69 results about "Spatial learning" patented technology

The invention discloses a multi-frame photoacoustic image reconstruction method based on optical flow alignment and depth feature fusion. The method comprises the following steps: S1, obtaining photoacoustic signal data; s2, reconstructing a photoacoustic cross-sectional image; s3, performing optical flow calculation and image alignment; and S4, training the deep feature fusion network. According to the method, through optical flow motion correction and a potential space learning mechanism, space-time information and complementary features in the aligned multiple frames of images are dynamically integrated, complementary information in the multiple frames of images is adaptively fused, noise is suppressed, and finally reconstruction of high signal-to-noise ratio and high spatial resolution images of biological tissues is achieved. Experimental results show that the method can significantly improve image quality, recover image distortion and detail loss caused by motion and noise, and provide a new effective scheme for promoting robust clinical application of a photoacoustic imaging technology.

The invention provides bacillus licheniformis F0726 for improving neuroinflammation of Alzheimer's disease and application of the bacillus licheniformis F0726. Relates to the microbial field. Experiments prove that the bacillus licheniformis F0726 can improve spatial learning and memory impairment of AD model mice and reduce TNF-alpha and IL-1beta in serum of the AD mice in Alzheimer's disease model mice induced by injecting A beta oligomer into the brain; according to the invention, the content of IL-10 is increased, the alteration of AD mouse intestinal flora is improved, and a new way and potential probiotic resources are provided for developing medicines related to prevention and treatment of Alzheimer's disease.

The invention discloses an emotion recognition method based on multi-band Riemannian manifold space learning, and belongs to the technical field of biological signal recognition, and the method comprises the steps: carrying out the self-adaptive frequency band division of an original signal through variational mode decomposition, restraining the mode aliasing, and extracting a multi-band sub-signal; further mapping the sub-signals to a Riemannian manifold space by using a multivariate phase space reconstruction + Riemannian manifold regularization Fisher discriminant analysis module, representing the geometrical characteristics of the signals through a covariance matrix, enhancing the characteristic separability by combining tangent space projection and discriminant analysis, and fusing the characteristic vectors by using an attention mechanism to obtain a multi-element phase space reconstruction and Riemannian manifold regularization Fisher discriminant analysis model; and finally, inputting the fused feature vectors into a classification model constructed by a graph convolutional neural network, and carrying out classification processing. According to the method, the problems of insufficient utilization of frequency band information and weak manifold feature discrimination in traditional emotion recognition are effectively solved, and the recognition precision and robustness are remarkably improved in scenes such as physiological signals.

The invention provides a circular RNA circ0004801 and an application of the circular RNA circ0004801 in treatment of an Alzheimer's disease. The nucleotide sequence of the circular RNA circ0004801 is as shown in SEQ ID NO: 1 (Sequence Identity Number 1). The expression of the circular RNA circ0004801 in the hippocampal tissue of the AD mouse is obviously up-regulated, and the circular RNA circ0004801 is used as a'sponge 'of miR-7688-5p and plays a role by adjusting the TTBK1 expression and tau phosphorylation level. By knocking down the circular RNA circ0004801, the spatial learning and memory of a 3 * Tg mouse can be obviously improved, and the tau protein phosphorylation level of a hippocampus region and the NFTs density of a hippocampus DG region can be obviously reduced. Therefore, the circular RNA circ0004801 and the miR-7688-5p can be used as new targets for treating the AD (Alzheimer's disease). According to the invention, it is found for the first time that circular RNA circ0004801 regulates TTBK1 through a ceRNA mechanism, so that excessive phosphorylation of tau is driven, and a new molecular target is provided for AD early intervention treatment based on circRNA and miRNA.

A method for evaluating the spatial learning and memory ability of tree shrews using a cheese board maze, which includes three steps: experimental preparation, experimental training, and determination of modeling indicators. In experimental preparation, various parameters such as the cheese board maze device, experimental layout, and environmental settings are optimized and designed. In experimental training, training is carried out before and after hippocampal lesion. A behavioral paradigm for evaluating the spatial memory of tree shrews is successfully established, and the modeling criteria and indicators are reasonably formulated. The successful establishment of the spatial learning and memory tree shrew model of the present invention provides a behavioral experimental method applicable to tree shrews for subsequent neuroscience research such as constructing AD models, in vivo electrophysiological recordings, and optical recordings, and can also provide more powerful scientific data for researchers in the research of multiple disciplines and fields such as spatial coding, learning and memory, new drug development, screening, and evaluation.

