Search

The main objective of scene text recognition is to recognize text in complex images and convert it into editable text. However, scene text recognition research has long been focused on English, and there is a lack of research on other small languages, such as Uyghur language. This paper conducts a series of studies on Uyghur text recognition in natural scenes, with the following main contributions: 1) To address the lack of Uyghur scene text recognition datasets, we established the Synthetic Uyghur Scene Text dataset SUST and the Real Uyghur Scene Text dataset RUST, and we augmented RUST with STR-Aug; 2) The contemporary existing STR models are selected to conduct experiments on the dataset proposed in this paper, through which we search for the model structure suitable for Uyghur scene text recognition; 3) According to the characteristics of Uyghur text, this paper designs a lightweight Uyghur text recognition model named LUSTR, which takes into account both lightweight and accuracy, and achieves good performance in the Uyghur scene text recognition dataset proposed in this paper. The SUST and RUST are now publicly available at https://github.com/kongfnajie/SUST-and-RUST-datasets-for-Uyghur-STR.

Aimed at the problems in continuous crowd sensing,such as the increased privacy requirements,the unreliable perception data collected and the low enthusiasm of users to participate,this paper proposes an incentive mechanism based on symmetric encryption and double-layer truth discovery(SDIM).First,the symmetric encryption algorithm is used to protect the privacy of the perceived data.When the privacy requirements are high and the number of perceptions is large,the computing overhead and the time of data encryption and reward computing will be greatly reduced.Second,based on a double-layer truth discovery model,an incentive mechanism supporting data reliability evaluation is proposed.The purpose is to simultaneously realize the real time reward of continuous crowd sensing,and improve the fairness of reward when the participants have malicious behavior.Finally,the dual privacy analysis of the proposed method is illustrated.The simulation results show that the proposed method can effectively calculate the true value and the reward according to the data reliability.Notably,it is obviously superior to the comparative model in the time of data encryption and reward computing,and can calculate the reward more fairly when the participants have malicious behavior.

Objectives: Overuse of antibiotics in healthcare remains prevalent and requires urgent attention in China, particularly in primary healthcare (PHC) facilities. This study aimed to describe the patterns of antibiotic prescriptions for acute upper respiratory tract infections (URTIs) in PHC facilities in China and to investigate how PHC physicians' knowledge influences their antibiotic prescribing behavior. Methods: A cross-sectional survey linking physician questionnaire responses and their prescription data was conducted in Shaanxi Province, China. The proportions of URTI visits that received at least one antibiotic, combined antibiotics, and broad-spectrum antibiotics were the main outcomes reflecting antibiotic prescribing behavior. Multivariate mixed-effects logistic regressions were applied to analyze the relationship between PHC physicians' knowledge about antibiotics and their antibiotic prescribing behavior. Results: A total of 108 physicians filled out the questionnaires between February 2021 and July 2021, and a sample of 11,217 URTI visits attended by these physicians from 1 January 2020 to 31 December 2020 were included in the analysis. The overall mean score of our respondents on the knowledge questions was 5.2 (total score of 10). Over sixty percent (61.2%; IQR 50.2–72.1) of the URTI visits received antibiotics. The percentages of URTI visits prescribed with combined and broad-spectrum antibiotics were 7.8% (IQR 2.3–10.2) and 48.3% (IQR 36.7–58.7), respectively. Third-generation cephalosporins were the most frequently used antibiotics. Physicians with lower antibiotic knowledge scores were more inclined to prescribe antibiotics (p < 0.001), combined antibiotics (p = 0.001), and broad-spectrum antibiotics (p < 0.001). Conclusions: Physicians' insufficient knowledge was significantly associated with antibiotic overprescribing. Professional training targeting physicians' knowledge of antibiotics is urgently needed to improve the rational use of antibiotics in grassroots healthcare facilities in China. [ABSTRACT FROM AUTHOR]

Macrophages, known for their phenotypic plasticity, play a critical role in maintaining homeostasis and inflammation-related pathogenesis. Although identifying diverse macrophage phenotypes holds promise for enhancing diagnoses and treatments of diseases mediated by macrophages, existing methodologies for differentiating macrophages often lack precision. They are limited by the cumbersome procedures that require large-scale equipment, such as flow cytometry and transcriptomic analysis. In this context, we have engineered fluorescent polyadenine (polyA)-mediated sticky flares that enable practical visualization of macrophages. This technology facilitates the highly sensitive detection of macrophage phenotypes through the specific recognition of intracellular mRNAs, permitting in situ imaging. Our approach demonstrates the potential for determining macrophage polarization status at the single-cell level within dynamic immune microenvironments, thereby providing crucial diagnostic and prognostic information that could guide the development of tailored treatments for macrophage-related diseases in personalized medicine. [ABSTRACT FROM AUTHOR]