Search

Scene text in natural images carries additional semantic information to aid in image classification. Existing methods lack full consideration of the deep understanding of the text and the visual text relationship, which results in the difficult to judge the semantic accuracy and the relevance of the visual text. This paper proposes image classification based on Cross modal Knowledge Learning of Scene Text (CKLST) method. CKLST consists of three stages: cross‐modal scene text recognition, text semantic enhancement, and visual‐text feature alignment. In the first stage, multi‐attention is used to extract features layer by layer, and a self‐mask‐based iterative correction strategy is utilized to improve the scene text recognition accuracy. In the second stage, knowledge features are extracted using external knowledge and are fused with text features to enhance text semantic information. In the third stage, CKLST realizes visual‐text feature alignment across attention mechanisms with a similarity matrix, thus the correlation between images and text can be captured to improve the accuracy of the image classification tasks. On Con‐Text dataset, Crowd Activity dataset, Drink Bottle dataset, and Synth Text dataset, CKLST can perform significantly better than other baselines on fine‐grained image classification, with improvements of 3.54%, 5.37%, 3.28%, and 2.81% over the best baseline in mAP, respectively. [ABSTRACT FROM AUTHOR]

The fates of terrestrial organic carbon (OCterr) during fluvial transport from land to ocean are still not well constrained. This study systematically examines the evolution and dynamics of OCterr along the river‐estuary‐coastal ocean continuum in three fluvial systems discharging to the Chinese marginal seas. The 14C‐depleted characteristics of bulk OC and molecular components of riverine suspended sediments and marine sediments suggest that the Chinese marginal seas are a significant sink of pre‐aged OCterr. Our study reveals significant apparent aging of OCterr within estuaries, likely due to degradation of (younger) labile components, and apparent rejuvenation of OCterr in shelf systems, likely reflecting inputs of younger OCterr from proximal sources along the sediment dispersal pathway. The aging and rejuvenation of OCterr along the river‐ocean continuum confounds the use of plant wax lipid 14C to constrain lateral transport times, and sheds light on more complex OCterr dynamics in marginal seas. Plain Language Summary: Determining the factors driving degradation of terrestrial organic carbon (OCterr) is important for understanding the fate of OCterr in marginal seas and its impact on the global carbon cycle. In this study, we use bulk and molecular‐level carbon isotopic measurements as well as sedimentological data to investigate how OCterr characteristics evolve along the river‐estuary‐coastal ocean continuum for three different river systems discharging into the Chinese marginal seas. Marked decreases in the OCterr abundance and 14C contents reveal marked aging associated with degradation processes within estuaries over relatively short transport distances, likely due to degradation of younger and labile OCterr. Both aging and apparent "rejuvenation" of OCterr after estuarine transport was observed in different shelf systems, the latter likely resulting from inputs of fresh OCterr from proximal sources. Our findings suggest both sources and transport processes influence terrestrial OC and biomarker signatures in marginal sea sediments, highlighting the complexity of OCterr dynamics in coastal systems. Key Points: Bulk and molecular 14C values suggest millennial‐aged terrestrial OC transport along the Chinese river‐estuary‐coastal ocean continuumRapid degradation and significant aging of terrestrial OC co‐occur during estuarine processingApparent aging and rejuvenation of terrestrial OC along‐shelf transport confounded the transport time estimated by plant wax lipid 14C [ABSTRACT FROM AUTHOR]

