Search

To evaluate the prevalence and influencing factors of long COVID, and measure the difference in health status between long COVID and non-long COVID cases. Methods: A cross-sectional survey was conducted from February 1 to 8, 2023, using a stratified random sampling method in four regions (eastern [Changzhou], central [Zhengzhou], western [Xining] and northeastern [Mudanjiang]) of China. The survey collected COVID-19 patients' socio-demographic characteristics and lifestyles information. The scores of lifestyles and health status range from 5 to 21 and 0 to 100 points, respectively. The criteria of "persistent health problems after 4 weeks of COVID-19 infection" issued by the US Centers for Disease Control and Prevention was used to assess long COVID. Multiple linear regression was used to analyze the influencing factors of the health. The bootstrap method was used to analyze the lifestyles' mediating effect. Propensity score matching (PSM) was used to evaluate the net difference in health scores between long COVID and non-long COVID cases. Results: The study included 3165 COVID-19 patients, with 308 (9.73%) long COVID cases. The health score of the long COVID cases (74.79) was lower than that of the non-long COVID cases (81.06). After adjusting for potential confounding variables, we found that never focused on mental decompression was a common risk factor for the health of both groups. Lifestyles was the mediating factor on individuals' health. After PSM, the non-long COVID cases' health scores remained higher than that of long COVID cases. Conclusion: The proportion of long COVID cases was low, but they were worse off in health. Given the positive moderating effect of healthy lifestyles on improving the health of long COVID cases, healthy lifestyles including mental decompression should be considered as the core strategy of primary prevention when the epidemic of COVID-19 is still at a low level. [ABSTRACT FROM AUTHOR]

In this paper, the automatic facial expression based on the combination of skin color and face motion is studied. The image is transformed into YIQ color space by RGB, and the image data is extracted in the I dimension of YIQ, and the background and skin color are segmented in the two valued image. By using Pareto optimization algorithm to select facial expression feature, this algorithm has less calculation, simple structure, fast running speed, accurate detection and can track the obstructions of small angle face skin color, facial expression, face rotation and face. Experiments show that the algorithm can adapt to small angle rotation of human face. For human eyes, the algorithm is not affected, can better adapt to the rich facial expression changes and different skin colors, and has a certain stability.

While existing supervised semantic segmentation methods have shown significant performance improvements, they heavily rely on large‐scale pixel‐level annotated data. To reduce this dependence, recent research has proposed semi‐supervised learning‐based methods that have achieved great success. However, almost all these works are mainly dedicated to image semantic segmentation, while semi‐supervised video semantic segmentation (SVSS) has been barely explored. Due to the significant difference between video data and image, simply adapting semi‐supervised image semantic segmentation approaches to SVSS may neglect the inherent temporal correlations in video frames. This paper presents a novel method (named TRACL) with temporal reconstruction (TR) and adaptive consistency loss (ACL) for SVSS, aiming to fully utilize the temporal relations of internal frames in video clip. The authors' TR method implements the reconstruction from the feature and output levels to narrow the distribution gap between internal video frames. Specifically, considering the underlying data distribution, the authors construct Gaussian models for each category, and use probability density function to obtain the similarity between different feature maps for temporal feature reconstruction. The authors' ACL can adaptively select two pixel‐wise consistency loss including Flow Consistency Loss and Reconstruction Consistency Loss, providing stronger supervision signals for unlabelled frames during model training. Additionally, the authors extend their method to unlabelled video for more training data by employing mean‐teacher structure. Extensive experiments on three datasets including Cityscapes, Camvid and VSPW demonstrate that the authors' proposed method outperforms previous state‐of‐the‐art methods. [ABSTRACT FROM AUTHOR]

During the COVID-19 pandemic, the vaccine hesitancy has significantly affected the vaccination. To evaluate the booster vaccine hesitancy and its influencing factors among urban and rural residents, as well as to estimate the net difference of booster vaccine hesitancy between urban and rural residents. We conducted a nationwide, cross-sectional Internet survey on 1-8 February 2023, and employed stratified random sampling technique to select participants (≥18 years old) from urban and rural areas. Multivariate logistic regression was used to determine the factors impacting booster vaccine hesitancy. Propensity Score Matching was used to estimate the net difference of COVID-19 booster vaccine hesitancy between urban and rural residents. The overall COVID-19 booster vaccine hesitancy rate of residents was 28.43%. The COVID-19 booster vaccine hesitancy rate among urban residents was found to be 34.70%, among rural residents was 20.25%. Chronic diseases, infection status, vaccination benefits, and trust in vaccine developers were associated with booster vaccine hesitancy among urban residents. Barriers of vaccination were associated with booster vaccine hesitancy among rural residents. PSM analysis showed that the urban residents have a higher booster vaccine hesitancy rate than rural residents, with a net difference of 6.20%. The vaccine hesitancy rate increased significantly, and the urban residents have a higher COVID-19 booster vaccine hesitancy than rural residents. It becomes crucial to enhance the dissemination of information regarding the advantages of vaccination and foster greater trust among urban residents toward the healthcare system.

