A Lightweight Vehicle Detection Method Fusing GSConv and Coordinate Attention Mechanism.

Aiming at the problems of target detection models in traffic scenarios including a large number of parameters, heavy computational burden, and high application cost, this paper introduces an enhanced lightweight real-time detection algorithm, which exhibits higher detection speed and accuracy for vehicle detection. This paper considers the YOLOv7 algorithm as the benchmark model, designs a lightweight backbone network, and uses the MobileNetV3 lightweight network to extract target features. Inspired by the structure of SPPF, the spatial pyramid pooling module is reconfigured by incorporating GSConv, and a lightweight SPPFCSPC-GS module is designed, aiming to minimize the quantity of model parameters and enhance the training speed even further. Furthermore, the CA mechanism is integrated to enhance the feature extraction capability of the model. Finally, the MPDIoU loss function is utilized to optimize the model's training process. Experiments showcase that the refined YOLOv7 algorithm can achieve 98.2% mAP on the BIT-Vehicle dataset with 52.8% fewer model parameters than the original model and a 35.2% improvement in FPS. The enhanced model adeptly strikes a finer equilibrium between velocity and precision, providing favorable conditions for embedding the model into mobile devices.

Spatiotemporal information enhanced multi-feature short-term traffic flow prediction.

Accurately predicting traffic flow is crucial for optimizing traffic conditions, reducing congestion, and improving travel efficiency. To explore spatiotemporal characteristics of traffic flow in depth, this study proposes the MFSTBiSGAT model. The MFSTBiSGAT model leverages graph attention networks to extract dynamic spatial features from complex road networks, and utilizes bidirectional long short-term memory networks to capture temporal correlations from both past and future time perspectives. Additionally, spatial and temporal information enhancement layers are employed to comprehensively capture traffic flow patterns. The model aims to directly extract original temporal features from traffic flow data, and utilizes the Spearman function to extract hidden spatial matrices of road networks for deeper insights into spatiotemporal characteristics. Historical traffic speed and lane occupancy data are integrated into the prediction model to reduce forecasting errors and enhance robustness. Experimental results on two real-world traffic datasets demonstrate that MFSTBiSGAT successfully extracts and captures spatiotemporal correlations in traffic networks, significantly improving prediction accuracy.

Calpain-cleaved collapsin response mediator protein-3 induces neuronal death after glutamate toxicity and cerebral ischemia.

Collapsin response mediator proteins (CRMPs) mediate growth cone collapse during development, but their roles in adult brains are not clear. Here we report the findings that the full-length CRMP-3 (p63) is a direct target of calpain that cleaves CRMP-3 at the N terminus (+76 amino acid). Interestingly, activated calpain in response to excitotoxicity in vitro and cerebral ischemia in vivo also cleaved CRMP-3, and the cleavage product of CRMP-3 (p54) underwent nuclear translocation during neuronal death. The expression of p54 was colocalized with the terminal deoxynucleotidyl transferase-mediated biotinylated UTP nick end labeling-positive nuclei in glutamate-treated cerebellar granule neurons (CGNs) and in ischemic neurons located in the infarct core after focal cerebral ischemia, suggesting that p54 might be involved in neuronal death. Overexpression studies showed that p54, but not p63, caused death of human embryonic kidney cells and CGNs, whereas knock-down CRMP-3 expression by selective small interfering RNA protected neurons against glutamate toxicity. Collectively, these results reveal a novel role of CRMP-3 in that calpain cleavage of CRMP-3 and the subsequent nuclear translocation of the truncated CRMP-3 evokes neuronal death in response to excitotoxicity and cerebral ischemia. Our findings also establish a novel route of how calpain signals neuron death.

A novel adenoviral vector-mediated neuronal selective gene expression in neonatal mouse brain in response to hypoxia.