The invention provides a traditional Chinese medicine preparation for improving or treating cognitive impairment, and relates to the technical field of traditional Chinese medicine pharmacy, and the traditional Chinese medicine preparation is prepared from morinda officinalis, eclipta, semen boitae, fructus alpiniae oxyphyllae, bighead atractylodes rhizome, valerian, turmeric, rhizoma acori graminei, leech, rhizoma anemarrhenae, dendrobium officinale, endothelium corneum gigeriae galli, hematoxylon and honey-fried licorice root. The traditional Chinese medicine preparation shows remarkable neuroprotection and cognition improvement effects on the cell level and the animal level, in the cell level, the preparation can effectively resist Abeta1-42 induced cell damage, and the protection effect is achieved through multiple ways of improving the cell activity, enhancing the oxidation resistance, inhibiting neuroinflammation and the like; in animal experiments, the preparation can comprehensively improve cognitive impairment of APP / PS1 model mice, including relieving behavioral abnormality, improving spatial learning and memory ability, enhancing new object recognition ability, and effectively reducing deposition of A beta in hippocampus on a molecular level. The invention provides a new thought for improving or treating cognitive impairment.

An experimental method for evaluating the spatial learning and memory ability of tree shrews in an eight-arm maze includes four steps: screening of animals before the experiment, preparation of experimental equipment, experimental operation, and modeling index. In the preparation of experimental equipment, various parameters such as ambient temperature, humidity, room built-in, and maze environment are optimized and set. In the experimental operation steps, the training of tree shrews is repeatedly demonstrated through multiple experiments in terms of training steps, number of times, time, frequency, and reward settings. In the formal training of the eight-arm maze, the experimental method provided by the present invention significantly reduces the number of times the tree shrews enter the arm without rewards (reference memory errors); the number of repeated entries into the arm with rewards (working memory errors) is significantly reduced; the average number of days for the 12 tree shrews experimented with the experimental method provided by the present invention to reach the modeling standard is basically consistent with the number of days entering the platform period, and the modeling standard is reasonably formulated.

The invention discloses a flavone derivative delivery system of targeted EGFR (Epidermal Growth Factor Receptor) protein as well as a preparation method and application of the flavone derivative delivery system, and belongs to the technical field of medicinal chemistry. The flavonoid derivative HT11 with anti-A beta and antioxidant activity is synthesized firstly, the depolymerization effect of the flavonoid derivative HT11 on A beta 1-42 reaches up to 92.40%, and the flavonoid derivative HT11 has good blood-brain barrier permeability. Then, the HT11 is loaded by taking the graphene oxide quantum dot as a carrier to obtain the HT11 coated GOQDs, so that the defect of low solubility of the HT11 is effectively overcome, and the targeting property and the curative effect of the HT11 coated GOQDs can be enhanced. Experimental results show that the HT11-coated GOQDs effectively solve the problem of reduction of AD animal space learning and memory ability, and possibility is provided for application of the HT11-coated GOQDs in the field of biological medicine.

This invention relates to the fields of aerospace vehicle overall design, and discloses a structured subspace learning-based solid rocket performance-cost analysis method, comprising: constructing a model sample dataset; constructing a structured sparse subspace feature selection model by combining a similarity graph matrix, and solving it to obtain an optimized row sparse orthogonal projection matrix, wherein the row index of the non-zero row corresponds to the feature parameter in the model sample as the selected cost driving factor; constructing a training sample dataset for a surrogate model; calculating the corresponding specific delivery cost as the output response based on the training samples, and training the surrogate model; for each cost driving factor, constructing a prediction sample dataset by random sampling; using a sequential quadratic programming method to find the optimal prediction sample in the prediction sample dataset, and determining the prediction sample with the minimum specific delivery cost as the final solid rocket design scheme.