Background: Abnormal DNA methylation is one of the most general epigenetic modifications in hepatocellular carcinoma (HCC). Recent research showed that DNA methylation was a prognostic indicator of all‐cause HCC and nonviral HCC. However, whether DNA methylation‐driver genes could be used for predicting survival, the probability of hepatitis‐positive HCC remains unclear. Methods: In this study, DNA methylation‐driver genes (MDGs) were screened by a joint analysis of methylome and transcriptome data of 142 hepatitis‐positive HCC patients. Subsequently, a prognostic risk score and nomogram were constructed. Finally, correlation analyses between the risk score and signaling pathways and immunity were conducted by GSVA and CIBERSORT. Results: Through random forest screening and Cox progression analysis, 10 prognostic methylation‐driver genes (AC008271.1, C11orf53, CASP8, F2RL2, GBP5, LUCAT1, RP11‐114B7.6, RP11‐149I23.3, RP11‐383 J24.1, and SLC35G2) were screened out. As a result, a prognostic risk score signature was constructed. The independent value of the risk score for prognosis prediction were addressed in the TCGA‐HCC and the China‐HCC cohorts. Next, clinicopathological features were analyzed and HBV status and histological grade were screened to construct a nomogram together with the risk score. The prognostic efficiency of the nomogram was validated by the calibration curves and the concordance index (C index: 0.829, 95% confidence interval: 0.794–0.864), while its clinical application ability was confirmed by decision curve analysis (DCA). At last, the relationship between the risk score and signaling pathways, as well as the correlations between immune cells were elucidated preliminary. Conclusions: Taken together, our study explored a novel DNA methylation‐driver gene risk score signature and an efficient nomogram for long‐term survival prediction of hepatitis‐positive HCC patients. [ABSTRACT FROM AUTHOR]

Lipid metabolic reprogramming is a prominent feature of clear cell renal cell carcinoma (ccRCC). Lipid accumulation affects cellular energy homeostasis, biofilm synthesis, lipid signal transduction, and phenotypic transformation in ccRCC. Herein, a prognostic‐related model was constructed, and the prognostic utility of AUP1, a lipid droplet–regulating very low–density lipoprotein assembly factor, in ccRCC was determined through multiparameter analysis. AUP1 expression was significantly higher in clinical samples than in normal tissues and was closely associated with the clinical stage. The inhibition of AUP1 expression impaired the proliferation, migration, and invasion of ACHN and A498 ccRCC cells in vitro and in vivo. RNA‐seq analysis revealed that AUP1 inhibition can significantly reduce the contents of intracellular triglyceride and cholesterol and regulate cell growth by cell cycle arrest, promoting apoptosis and reversing epithelial‐mesenchymal transition. AUP1 regulated the synthesis of cholesterol esters and fatty acids (FAs) in ccRCC cells by targeting sterol O‐acyltransferase 1 and partially promoted the progression of ccRCC. AUP1 also induced lipid accumulation in ccRCC by promoting the de novo synthesis of FAs (inhibiting protein kinase AMP‐activated catalytic subunit alpha 2), inhibiting the rate‐limiting enzyme of FA β oxidation (carnitine palmitoyltransferase 1A), regulating the key enzyme of lipolysis (monoglyceride lipase, MGLL), and inhibiting the lipid transporter StAR‐related lipid transfer domain containing 5 (STARD5). However, it did not affect the intracellular cholesterol synthesis pathway. The differential expression and prognostic significance of MGLL and STARD5 in ccRCC should be further studied. AUP1 may serve as a new and effective potential target and prognostic marker for ccRCC. [ABSTRACT FROM AUTHOR]

To effectively optimise pole shape for surface‐mounted permanent magnet synchronous machine (SPMSM), an analytical model of SPMSM with the chamfered pole is proposed for no‐load and on‐load magnetic field calculation. In this model, the pole of SPMSM is approximately segmented into pieces with regular shape and uniform magnet property, and magnetic field parameters can be obtained by subdomain superposition. Meanwhile, iron core saturation is considered to get a correction coefficient to revise air‐gap flux density, inducing flux linkage, line no‐load back electromotive force (EMF) and electromagnetic torque. Based on this analytical model, the geometry optimisation of PM can be optimised by end thickness of pole, pole‐arc coefficient of cutting and remanence, and the waveform total harmonic distortion of air‐gap flux density and no‐load back EMF are greatly reduced. To verify the analytical model, an outer rotor type SPMSM with 20‐poles and 72‐slots for tractor application is designed and manufactured. Finally, the effectiveness of the improved method is validated by the finite‐element method and experimental test. [ABSTRACT FROM AUTHOR]

