Postoperative neurocognitive impairment is an urgent problem with global aging accelerating. The prevention and treatment of postoperative neurocognitive impairment have been widely investigated but lack effective strategies. Low-intensity pulsed ultrasound (LIPUS), a non-invasive tool, has shown an effect on neuroprotection, but whether it could attenuate the postoperative neurocognitive impairment and the underlying mechanisms remains unknown. An experimental setup for LIPUS stimulation of the hippocampus was well established. A laparotomy model in aged mice was applied, and a Morris water maze was used to assess cognitive function. RT-qPCR and western blotting were used to detect levels of Piezo1, synapse-associated proteins in the hippocampus, respectively. Immunofluorescent staining was also used to determine the neural activation and Piezo1 expression. The results showed that LIPUS increased synapse-related proteins of the hippocampus and attenuated cognitive impairment in aged mice. Meanwhile, LIPUS suppressed the overexpression of Piezo1 in the hippocampus. We further found that LIPUS promoted Calpain1 activity and increased extracellular regulated protein kinases (Erk) phosphorylation. Our results suggested that LIPUS could improve cognitive impairment and increase hippocampal synaptogenesis through the Piezo1-mediated Calpain1/ Erk pathway. LIPUS could be used as an effective physical intervention to alleviate postoperative cognitive dysfunction in the aged population.

A DLP (data loss prevention) system usually arranges network monitors at the network boundary to perform network traffic capture, file parsing, and strategy matching procedures. Strategy matching is a key process to prevent corporate secret-related documents from leaking. This paper adopts the document fingerprint similarity detection method based on the SimHash principle and customizes the KbS (Keyword-based SimHash) fingerprint, PbS (Paragraph-based SimHash) fingerprint, and SoP (SimHash of Paragraph) fingerprint, three different feature extraction SimHash algorithms for strategy matching to detect. The parsed unstructured data is stored as a file type in.txt format, and then a file fingerprint is generated. Matching the established sensitive document library to calculate the Hamming distance between the fingerprints, the Hamming distance values under different modification degrees are summarized. The experimental results reveal that the hybrid algorithmic strategy matching rules with different levels and accuracy are established. This paper has a reference role for the leakage prevention research of enterprise sensitive data.

Classification and gradation system adopts different security protection schemes for different types of data by implementing classification and gradation management of data, which is an important pretechnical means for data security protection and prevention of data leakage. This paper introduces artificial intelligence classification, machine learning, and other means to learn and train enterprise documents according to the characteristics of enterprise sensitive data. The generated training model can intelligently identify and classify file streams, improving work efficiency and accuracy of classification and gradation. At the same time, the differences, advantages, and disadvantages of K-NN (K-Nearest Neighbors), DT (Decision Tree), and LinearSVC algorithms are compared. The experimental data shows that LinearSVC algorithm is applicable to high-dimensional data, with discrete, sparse data features and large number of features, which is more suitable for classification of sensitive data of enterprises.