Search

With the development of society, people are going out more and more, which leads to more and more crowded scenes. The detection of crowd density is particularly important. Aiming at the multi-scale problem of different human scales caused by camera angles in the crowd, a crowd density detection method based on deep semantic segmentation is proposed. The front-end of the network uses an improved VGG network to extract the crowd characteristics, so that the output feature map is 1/8 of the original image to improve the accuracy of the predicted density map. The back-end designs two atrous convolution modules with different array dilation rates to capture the multi-scale features of the crowd. The multi-scale features of the network enable the network to capture more scale details and edge information. The network finally uses 1×1 convolution to cascade the output to obtain a high-quality prediction density map. At the same time, in order to solve the grid effect caused by atrous convolution, a zigzag network structure is designed, so that every pixel in the convolution operation after zero filling is calculated to ensure the continuity of the information, thereby improving the accuracy of the network. The network performance is tested on the ShanghaiTech and UCF_CC_50 datasets. The test results are better than the current mainstream crowd density detection methods. The MAE value obtained from the test is 42.4% and 38.1% higher than that of the MCNN network. Compared with SANet, network performance is increased by 5.3% and 9.6%.

Image semantic segmentation is a necessary part of modern autonomous driving systems, because real-time and accurate capture of road condition information is the key to navigation and action planning. Traffic signs are important road condition information. The traffic sign semantic algorithm with stable performance, high real-time performance and accuracy that can meet the application needs is the basis for the realization of active safe driving systems and automatic driving systems. First, based on the analysis of actual application needs, the GTSDB database is selected as the original data, and a traffic sign data set that can comprehensively evaluate the perfor-mance of the semantic segmentation algorithm is designed. Then, based on the classical semantic segmentation network with stable performance U-Net, the D-Unet (D means dilated convolution), a deep neural network structure is proposed with better segmentation performance and higher real-time performance for small targets such as traffic signs. This method uses fewer pooling layers to retain more image information, and uses dilated convolution instead of conventional convolution to expand the receptive field of convolution, better overall planning global information. Finally, tested on the data set designed in this paper, compared with FCN-8s, SegNet, U-Net and other image segmentation network models, the mean intersection over union (MIoU) of the model is increased by about 11.9 percentage points, 6.09 percentage points and 3.71 percentage points, and the parameter amount is only 4.94%, 22.5% and 85.5% of the other three network models.

PoRVA and PEDV coinfections are extremely common in clinical practice. Although coinfections of PoRVA and PEDV are known to result in increased mortality, the underlying mechanism remains unknown. Here, we found that PoRVA infection promoted PEDV infection in vivo and in vitro and that PoRVA G9P[23] (RVA-HNNY strain) enhanced PEDV replication more significantly than did PoRVA G5P[7] (RVA-SXXA strain). Metabolomic analysis revealed that RVA-HNNY more efficiently induced an increase in the intracellular glutamine content in porcine small intestinal epithelial cells than did RVA-SXXA, which more markedly promoted ATP production to facilitate PEDV replication, whereas glutamine deprivation abrogated the effect of PoRVA infection on promoting PEDV replication. Further studies showed that PoRVA infection promoted glutamine uptake by upregulating the expression of the glutamine transporter protein SLC1A5. In SLC1A5 knockout cells, PoRVA infection neither elevated intracellular glutamine nor promoted PEDV replication. During PoRVA infection, the activity and protein expression levels of glutamine catabolism-related enzymes (GLS1 and GLUD1) were also significantly increased promoting ATP production through glutamine anaplerosis into the TCA cycle. Consistent with that, siRNAs or inhibitors of GLS1 and GLUD1 significantly inhibited the promotion of PEDV replication by PoRVA. Notably, RVA-HNNY infection more markedly promoted SLC1A5, GLS1 and GLUD1 expression to more significantly increase the uptake and catabolism of glutamine than RVA-SXXA infection. Collectively, our findings illuminate a novel mechanism by which PoRVA infection promotes PEDV infection and reveal that the modulation of glutamine uptake is key for the different efficiencies of PoRVA G9P[23] and PoRVA G5P[7] in promoting PEDV replication. Author summary: All phases of the viral life cycle are heavily dependent on host metabolism, thus virtually all viruses actively reprogram host metabolic networks to support their replication and spread. However, in the case of PoRVA and PEDV coinfection, the impact of host cell metabolic reprogramming induced by PoRVA infection on PEDV replication has not been reported. Herein, we demonstrated that PoRVA infection promotes PEDV replication by targeting glutamine metabolism in porcine intestinal epithelial cells. PoRVA infection induces increased expression levels of SLC1A5, GLS1, and GLUD1, which enhance glutamine uptake, TCA cycle flux, and ATP production. Notably, we found that PoRVA G9P[23] infection more effectively promoted PEDV replication than PoRVA G5P[7] infection by significantly increasing glutamine uptake and anaplerosis into the TCA cycle. [ABSTRACT FROM AUTHOR]

Background: Glioma, an aggressive brain tumor, poses a challenge in understanding the mechanisms of treatment resistance, despite promising results from immunotherapy., Methods: We identified genes associated with immunotherapy resistance through an analysis of The Cancer Genome Atlas (TCGA), Chinese Glioma Genome Atlas (CGGA), and Gene Expression Omnibus (GEO) databases. Subsequently, qRT-PCR and western blot analyses were conducted to measure the mRNA and protein levels of TBC1 Domain Family Member 1 (TBC1D1), respectively. Additionally, Gene Set Enrichment Analysis (GSEA) was employed to reveal relevant signaling pathways, and the expression of TBC1D1 in immune cells was analyzed using single-cell RNA sequencing (scRNA-seq) data from GEO database. Tumor Immune Dysfunction and Exclusion (TIDE) database was utilized to assess T-cell function, while Tumor Immunotherapy Gene Expression Resource (TIGER) database was employed to evaluate immunotherapy resistance in relation to TBC1D1. Furthermore, the predictive performance of molecules on prognosis was assessed using Kaplan-Meier plots, nomograms, and ROC curves., Results: The levels of TBC1D1 were significantly elevated in tumor tissue from glioma patients. Furthermore, high TBC1D1 expression was observed in macrophages compared to other cells, which negatively impacted T cell function, impaired immunotherapy response, promoted treatment tolerance, and led to poor prognosis. Inhibition of TBC1D1 was found to potentially synergistically enhance the efficacy of immunotherapy and prolong the survival of cancer patients with gliomas., Conclusion: Heightened expression of TBC1D1 may facilitate an immunosuppressive microenvironment and predict a poor prognosis. Blocking TBC1D1 could minimize immunotherapy resistance in cancer patients with gliomas., 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 Song, Yang, Li, Wei, Zhang, Du, Ren and Li.)