visual regions. As one of the dual modules, the textual attention module is designed to explore the fine-grained interaction between vision and language. On the other hand, visual attention module models internal representations of vision from channel and spatial axes, which can better capture the global structures. Meanwhile, we apply an attention embedding module to merge multi-path features. Furthermore, we present an inverted residual structure to boost representation power of CNNs and apply spectral normalization to stabilize GAN training. With extensive experimental validation on two benchmark datasets, our method significantly improves stateof-the-art models over the evaluation metrics of inception score and Fréchet inception distance.INDEX TERMS Generative adversarial network, textual attention, visual attention, inverted residual structure, spectral normalization.
Tibetans have a higher essential hypertension prevalence compared with other ethnics in China. The reason might be due to their unique environmental influence, as well as genetic factor. However, limited studies focus on Tibetan genetics and its association with hypertension. The aim of this study was to investigate the association between With-No-Lysine (K) Kinase 1 (WNK1), Serine/Threonine kinase 39(STK39) genes variants and hypertension in the Tibetan population. 204 Tibetan hypertensive patients and 305 normotensive controls were recruited in an epidemiological survey conducted at 2 sites in the Ganzi Tibetan autonomous region. Patients were genotyped for nineteen WNK1 candidate tag single nucleotide polymorphisms (SNPs) and three STK39 SNPs, and haplotype analysis was performed. Results showed that the allele A in rs1468326 was overrepresented in hypertensive patients versus control (53.4% vs 42.9%, P < 0.05). The multivariable-adjusted odds ratio (OR) for hypertension among CA + AA genotypes carriers was 1.60 (95% CI: 1.02-2.62, P < 0.05), and they also had a higher systolic blood pressure (136.5 ± 28.6 vs 131.7 ± 24.8 mmHg, P < 0.05). However, the TT genotype ratio in rs6749447 was lower in hypertensives (5.4% vs 10.8%, P < 0.05), and the hypertension risk for the TT genotype carriers in rs6749447 decreased after adjustment (OR 0.49, 95% CI 0.19-0.95, P < 0.05). Subjects with haplotype AGACAGGAATCGT showed 1.57 times higher risk of hypertension (95% CI 1.02-2.41, P < 0.05). In conclusion, SNP rs1468326 of WNK1, rs6749447 of STK39, and WNK1 haplotype AGACAGGAATCGT were associated with hypertension in Tibetan individuals. Environ. Mol. Mutagen. 59:151-160, 2018. © 2017 Wiley Periodicals, Inc.
We performed long non-coding RNA (lncRNA) microarray assay to identify lncRNAs with differential expression between patients with intracranial aneurysm (IA) and healthy control individuals to evaluate their potential use as biomarkers of IA. Arraystar Human lncRNA Microarray v3.0 was performed to identify differentially expressed lncRNAs and mRNAs in plasma samples (4 ml). lncRNAs with the most pronounced differential expression were used to select gene markers, and results were validated by quantitative reverse-transcription polymerase chain reaction (qRT-PCR). Plasma levels of TCONS_00000200 (fold change: 2.28) and ENST00000511927 (fold change: 2.50) were significantly higher in IA patients than in healthy individuals (P<0.001), and plasma levels of ENST00000421997 (fold change: 0.45) and ENST00000538202 (fold change: 0.43) were significantly lower in IA patients than in healthy individuals (P<0.001). qRT-PCR confirmed the same trends of up- and down-regulation of these four lncRNAs. A receiver operating characteristic (ROC) curve for TCONS_00000200 showed that the area under the curve (AUC) was 0.963 (95% confidence interval, 0.919–1.000), optimal cut-off point was 0.0081, sensitivity was 90.0%, and specificity was 96.7%. These results indicate that the lncRNA TCONS_00000200 is differentially expressed in the plasma of IA patients and could serve as a biomarker of IA.
CuInP2S6 (CIPS) is a novel two-dimensional (2D) van der Waals (vdW) ferroelectric layered material with a Curie temperature of TC~315 K, making it promising for great potential applications in electronic and photoelectric devices. Herein, the ferroelectric and electric properties of CIPS at different thicknesses are carefully evaluated by scanning probe microscopy techniques. Some defects in some local regions due to Cu deficiency lead to a CuInP2S6–In4/3P2S6 (CIPS–IPS) paraelectric phase coexisting with the CIPS ferroelectric phase. An electrochemical strain microscopy (ESM) study reveals that the relaxation times corresponding to the Cu ions and the IPS ionospheres are not the same, with a significant difference in their response to DC voltage, related to the rectification effect of the ferroelectric tunnel junction (FTJ). The electric properties of the FTJ indicate Cu+ ion migration and propose that the current flow and device performance are dynamically controlled by an interfacial Schottky barrier. The addition of the ferroelectricity of CIPS opens up applications in memories and sensors, actuators, and even spin-orbit devices based on 2D vdW heterostructures.
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.