Detecting and controlling tea pests promptly are crucial for safeguarding tea production quality. Due to the insufficient feature extraction ability of traditional CNN-based methods, they face challenges such as inaccuracy and inefficiency of detecting pests in dense and mimicry scenarios. This study proposes an end-to-end tea pest detection and segmentation framework, TeaPest-Transfiner (TP-Transfiner), based on Mask Transfiner to address the challenge of detecting and segmenting pests in mimicry and dense scenarios. In order to improve the feature extraction inability and weak accuracy of traditional convolution modules, this study proposes three strategies. Firstly, a deformable attention block is integrated into the model, which consists of deformable convolution and self-attention using the key content only term. Secondly, the FPN architecture in the backbone network is improved with a more effective feature-aligned pyramid network (FaPN). Lastly, focal loss is employed to balance positive and negative samples during the training period, and parameters are adapted to the dataset distribution. Furthermore, to address the lack of tea pest images, a dataset called TeaPestDataset is constructed, which contains 1,752 images and 29 species of tea pests. Experimental results on the TeaPestDataset show that the proposed TP-Transfiner model achieves state-of-the-art performance compared with other models, attaining a detection precision (AP50) of 87.211% and segmentation performance of 87.381%. Notably, the model shows a significant improvement in segmentation average precision (mAP) by 9.4% and a reduction in model size by 30% compared to the state-of-the-art CNN-based model Mask R-CNN. Simultaneously, TP-Transfiner's lightweight module fusion maintains fast inference speeds and a compact model size, demonstrating practical potential for pest control in tea gardens, especially in dense and mimicry scenarios., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Wu, He, Rong, Liang, Xu, Ni and Dong.)

Background: Vaccination plays an important role in preventing COVID-19 infection and reducing the severity of the disease. There are usually differences in vaccination rates between urban and rural areas. Measuring these differences can aid in developing more coordinated and sustainable solutions. This information also serves as a reference for the prevention and control of emerging infectious diseases in the future., Objective: This study aims to assess the current coverage rate and influencing factors of COVID-19 (second booster) vaccination among Chinese residents, as well as the disparities between urban and rural areas in China., Methods: This cross-sectional study used a stratified random sampling approach to select representative samples from 11 communities and 10 villages in eastern (Changzhou), central (Zhengzhou), western (Xining), and northeast (Mudanjiang) Mainland China from February 1 to February 18, 2023. The questionnaires were developed by experienced epidemiologists and contained the following: sociodemographic information, health conditions, vaccine-related information, information related to the Protective Motivation Theory (PMT), and the level of trust in the health care system. Vaccination rates among the participants were evaluated based on self-reported information provided. Binary logistic regression models were performed to explore influencing factors of vaccination among urban and rural participants. Urban-rural disparities in the vaccination rate were assessed using propensity score matching (PSM)., Results: A total of 5780 participants were included, with 53.04% (3066/5780) being female. The vaccination rate was 12.18% (704/5780; 95% CI 11.34-13.02) in the total sample, 13.76% (341/2478; 95% CI 12.40-15.12) among the rural participants, and 10.99% (363/3302; 95% CI 9.93-12.06) among the urban participants. For rural participants, self-reported health condition, self-efficacy, educational level, vaccine knowledge, susceptibility, benefits, and trust in the health care system were independent factors associated with vaccination (all P＜.05). For urban participants, chronic conditions, COVID-19 infection, subjective community level, vaccine knowledge, self-efficacy, and trust in the health care system were independent factors associated with vaccination (all P＜.05). PSM analysis uncovered a 3.42% difference in vaccination rates between urban and rural participants., Conclusions: The fourth COVID-19 vaccination coverage rate (second booster) among the Chinese population was extremely low, significantly lower than the previous vaccine coverage rate. Given that COVID-19 infection is still prevalent at low levels, efforts should focus on enhancing self-efficacy to expand the vaccine coverage rate among the Chinese population. For rural residents, building awareness of the vaccine's benefits and improving their overall health status should be prioritized. In urban areas, a larger proportion of people with COVID-19 and patients with chronic illness should be vaccinated., (©Wenyong Dong, Yudong Miao, Zhanlei Shen, Wanliang Zhang, Junwen Bai, Dongfang Zhu, Ruizhe Ren, Jingbao Zhang, Jian Wu, Clifford Silver Tarimo, Theodora Ojangba, Yi Li. Originally published in JMIR Public Health and Surveillance (https://publichealth.jmir.org), 19.07.2024.)

One of the main unsolved problems in computer algebra is to determine the minimal number of multiplications which is necessary to compute the product of two matrices. For practical value, the small format is of special interest. This leads to a combinatorial optimization problem which is unlikely solved in polynomial time. In this paper, we present a method called combining Gaussian eliminations to reduce the number of variables in this optimization problem and use heuristic ant colony algorithm to solve the problem. The results of experiments on 2 × 2 case show that our algorithm achieves significant performance gains. Extending this algorithm from 2 × 2 case to 3 × 3 case is also discussed. Index Terms—Ant colony optimization (ACO), evolutionary algorithms, Gaussian eliminations, matrix multiplication, multiplicative complexity, Strassen's algorithm.