A Robin type boundary condition (BC), commonly adopted at stream‐aquifer interface, excludes a term associated with streambed accounting for the effects of streambed storage and width. This study presents two new analytical models for describing confined flow induced by pumping in a stream‐aquifer system. One model considers a single‐zone aquifer and treats the streambed as a lagging Robin BC with a time lag parameter related to the effects. The other considers a two‐zone aquifer consisting of aquifer and streambed zones. A Dirichlet BC for stream water level is specified at the edge of the streambed. The time‐domain solutions of both models are developed to describe spatiotemporal drawdown and temporal stream filtration/depletion rate (SDR). The finite element solutions (FESs) of both models are also built. Results suggest the lag time equals the half of the squared streambed width divided by the streambed hydraulic diffusivity. The effects of streambed width and storage on SDR should be considered when their lumped parameter exceeds 0.1. Neglecting their effects causes 25% difference in SDR when the lumped parameter equals 10. Based on the FESs, the use of the lagging Robin BC takes nearly a tenth of computing time of obtaining accurate steady‐state SDR for the simulation of the two‐zone aquifer. In addition, the present solutions agree to a field SDR experiment conducted by Hunt et al. (2001, https://doi.org/10.1111/j.1745‐6584.2001.tb02310.x). To conclude, this study presents two new models for describing groundwater flow in a stream‐aquifer system and explores the joint effect of streambed width and storage on SDR. Key Points: Two new analytical solutions are developed to estimate aquifer drawdown and stream depletion rate in a pumped stream‐aquifer systemOne treats streambed as a Robin type boundary based on the lagging theory while the other considers streambed as a finite‐width zoneThe lag time in the Robin boundary equals the half of squared streambed width divided by the streambed hydraulic diffusivity [ABSTRACT FROM AUTHOR]

Real‐time visualization of the positions of micro/nanomotors is highly desired for practical applications due to the complex environments that they are used in. However, most reported photocatalytic micromotors can only be tracked in transparent solutions under bright fields, which greatly limits their applications. Herein, photocatalytic micromotors based on Cu2O@carbon dot (CDot) particles with excellent fluorescent properties and precise propulsion capabilities in 3D are reported. In the fully green environment of malic acid, the Cu2O@CDot micromotors display efficient propulsion under visible light irradiation (as high as 40.85 μm s−1). In addition, the propulsion direction of the Cu2O@CDot micromotors can be effectively adjusted by changing the direction of light via positive phototaxis. Interestingly, due to the presence of CDots, these new Cu2O@CDot micromotors display excellent and stable fluorescence. This study reports the first photocatalytic micromotors with both novel fluorescence properties and excellent propulsion characteristics in biocompatible environments. Due to their excellent fluorescent properties and precise motion control, these Cu2O@CDot micromotors hold great potential to perform complex tasks across diverse practical applications. [ABSTRACT FROM AUTHOR]

Background: Evidence from cohort studies indicates a bidirectional relationship between periodontal disease and type 2 diabetes (T2D), but the underlying mechanisms remain unknown. In this study, we aimed to (1) identify saliva, plasma, and multifluid metabolomic signatures associated with periodontal disease and (2) determine if these signatures predict T2D progression and cardiometabolic biomarkers at year 3., Methods and Results: We included participants from the SOALS (San Juan Overweight Adult Longitudinal Study) (n=911). Metabolites from saliva (k=635) and plasma (k=1051) were quantified using liquid chromatography-mass spectrometry. We applied elastic net regression with 10-fold cross-validation to identify baseline metabolomic signatures of periodontal disease. Multivariable Cox proportional hazards regression and linear regression were used to evaluate the association with T2D progression and biomarker concentrations. Metabolomic profiles included highly weighted metabolites related to lysine and pyrimidine metabolism. Periodontal disease or its 3 metabolomic signatures were not associated with T2D progression in 3 years. Prospectively, 1-SD increments in the multifluid and saliva metabolomic signatures were associated with higher low-density lipoprotein (multifluid: 12.9±5.70, P =0.02; saliva: 13.3±5.11, P =0.009). A 1-SD increment in the plasma metabolomic signature was also associated with Homeostatic Model Assessment for Insulin Resistance (2.67±1.14, P =0.02) and triglyceride (0.52±0.18, P =0.002)., Conclusions: Although metabolomic signatures of periodontal disease could not predict T2D progression, they were associated with low-density lipoprotein, triglyceride, and Homeostatic Model Assessment for Insulin Resistance levels at year 3.