The invention relates to the technical field of learning and memory ability evaluation, and provides a mouse hippocampus learning and memory ability comprehensive evaluation method and system based on a Morris water maze, the method is different from a traditional method for evaluating learning and memory only through an incubation period, a mouse behavior character stability and swimming skill dual evaluation mechanism is innovatively introduced, and the method is suitable for comprehensive evaluation of learning and memory ability of a mouse hippocampus. And in combination with a training adaptation period, behavior path analysis and weighted scoring of multi-dimensional behavior indexes, precise quantification of the hippocampus dependency space learning and memory ability is realized. The experimental indexes comprise swimming speed, path length, target quadrant residence time and incubation period, and performing standardized calculation after weight assignment to generate a memory ability evaluation score. According to the invention, the scientificity, stability and experimental repeatability of an evaluation result are improved, and the method is suitable for mouse cognitive function evaluation under various research situations such as drug intervention, neurodegenerative disease models and the like.

A pancreatic postoperative diabetes prediction system based on supervised deep subspace learning. A deep convolutional neural network and the MITK software are used to obtain postoperative residual pancreas area, so as to taken as the region-of-interest. Traditional image radiomics features and deep semantic features are extracted from the residual pancreas area, and a high-dimensional image feature set is constructed. Clinical factors related to diabetes, including pancreatic excision rate, fat and muscle tissue components, demographic information and living habits are extracted, and a clinical feature set is constructed. Based on a supervised deep subspace learning network, image and clinical features are represented and fused in subspace in dimensionality reduction, while a prediction model is trained to mine sensitive features highly relevant to the prediction risk of a patient suffering postoperative diabetes mellitus with a high degree of automation and discriminative accuracy.

The invention discloses a strain of lactobacillus acidophilus for treating chronic cerebral ischemia, which is preserved in the China Center for Type Culture Collection (CCTCC) and has the preservation number of CCTCC NO: M 20251505. Animal experiments show that the strain can reduce the death rate of chronic cerebral ischemia model mice and improve neurological impairment, motion coordination and spatial learning and memory ability of the chronic cerebral ischemia model mice. Mechanism research shows that the strain can increase histone acetylation and BDNF secretion by inhibiting the activity of astrocyte HDAC1, and meanwhile, can inhibit an NF-kappa B signal channel and reduce inflammatory factor release by activating a microglial cell GPR120 receptor, so that a neuroprotective effect is achieved through an'intestine-brain axis'. The invention provides a new candidate strain for the development of microbial preparations for preventing and / or treating chronic cerebral ischemia and related cognitive impairment.

The invention discloses a method for synergistically treating Alzheimer's disease through gene-modified human interdental stem cells, and belongs to the field of drug therapy. According to the method, healthy adult dental pulp mesenchymal stem cells (hDPSCs) are extracted, BDNF and NRF2 genes are introduced through lentiviral vectors, the BDNF and NRF2 genes have neurotrophic, antioxidant and anti-inflammatory capacities, and then the hDPSCs are combined with curcumin nanoparticles and injected to the hippocampus of an AD mouse through stereotactic injection. The BDNF can promote nerve regeneration, the NRF2 can improve oxidative stress and inflammation environment, and the curcumin nanoparticles regulate A beta metabolism and activate a BDNF signal channel. Experiments show that the combined therapy can significantly enhance hippocampal neurogenesis of AD mice and improve expression of neuron markers, and it is proved through Morse water maze tests that the combined therapy can improve the spatial learning and memory ability of the mice. The method breaks through the limitation of traditional single-target treatment, multi-dimensional pathological intervention is achieved, the stem cells are rich in source and high in safety, and a new strategy is provided for treatment of the Alzheimer's disease.

The invention discloses application of lactobacillus rhamnosus to preparation of a product for improving cognitive impairment or neuroinflammation caused by chronic fatigue. The lactobacillus rhamnosus BD4047 is an edible safety strain, and can improve cognitive impairment of a chronic fatigue animal model by reversing short-term spatial learning memory ability and working memory ability; by improving the indole level, lowering the IL-33 level and balancing intestinal flora, the neuroinflammation of a chronic fatigue animal model is relieved, so that the lactobacillus rhamnosus BD4047 can be used as an active ingredient for developing functional foods such as fermented milk, lactobacillus beverages, probiotic powder, viable bacteria water, milk powder and cheese, and has large-scale production and market popularization prospects.

