OBJECTIVE:To elucidate the anti-inflammatory mechanisms of modified Shoutai pills(改良寿胎丸,MSTP)in miscarriages,we performed transcriptome sequencing on the decidua and placental tissues of pregnancy mice.METHODS:Th...OBJECTIVE:To elucidate the anti-inflammatory mechanisms of modified Shoutai pills(改良寿胎丸,MSTP)in miscarriages,we performed transcriptome sequencing on the decidua and placental tissues of pregnancy mice.METHODS:The therapeutic effects and antiinflammatory mechanisms of MSTP were studied in mice with lipopolysaccharide(LPS)-induced miscarriage.First,the effects of MSTP on pregnancy outcomes and the maternal-fetal interface,in LPS-induced miscarriage mice were examined.RNA sequencing was used to further investigate gene expression changes in LPS-induced miscarriage mice and to assess the effects of MSTP intervention.Finally,the expression levels of inflammation-related genes in the decidua and placental tissues were determined using quantitative real-time polymerase chain reaction(q RT-PCR).RESULTS:A high dose of MSTP significantly decreased the resorption rate(P<0.05)and reduced apoptosis of the decidua and placental tissues in mice.Gene ontology and Kyoto Encyclopedia of Genes and Genomes enrichment analyses showed that inflammatory and immune-related signals were enriched.q RT-PCR results confirmed that in decidual and placental tissues,MSTP reduced the gene expression levels of toll-like receptor 4(TLR4),nuclear factor kappa-B(NF-κB),c-Jun N-terminal kinase 1,p38,and tumor necrosis factor-α.CONCLUSIONS:In this study,we demonstrated that MSTP effectively prevented embryo loss with an antiinflammatory mechanism through downregulation of the TLR4-NF-κB/MAPK signaling pathway,in LPS-induced miscarriage mice model.To our knowledge,this is the first study to reveal the therapeutic mechanism of MSTP in LPS-induced miscarriage in mice.展开更多
针对当前多模态情感识别算法在模态特征提取、模态间信息融合等方面存在识别准确率偏低、泛化能力较差的问题,提出了一种基于语音、文本和表情的多模态情感识别算法。首先,设计了一种浅层特征提取网络(Sfen)和并行卷积模块(Pconv)提取...针对当前多模态情感识别算法在模态特征提取、模态间信息融合等方面存在识别准确率偏低、泛化能力较差的问题,提出了一种基于语音、文本和表情的多模态情感识别算法。首先,设计了一种浅层特征提取网络(Sfen)和并行卷积模块(Pconv)提取语音和文本中的情感特征,通过改进的Inception-ResnetV2模型提取视频序列中的表情情感特征;其次,为强化模态间的关联性,设计了一种用于优化语音和文本特征融合的交叉注意力模块;最后,利用基于注意力的双向长短期记忆(BiLSTM based on attention mechanism,BiLSTM-Attention)模块关注重点信息,保持模态信息之间的时序相关性。实验通过对比3种模态不同的组合方式,发现预先对语音和文本进行特征融合可以显著提高识别精度。在公开情感数据集CH-SIMS和CMU-MOSI上的实验结果表明,所提出的模型取得了比基线模型更高的识别准确率,三分类和二分类准确率分别达到97.82%和98.18%,证明了该模型的有效性。展开更多
基金Supported by the‘Pioneer’R&D Program of Zhejiang:Research on Key Technologies for the Development of Traditional Chinese Medicine New Drugs(No.2023C03004)National Key Research and Development Program of China:Mechanism Study and Clinical Exploration of Electroacupuncture Promoting Immune Normalization+4 种基金Supporting the Body and Inhibiting Cancer,Synergistic Programmed Death Receptor 1/Programmed Cell Death Ligand 1 Monoclonal Antibody therapy for Intestinal and Biliary Tumors(No.2023YFC3504600)Zhejiang Province Traditional Chinese Medicine Science and Technology Project:Clinical Metabolomics Based Discovery of Effective Markers for Nourishing Yin of Radix Ophiopogonis in the Treatment of Gestational Diabetes(No.GZY-ZJKJ-24076)the‘Pioneer’R&D Program of Zhejiang:Analysis of the Complex System of Traditional Chinese Medicine and Development of New Chinese Medicine Drugs-Analysis of the Complex Cross Organ Action Mode of Traditional Chinese Medicine for Anti-Coronary Heart Disease and Blood Stasis Syndrome and Development of New Drugs(No.2024C03106)Health and Medicinal Research Fund from Health Burden,Hong Kong Special Administrative Region of the People's Republic of China:Chinese Versus Western Medicine for Threatened Miscarriage:Abridged Secondary Publication(No.15160971)Transverse Research Project of Zhejiang University:Development of Traditional Chinese Medicine Big Health Formula for Reproductive Health(No.2023-KYY-A070350007)。
文摘OBJECTIVE:To elucidate the anti-inflammatory mechanisms of modified Shoutai pills(改良寿胎丸,MSTP)in miscarriages,we performed transcriptome sequencing on the decidua and placental tissues of pregnancy mice.METHODS:The therapeutic effects and antiinflammatory mechanisms of MSTP were studied in mice with lipopolysaccharide(LPS)-induced miscarriage.First,the effects of MSTP on pregnancy outcomes and the maternal-fetal interface,in LPS-induced miscarriage mice were examined.RNA sequencing was used to further investigate gene expression changes in LPS-induced miscarriage mice and to assess the effects of MSTP intervention.Finally,the expression levels of inflammation-related genes in the decidua and placental tissues were determined using quantitative real-time polymerase chain reaction(q RT-PCR).RESULTS:A high dose of MSTP significantly decreased the resorption rate(P<0.05)and reduced apoptosis of the decidua and placental tissues in mice.Gene ontology and Kyoto Encyclopedia of Genes and Genomes enrichment analyses showed that inflammatory and immune-related signals were enriched.q RT-PCR results confirmed that in decidual and placental tissues,MSTP reduced the gene expression levels of toll-like receptor 4(TLR4),nuclear factor kappa-B(NF-κB),c-Jun N-terminal kinase 1,p38,and tumor necrosis factor-α.CONCLUSIONS:In this study,we demonstrated that MSTP effectively prevented embryo loss with an antiinflammatory mechanism through downregulation of the TLR4-NF-κB/MAPK signaling pathway,in LPS-induced miscarriage mice model.To our knowledge,this is the first study to reveal the therapeutic mechanism of MSTP in LPS-induced miscarriage in mice.
文摘针对当前多模态情感识别算法在模态特征提取、模态间信息融合等方面存在识别准确率偏低、泛化能力较差的问题,提出了一种基于语音、文本和表情的多模态情感识别算法。首先,设计了一种浅层特征提取网络(Sfen)和并行卷积模块(Pconv)提取语音和文本中的情感特征,通过改进的Inception-ResnetV2模型提取视频序列中的表情情感特征;其次,为强化模态间的关联性,设计了一种用于优化语音和文本特征融合的交叉注意力模块;最后,利用基于注意力的双向长短期记忆(BiLSTM based on attention mechanism,BiLSTM-Attention)模块关注重点信息,保持模态信息之间的时序相关性。实验通过对比3种模态不同的组合方式,发现预先对语音和文本进行特征融合可以显著提高识别精度。在公开情感数据集CH-SIMS和CMU-MOSI上的实验结果表明,所提出的模型取得了比基线模型更高的识别准确率,三分类和二分类准确率分别达到97.82%和98.18%,证明了该模型的有效性。