Selective gene expression targeting neurons is a challenge, which, if successfully overcome, carries an enormous potential for clinical applications in therapeutics against neurodegenerative diseases. We have reported previously the construction of a series of adenoviral vectors capable of selectively expressing a reporter gene luciferase in cultured neurons [D. Huang, A. Desbois, S.T. Hou, A novel adenoviral vector which mediates hypoxia-inducible gene expression selectively in neurons, Gene Ther. 12 (2005) 1369-1376]. A combination of neuron restrictive silencer elements (NRSEs), hypoxia responsive elements (HREs) and CMV minimal promoter (CMVmp) was packaged into replication defective adenovirus to target gene expression selectively in neurons in a hypoxia-regulated manner. In the present study, we injected the adenoviral vectors into the neonatal mouse brain followed by treatment with hypoxia. The expression of the reporter luciferase gene was examined by luciferase assay and fluorescent immunostaining. Neurons and glial cells were identified by staining with antibodies against NeuN and GFAP, respectively. Remarkably, in response to hypoxia, Ad/5HRE-3NRSE showed strong hypoxia-inducible gene expression of the reporter luciferase selectively in neurons but not in glial cells. In contrast, brains infected with the control vector Ad/5HRE showed no selectivity in luciferase expression (in both neurons and glial cells) under the hypoxic condition. Taken together, these studies demonstrated that this vector (Ad/5HRE-3NRSE) can mediate gene expression selectively in neurons both in vitro and in vivo, supporting the suggestion that further refinement of this vector may lead to the development of a useful tool to investigate mechanisms of neuronal damage following cerebral ischemia and a possible effective gene therapy vector to stroke.

The gas7 protein potentiates NGF-mediated differentiation of PC12 cells.

The growth-arrest-specific protein gas7 is required for morphological differentiation of cultured mouse cerebellar neurons and PC12 cells. Moreover, its overexpression in various cell types induces neurite-like outgrowth. The role of gas7 in neuronal differentiation was further characterized by adenovirus-mediated overexpression in PC12 cells and quantification of the expression of various neuronal markers, in the absence and presence of different concentrations of nerve growth factor (NGF). The potential neuroprotective activity of gas7 against various neurotoxic insults was also assessed. In addition to promoting the formation of neurite-like extensions, overexpression of gas7 potentiated NGF-mediated neuronal differentiation of PC12 cells, as shown by the enhanced expression of the neuronal proteins betaIII-tubulin, synaptotagmin, alpha7 subunit of the acetylcholine receptor, and dihydropyrimidinase related protein-3. This effect was exerted independently of cell cycle progression, as gas7 did not affect proliferation of PC12 cells. While some differentiation enhancers protect PC12 cells against lethal insults, gas7 overexpression in PC12 cells did not protect against oxygen-glucose deprivation, the calcium ionophore A23187, or the nitric oxide donor sodium nitroprusside, suggesting that gas7 is not neuroprotective. The ability of gas7 to potentiate neuronal differentiation makes it a potential therapeutic target to promote re-establishment of neuronal connections in the injured or diseased brain, such as following stroke.

Characterization of the expression of key adenoviral receptors CAR and integrin beta3/beta5 subunits on the membrane of human NT2 neurons.

Expression of therapeutic gene products in differentiated human NT2 neurons (NT2/Ns) is being explored for ex vivo gene therapy of human neurological diseases. In this study we determined the efficiency of adenovirus (Ad)-mediated gene delivery into NT2/Ns and characterized the expression of several key receptors known to be required for efficient Ad-mediated gene delivery. Undifferentiated NT2 cells and NT2/Ns were infected by Ad expressing green fluorescent protein at an efficiency of 33% and 17%, respectively percentages much lower than the 92% infectivity obtained from a human non-neuronal cell line A549 cells. This relatively low infectivity of NT2/Ns might be caused by the extremely low expression of integrin subunit beta3 and the reduced expression of beta5 during differentiation. The expression of coxsackie-Ad receptor (CAR) was relatively high and remained constant during differentiation. Blocking CAR receptor using an antibody specific against CAR reduced Ad infectivity in a dose-dependent manner. These observations suggest that modulating the expression of integrin subunits beta3/5 or the functional heterodimer alphavbeta3/5 in human NT2/Ns may enhance adenoviral infectivity of NT2/Ns.