The invention discloses a nutritional composition for promoting brain development and application thereof, and relates to the technical field of biological medicines. The compound composition is selected from a composition A (melatonin + uridylic acid + inositol), a composition B (melatonin + uridylic acid) and a composition C (uridylic acid + inositol). Experiments prove that the combination B can significantly improve elevated anxiety behaviors and learning and memory ability caused by rhythm disorder, the combination C can significantly improve open field anxiety behaviors caused by rhythm disorder, and the combination A can significantly relieve anxiety behaviors (including elevated and open field) and spatial learning and memory ability disorder caused by rhythm disorder at the same time. The active components in the compound composition disclosed by the invention are remarkable in synergistic effect, high in safety and good in biocompatibility, and can be used for preparing medicines, food additives or feed additives for improving rhythm disorder related brain dysplasia.

The invention discloses application of Cacumen biotae-derived extracellular vesicles in preparation of drugs for treating anxiety, depression or insomnia. The EVs are separated and purified from cacumen biotae through a specific differential ultracentrifugation method, the EVs have a typical double-layer membrane structure, the average particle size is about 100 nm, about 1012 EVs can be extracted from each gram of cacumen biotae, and 62 volatile components are contained. Cacumen biotae EVs can be taken by PC12 cells, and corticosterone-induced neuron damage can be relieved; for chronic constraint stress model mice, anxiety behaviors and spatial learning and memory disorders can be improved, and central nervous inflammation can be inhibited; for chronic unpredictable mild stress model mice, the weight, the sucrose preference rate and the cognitive ability can be recovered, and the levels of 5-hydroxytryptamine and gamma-aminobutyric acid are increased; and for 4-chloro-DL-phenylalanine induced insomnia model mice, the sleep condition can be improved, the neurotransmitter level can be recovered, and the effect is equivalent to that of positive drugs such as citalopram, fluoxetine, diazepam and the like.

The invention discloses calf animal coccus PD13 for improving Alzheimer's disease and application of the calf animal coccus PD13, and relates to the technical field of biology. Experiments prove that in an Alzheimer's disease model mouse induced by injecting A beta oligomer into the brain, the Azococcus calvus PD13 can improve spatial learning and memory disorder and autonomous exploration ability of the AD model mouse and relieve damage to neurons in a hippocampal region of the Alzheimer's disease model mouse, so that the effect of protecting the neurons is achieved, and the Azococcus calvus PD13 can be used for preparing a medicine for treating the Alzheimer's disease. According to the present invention, with the application of the animal coccus calves, the content of glutathione in mouse serum and the enzyme activity of superoxide dismutase are increased, and the content of malondialdehyde is reduced, such that the animal coccus calves provides good prevention and treatment effects on the Alzheimer's disease, and the new approach is provided for developing the Alzheimer's disease-related drugs.

The invention discloses a cross-modal bearing fault diagnosis method based on self-supervised space learning, which mainly constructs a unified optimization model of class labels and a cross-modal space projection direction, so that the self-supervised learning of the class labels can be realized under the condition of unsupervised fault diagnosis, and the accuracy of fault diagnosis is effectively improved. The specific implementation process is as follows: (1) constructing a cross-modal self-supervision space learning model by using cross-modal class label information and a related space theory; (2) theoretically deriving indirect representation of a cross-modal class center, and further obtaining an analytical solution of a class label in a learned cross-modal consistent space; and (3) obtaining cross-modal fault features of the fault sample data on the basis of the cross-modal self-supervised space learning projection direction, and inputting the low-dimensional discrimination unified fault features into the classifier to obtain a final bearing fault diagnosis result. Compared with the prior art, the fault diagnosis method is more effective and robust.

The present invention provides a Chinese medicine composition for treating or preventing Alzheimer's disease, and its preparation method and application, wherein the Chinese medicine composition comprises the following raw materials in parts by mass: 10-20 parts of ginseng, 10-20 parts of polygala, 10-20 parts of calamus, 10-20 parts of tuckahoe, 10-20 parts of yam, 10-20 parts of polygonatum, 10-20 parts of green tea pigments and 10-20 parts of quercetin, and optionally comprises 0.1-0.5 parts of black pepper. The Chinese medicine composition of the present invention has a significant therapeutic effect on 5XFAD transgenic AD mice, which is specifically manifested in that daily oral administration of the Chinese medicine composition of the present invention improves the spatial learning and memory disorders of dementia mice, reduces the deposition of amyloid plaques in the dentate gyrus of the hippocampus of mice, and significantly inhibits the generation of Aβ without causing liver and kidney toxicity and weight loss. Therefore, the Chinese medicine composition of the present invention is a safe and effective method for treating AD.

