Joint superpixel and Transformer for high resolution remote sensing image classification.

Deep neural networks combined with superpixel segmentation have proven to be superior to high-resolution remote sensing image (HRI) classification. Currently, most HRI classification methods that combine deep learning and superpixel segmentation use stacking on multiple scales to extract contextual information from segmented objects. However, this approach does not take into account the contextual dependencies between each segmented object. To solve this problem, a joint superpixel and Transformer (JST) framework is proposed for HRI classification. In JST, HRI is first segmented into superpixel objects as input, and Transformer is used to model the long-range dependencies. The contextual relationship between each input superpixel object is obtained and the class of analyzed objects is output by designing an encoding and decoding Transformer. Additionally, we explore the effect of semantic range on classification accuracy. JST is also tested by using two HRI datasets with overall classification accuracy, average accuracy and Kappa coefficients of 0.79, 0.70, 0.78 and 0.91, 0.85, 0.89, respectively. The effectiveness of the proposed method is compared qualitatively and quantitatively, and the results achieve competitive and consistently better than the benchmark comparison method.

First-principle study on the stability of Cd passivates in soil.

The stable existence of heavy metals in soil under natural conditions is the core issue in heavy metal pollution solidification and remediation technology. However, the existing research is limited to soil passivation tests of different materials or biochar adsorption tests and cannot reveal the internal mechanism of functional groups of different compounds in soil passivation. This paper takes the common heavy metal ion Cd2+ as an example to analyze the stability of the combination of heavy metal ions and common ion groups in soil. The stability and existing form of Cd are analyzed by using first-principle calculations, and the free energy, band structure, and partial density of states of CdCO3, CdSO4, CdCl2, and CdSiO3 are computed. The stability of Cd binding to common anions in soil is determined. Results show the descending order of structural stability of cadmium compounds is CdSiO3, CdSO4, CdCO3, and CdCl2. SO42- and SiO32- can be used as preferred functional groups for cadmium pollution passivation. Anhydrous sodium sulfate and sodium silicate are promising passivators.

Vitamin C attenuates vancomycin induced nephrotoxicity through the reduction of oxidative stress and inflammation in HK-2 cells.

BACKGROUND: Oxidative stress is one of the possible mechanisms in vancomycin (VCM) induced nephrotoxicity. Some studies suggested that high dose Vitamin C (VC) has protective effect against the nephrotoxicity in mice, but the underlying molecular mechanism is not mentioned. We investigated the potential targets of high dose VC against oxidative stress and inflammation induced by VCM in renal tubular epithelial cells. METHODS: We conducted an in vitro study using an immortalized proximal tubule epithelial cell line from a normal adult human kidney (HK-2). RESULTS: VCM added to HK-2 cells caused an increase of cell death, oxidative stress and expression of inflammatory cytokines. Co-treatment with 0.5 and 1 mM VC attenuated 4-8 mM VCM induced cell death and increased the cell viability to 58-90%. VC significantly decreased lipid peroxidation and increased superoxide dismutase activity. The upregulations of NF-κB, TNF-α and IL-6 in HK-2 cells under 4 mM VCM were also reversed by 0.5 mM VC through the inhibition of oxidative stress. CONCLUSIONS: This study showed for the first time that VC can attenuate the VCM induced nephrotoxicity by decreasing lipid peroxidation and expression of inflammatory cytokines, and increasing superoxide dismutase 2 (SOD2) activity, this effect may relate to the regulation of ROS/NF-κB pathway.