Huatan Jiangzhuo decoction(HJD)is a combination of six traditional Chinese medicines that were used for lipid metabolism-related disorders,but its efficacy and underlying mechanisms have not been explored by modern re...Huatan Jiangzhuo decoction(HJD)is a combination of six traditional Chinese medicines that were used for lipid metabolism-related disorders,but its efficacy and underlying mechanisms have not been explored by modern research strategies.This study aimed to investigate the therapeutic role of HJD in determining the transcriptome level.Hyperlipidemia model was established by feeding Sprague–Dawley rats with high-fat diet.Differentially expressed genes(DEGs)were detected by high-through transcriptome sequencing,followed by gene ontology(GO)and Kyoto Encyclopedia of Genes and Genomes(KEGG)analysis.The total cholesterol(TC)and triglyceride(TG)levels in hyperlipidemia model rats were significantly increased,whereas high-density lipoprotein(HDL)concentration decreased when compared to normal rats,and HJD significantly downregulated TC concentrations and liver coefficient in the hyperlipidemia rats.Histology staining showed that HDJ greatly recovered the lipid accumulation in rat hepatic stellate cells and aortic arch vascular wall thickness of hyperlipidemia rats.One thousand nine hundred and thirty-six DEGs were identified in the HJD-treated hyperlipidemia rats,which were associated with various biological processes and signaling pathways such as peroxisome proliferator-activated receptors,AMP-activated Protein Kinase,and insulin signaling pathways.Quantitative reverse transcription–polymerase chain reaction further confirmed the downregulated expression of cholesterol 7-α-hydroxylase(CYP7A1),liver orphan receptor(LXRα),peroxisome proliferator-activated receptor gamma(PPARγ),and Sterol Response Element-Binding Protein 1c(SREBP1c)genes in hyperlipidemia rats treated with HJD.Our data first elucidated the gene expression profile of high-fat diet-induced hyperlipidemia in rats after HJD treatment,and lipid metabolism-related genes(CYP7A1,LXRα,PPARγ,and SREBP1c)may be potentially biomarkers for HJD-alleviated hyperlipidemia.展开更多
为了提高活体检测的准确度,提出一种基于改进MobileNetV2算法与频谱图的静默活体检测方法,该方法利用深度学习和图像处理技术,通过分析图像中人脸的生物特征和频谱特征来判断人脸的真实性。文中对Mobile Net V2算法进行改进,采用挤压和...为了提高活体检测的准确度,提出一种基于改进MobileNetV2算法与频谱图的静默活体检测方法,该方法利用深度学习和图像处理技术,通过分析图像中人脸的生物特征和频谱特征来判断人脸的真实性。文中对Mobile Net V2算法进行改进,采用挤压和激励(Squeeze and Excitation,SE)注意力机制和傅里叶频谱辅助学习的方法来提高静默活体检测的准确度。此外,在Print-Attack、OULU和NUAA数据集上进行实验评估,对Mobile Net V2+FFT+SE方法进行验证。实验结果显示,当R_(FP)=10^(-2)时,R_(TP)为97.23%,R_(EE)为2.07%,A_(UC)为96.27%,A_(CC)为92.31%。这些指标证明该方法具有优异的分类性能、较高的A_(UC)值和较低的错误率,同时此方法在实际应用中能够提高人脸识别的安全性,保护个人隐私和信息安全,并具有重要的应用前景。展开更多
基金the Natural Science Foundation of Guangdong Province,China(No.2020A1515010725).
文摘Huatan Jiangzhuo decoction(HJD)is a combination of six traditional Chinese medicines that were used for lipid metabolism-related disorders,but its efficacy and underlying mechanisms have not been explored by modern research strategies.This study aimed to investigate the therapeutic role of HJD in determining the transcriptome level.Hyperlipidemia model was established by feeding Sprague–Dawley rats with high-fat diet.Differentially expressed genes(DEGs)were detected by high-through transcriptome sequencing,followed by gene ontology(GO)and Kyoto Encyclopedia of Genes and Genomes(KEGG)analysis.The total cholesterol(TC)and triglyceride(TG)levels in hyperlipidemia model rats were significantly increased,whereas high-density lipoprotein(HDL)concentration decreased when compared to normal rats,and HJD significantly downregulated TC concentrations and liver coefficient in the hyperlipidemia rats.Histology staining showed that HDJ greatly recovered the lipid accumulation in rat hepatic stellate cells and aortic arch vascular wall thickness of hyperlipidemia rats.One thousand nine hundred and thirty-six DEGs were identified in the HJD-treated hyperlipidemia rats,which were associated with various biological processes and signaling pathways such as peroxisome proliferator-activated receptors,AMP-activated Protein Kinase,and insulin signaling pathways.Quantitative reverse transcription–polymerase chain reaction further confirmed the downregulated expression of cholesterol 7-α-hydroxylase(CYP7A1),liver orphan receptor(LXRα),peroxisome proliferator-activated receptor gamma(PPARγ),and Sterol Response Element-Binding Protein 1c(SREBP1c)genes in hyperlipidemia rats treated with HJD.Our data first elucidated the gene expression profile of high-fat diet-induced hyperlipidemia in rats after HJD treatment,and lipid metabolism-related genes(CYP7A1,LXRα,PPARγ,and SREBP1c)may be potentially biomarkers for HJD-alleviated hyperlipidemia.
文摘为了提高活体检测的准确度,提出一种基于改进MobileNetV2算法与频谱图的静默活体检测方法,该方法利用深度学习和图像处理技术,通过分析图像中人脸的生物特征和频谱特征来判断人脸的真实性。文中对Mobile Net V2算法进行改进,采用挤压和激励(Squeeze and Excitation,SE)注意力机制和傅里叶频谱辅助学习的方法来提高静默活体检测的准确度。此外,在Print-Attack、OULU和NUAA数据集上进行实验评估,对Mobile Net V2+FFT+SE方法进行验证。实验结果显示,当R_(FP)=10^(-2)时,R_(TP)为97.23%,R_(EE)为2.07%,A_(UC)为96.27%,A_(CC)为92.31%。这些指标证明该方法具有优异的分类性能、较高的A_(UC)值和较低的错误率,同时此方法在实际应用中能够提高人脸识别的安全性,保护个人隐私和信息安全,并具有重要的应用前景。