The invention relates to the technical field of traditional Chinese medicine compositions and cosmetics, and discloses an anti-aging traditional Chinese medicine compound composition which is prepared from cistanche, radix achyranthis bidentatae, eucommia ulmoides, dendrobium flowers, polygala tenuifolia, roses, Chinese yam, angelica sinensis, salvia miltiorrhiza and liquorice. The traditional Chinese medicine compound composition prepared by the method can significantly inhibit the activity of an intracellular senescence marker beta-galactosidase and the expression of P16 or P21, improve the oxidation resistance of cells, improve the spatial learning and memory ability of senescent mice and recover the cardiac ejection ability of mice, exert a significant anti-senescence effect, and has good application prospects. The method has great popularization and application values.

The invention relates to a composition for improving cognitive decline and application thereof, a mouse aging model is constructed through a D-galactose injection method, and then mice are randomly divided into a control group, a D-gal model group, a DHA group, and a DHA, folic acid and phosphatidylserine combined intervention group (n = 12). And testing. Test results show that the combined intervention of DHA, folic acid and phosphatidylserine has a certain improvement effect on the cognitive function of the brain of an aging model mouse, can improve the inflammatory response in hippocampus tissues and the expression conditions of neurotrophic factors and synaptic associated proteins, regulates the in-vivo antioxidant level, and has a good anti-aging effect. And the new object identification and spatial learning memory ability of the mouse is improved. Therefore, the invention provides a theoretical basis for improving cognition and realizing functional dairy product development and industrial production through combined intervention of DHA, folic acid and phosphatidylserine.

The invention relates to the field of bearing fault diagnosis, in particular to a variable-speed bearing fault diagnosis method based on physical feature guidance and dynamic tensor projection. The method comprises the following steps: acquiring a speed change historical time domain vibration signal when a bearing has a fault; an AWFE-CDSP-DDPL fault diagnosis model is constructed, and a fault diagnosis model is constructed; historical time domain vibration signals are divided into a training set and a test set by adopting a sliding window technology, and the AWFE-CDSP-DDPL fault diagnosis model is trained by using the training set; and obtaining a current variable speed time domain vibration signal of the bearing, and inputting the current variable speed time domain vibration signal into the trained AWFE-CDSP-DDPL fault diagnosis model to complete fault diagnosis. By means of the signal processing technology and the subspace learning method, the convolutional neural network is endowed with certain physical significance, meanwhile, the influence of the rotating speed is eliminated in network module construction, the bearing fault diagnosis precision under the variable speed condition is improved, and the research and economic production efficiency is improved.

The application relates to the technical field of animal learning ability testing, in particular to a ring-shaped runway for detecting the space learning and memory ability of small animals, which comprises a ring-shaped bottom plate, an inner baffle is arranged on one side of the inner ring of the ring-shaped bottom plate, an outer baffle is arranged on one side of the outer ring of the ring-shaped bottom plate, an experimental area for placing experimental objects is arranged on the ring-shaped bottom plate, and a water drinking area for water drinking of research objects is arranged on the ring-shaped bottom plate. The application has the beneficial effect that in the behavioral study of animal space memory, the application can reduce the information interference suffered by the research objects and can truly reflect the space memory of the research objects under normal conditions.

The present invention discloses the use of Xinshubao tablets in preparing a drug for preventing and / or treating senile dementia. The present invention uses an AD animal model experiment constructed with scopolamine to determine that Xinshubao tablets improve the spatial learning and memory function of senile dementia mice by improving cortical and hippocampal tissue pathological damage, hippocampal neuron ultrastructure damage, and the cholinergic nervous system in the brain, and improving the neurotrophic state and synaptic function in the brain. The application of Xinshubao tablets in preventing and treating senile dementia has broad application prospects.

This invention discloses a remote sensing image wind turbine detection method based on attention mechanisms and feature alignment, relating to the fields of computer vision and remote sensing. The method includes inputting the acquired remote sensing wind turbine image into a deep residual network; sequentially passing the image through layers C2, C3, C4, and C5 of the deep residual network to extract feature maps at different scales; performing feature enhancement on the feature map extracted by layer C5 using a thresholding attention module with cross-spatial learning; inputting the features extracted by layers C3 and C4, and the features enhanced by layer C5 (using layer S2), into a feature pyramid after passing through a two-branch semantic flow alignment module; connecting each layer of the pyramid feature layers aligned by process S3 to a detection head, and finally outputting the detection result. This invention helps reduce background interference while preserving local information of the wind turbine, and utilizes spatial transformation techniques for precise alignment, thereby improving the accuracy of feature fusion and increasing the accuracy of the detection results.

