Optimized Dropkey-Based Grad-CAM: Toward Accurate Image Feature Localization.

Regarding the interpretable techniques in the field of image recognition, Grad-CAM is widely used for feature localization in images to reflect the logical decision-making information behind the neural network due to its high applicability. However, extensive experimentation on a customized dataset revealed that the deep convolutional neural network (CNN) model based on Gradient-weighted Class Activation Mapping (Grad-CAM) technology cannot effectively resist the interference of large-scale noise. In this article, an optimization of the deep CNN model was proposed by incorporating the Dropkey and Dropout (as a comparison) algorithm. Compared with Grad-CAM, the improved Grad-CAM based on Dropkey applies an attention mechanism to the feature map before calculating the gradient, which can introduce randomness and eliminate some areas by applying a mask to the attention score. Experimental results show that the optimized Grad-CAM deep CNN model based on the Dropkey algorithm can effectively resist large-scale noise interference and achieve accurate localization of image features. For instance, under the interference of a noise variance of 0.6, the Dropkey-enhanced ResNet50 model achieves a confidence level of 0.878 in predicting results, while the other two models exhibit confidence levels of 0.766 and 0.481, respectively. Moreover, it exhibits excellent performance in visualizing tasks related to image features such as distortion, low contrast, and small object characteristics. Furthermore, it has promising prospects in practical computer vision applications. For instance, in the field of autonomous driving, it can assist in verifying whether deep learning models accurately understand and process crucial objects, road signs, pedestrians, or other elements in the environment.

High-intensity interval training and medium-intensity continuous training may affect cognitive function through regulation of intestinal microbial composition and its metabolite LPS by the gut-brain axis.

AIMS: The gut-brain axis is the communication mechanism between the gut and the central nervous system, and the intestinal flora and lipopolysaccharide (LPS) play a crucial role in this mechanism. Exercise regulates the gut microbiota composition and metabolite production (i.e., LPS). We aimed to investigate the effects of high-intensity interval training (HIIT) and moderate-intensity continuous training (MICT) on cognitive function in C57BL/6 J mice through gut-brain axis regulation of gut microbiota composition and LPS displacement. MAIN METHODS: C57BL/6 J male mice were randomly divided into sedentary, HIIT, and MICT groups. After 12 weeks of exercise intervention, the cognitive function of the brain and mRNA levels of related inflammatory factors were measured. RNA sequencing, Golgi staining, intestinal microbial 16 s rDNA sequencing, and ELISA were performed. KEY FINDINGS: HIIT and MICT affect brain cognitive function by regulating the gut microbiota composition and its metabolite, LPS, through the gut microbiota-gut-brain axis. HIIT is suspected to have a risk: it can induce "intestinal leakage" by regulating intestinal permeability-related microbiota, resulting in excessive LPS in the blood and brain and activating M1 microglia in the brain, leading to reduced dendritic spine density and affecting cognitive function. SIGNIFICANCE: This study revealed a potential link between changes in the gut microbiota and cognitive function. It highlighted the possible risk of HIIT in reducing dendritic spine density and affecting cognitive function.

Estimation formula of finished bridge pre-camber in continuous rigid-frame bridges.

Continuous rigid-frame bridges are widely used, but the large deflection in the mid-span during operation has always been their disease. This problem is generally solved by setting the finished bridge pre-camber. There are many calculation methods for pre-camber, and the effects are different. In this paper, based on a large number of design parameters of continuous rigid-frame bridges obtained from the investigation, 18 finite element analysis models of different span combinations were established, and 30 sets of valid data were obtained under the action of multi-factor. The results show that the shrinkage and creep of concrete is the most important factor for the mid-span deflection of continuous rigid frame bridges, and the deflection amount has an obvious functional relationship with the span. The effect of prestress loss on mid-span deflection is second, and stiffness reduction has little effect on mid-span long-term deflection. In this paper, the least-squares method is used to perform polynomial fitting, and the fitting formula for the mid-span finished bridge pre-camber is finally obtained. The applicability of the calculation formula is proved by comparing it with the specification solution, the empirical solution, and the measured value.

Treadmill Exercise Modulates Intestinal Microbes and Suppresses LPS Displacement to Alleviate Neuroinflammation in the Brains of APP/PS1 Mice.

Neuroinflammation occurs throughout the pathogenesis of Alzheimer's disease (AD). Here, we investigated the effects of treadmill exercise on neuroinflammation in APP/PS1 transgenic AD mice and the potential involvement of microbe-gut-brain axis (MGB) mechanisms based on growing evidence that AD's pathogenesis is correlated with a deterioration in the function of gut microbiota. APP/PS1 transgenic AD mice were subjected to 12 weeks of treadmill exercise, followed by spatial memory tests. After the behavioral study, the amyloid (Aß) pathology, gut microbes and metabolites, bacterial lipopolysaccharide (LPS) displacement, and degree of neuroinflammation were analyzed. We found that this strategy of exercise enriched gut microbial diversity and alleviated neuroinflammation in the brain. Notably, exercise led to reductions in pathogenic bacteria such as intestinal Allobaculum, increases in probiotic bacteria such as Akkermansia, increased levels of intestine-brain barrier proteins, and attenuated LPS displacement. These results suggest that prolonged exercise can effectively modulate gut microbes and the intestinal barrier and thereby reduce LPS displacement and ultimately alleviate AD-related neuroinflammation.