Pharmacological or gene inactivation of MAGL reduces neuroinflammation and neuropathology in AD animal models. However, global MAGL inactivation induces functional tolerance of CB1R, which reduces the effectiveness of drug therapy. Evidence has shown that selective inactivation of MAGL in astrocytes rather than neurons reduces neuropathological and synaptic and cognitive disorders in TBI. Specifically, results show that AD neuropathology is weakened by inactivation of the astrocyte MAGL, and LTP, spatial learning and memory deterioration of AD animals are prevented. This shows that the neuroprotection effect of global MAGL inactivation is mainly caused by limiting the degradation of 2-AG in astrocytes rather than neurons. Thus, the selective inactivation of the astrocyte MAGL provides a better therapeutic outcome for AD because this minimizes the potential side effects arising from the global inactivation-induced disruption of 2-AG degradation in neurons and other peripheral tissues. For this purpose, an AAV-mediated gene silencing method for selectively knocking out MAGL in astrocytes is proposed.

The invention provides application of a compound shown in a formula I or pharmaceutically acceptable salt thereof and / or hyodeoxycholic acid or pharmaceutically acceptable salt thereof in preparation of a medicine for preventing, delaying or treating cognitive dysfunction and / or motor dysfunction. Muscle strength, athletic ability, active exploration ability and spatial learning and memory ability can be effectively improved, antioxidant pressure resistance and proinflammatory state can be improved, expression of neurotrophic factors is increased, and nerve injury repair is improved.

The invention belongs to the technical field of microorganisms, and particularly relates to application of lactobacillus plantarum LLY-606 in preparation of a product for improving the body capacity and a composition. The invention aims to solve the problem of lack of probiotics with the corresponding effects. The strain is proved to be capable of remarkably improving the survival rate of an aging model animal and comprehensively improving the working memory, spatial learning and recognition memory ability of the aging model animal. The action mechanism is that systemic neuroinflammation is effectively inhibited through an intestine-brain axis, and the level of proinflammatory factors in serum is remarkably reduced, so that a safe and effective new scheme for resisting aging and improving cognition is provided.

The invention relates to the related technical field of vision regulation and control, in particular to a regulation and control device based on low-frequency flickering red light and a use method. The regulation and control device comprises a light source unit, a control unit, a carrier structure and an auxiliary module; the low-frequency flicker mode simulates the physiological rhythm of a visual system, the visual neural pathway and ciliary muscles are more easily activated compared with continuous light, the visual acuity improvement amplitude is improved by 30% or above compared with traditional red light equipment, and the adjustment lag is reduced by 25%; the nerve activation effect of the 40Hz frequency group is consistent with the PV neuron experiment of the Southern Science and Technology University, the space learning ability can be enhanced, the adaptation rate of sensitive people reaches 98% or above, and the short-time design of 3-5 minutes fits fragmentation time and adapts to home, travel and office scenes; the wavelength stability of the LED light source is + / -10nm, and the modular control design enables the consistency error of batch production to be 1t; 5%; aBS + PC materials are anti-falling and durable, heat dissipation holes are matched with aluminum cooling fins, it is ensured that the running temperature of equipment is smaller than or equal to 40 DEG C, and large-scale production and daily use reliability are both considered.

The invention discloses application of an NLRP3 inhibitor in preparation of a medicine for treating cognitive impairment of diabetes, and belongs to the technical field of medicines. And the NLRP3 inhibitor is 1, 2, 4-trimethoxybenzene (ZY16), and the NLRP3 inhibitor is 1, 2, 4-trimethoxybenzene. Experimental results show that 1, 2, 4-trimethoxybenzene can effectively relieve cognitive impairment of diabetes mellitus by inhibiting activation of NLRP3 inflammasomes, reducing blood sugar level, improving insulin resistance and glycolipid metabolism, relieving neuroinflammation and oxidative stress, improving pathological damage of hippocampus tissues and improving spatial learning and memory ability of model animals. The invention provides a new effective way for clinical treatment of cognitive impairment of diabetes, and has important clinical application value.