针对输入数据特征多时负荷预测模型精度提升难的问题,文章提出一种并行多模型融合的混合神经网络超短期负荷预测方法。将卷积神经网络(convolutional neural network,CNN)与门控循环单元神经网络(gated recurrent unit neural network,G...针对输入数据特征多时负荷预测模型精度提升难的问题,文章提出一种并行多模型融合的混合神经网络超短期负荷预测方法。将卷积神经网络(convolutional neural network,CNN)与门控循环单元神经网络(gated recurrent unit neural network,GRU-NN)并行,分别提取局部特征与时序特征,将2个网络结构的输出拼接并输入深度神经网络(deep neural network,DNN),由DNN进行超短期负荷预测。最后应用负荷与温度数据进行预测实验,结果表明相比于GRUNN网络结构、长短期记忆(long short term memory,LSTM)网络结构、串行CNN-LSTM网络结构与串行CNN-GRU网络结构,所提方法具有更好的预测性能。展开更多
目的:探讨NSCLC患者癌组织中Ki-67与PD-L1表达的相关性及两者对患者预后的影响。方法:选取符合纳入标准的2012年1月至2018年8月在海军军医大学附属长海医院,手术确诊为NSCLC并进行免疫组化PD-L1和Ki-67检测的患者401例,收集其临床病理资...目的:探讨NSCLC患者癌组织中Ki-67与PD-L1表达的相关性及两者对患者预后的影响。方法:选取符合纳入标准的2012年1月至2018年8月在海军军医大学附属长海医院,手术确诊为NSCLC并进行免疫组化PD-L1和Ki-67检测的患者401例,收集其临床病理资料,定期生存随访,应用统计学方法分析Ki-67与PD-L1表达的相关性及两者对患者术后DFS和化疗后PFS的影响。结果:NSCLC组织中PD-L1和Ki-67表达阳性率分别为37.9%(152/401)和96.3%(386/401),单因素分析显示Ki-67为PDL1表达相关的影响因素(OR=0.33,95%CI=0.28～0.39,P<0.0001),曲线拟合分析显示Ki-67与PD-L1表达显著正相关,阈值效应分析、分段多因素Logistic和ROC曲线分析表明14%是Ki-67较适宜与PD-L1联用的阈值。Kaplan-Meier分析显示,术后DFS,Ki-67高表达组显著短于Ki-67低表达组[(21.88±11.25) vs (41.22±16.25)个月,P<0.0001],PD-L1阳性组显著短于PD-L1阴性组[(24.75±14.59) vs (38.27±16.75)个月,P<0.0001],Ki-67高表达/PD-L1阳性组与其余3组相比术后DFS最短[(20.57±11.33) vs(24.11±10.79) vs (36.00±16.79) vs (42.91±15.77)个月,P<0.0001];化疗PFS,Ki-67高表达组显著长于Ki-67低表达组[(7.70±3.01) vs(5.80±2.99)个月,P=0.016],PD-L1阳性组与阴性组相比差异无统计学意义[(7.04±3.21) vs (6.33±3.06)个月,P=0.22],Ki-67与PDL1联合测评,Ki-67高表达两组的PFS显著长于Ki-67低表达两组[(7.74±3.25) vs (7.43±2.38) vs(4.91±1.97) vs (6.02±3.19)个月,P=0.041]。结论:NSCLC组织中Ki-67与PD-L1表达呈正相关,Ki-67 14%是适宜与PD-L1联用的阈值,Ki-67和PD-L1均为患者预后不良的预测因子,两者联合对预后不良的预测有"叠加效应",同时Ki-67高表达患者对化疗的敏感性较好。展开更多
Given that only a subset of patients with colorectal cancer(CRC)benefit from immune checkpoint therapy,efforts are ongoing to identify markers that predict immunotherapeutic response.Increasing evidence suggests that ...Given that only a subset of patients with colorectal cancer(CRC)benefit from immune checkpoint therapy,efforts are ongoing to identify markers that predict immunotherapeutic response.Increasing evidence suggests that microbes influence the efficacy of cancer therapies.Fusobacterium nucleatum induces different immune responses in CRC with different microsatellite-instability(MSI)statuses.Here,we investigated the effect of F.nucleatum on anti-PD-L1 therapy in CRC.We found that high F.nucleatum levels correlate with improved therapeutic responses to PD-1 blockade in patients with CRC.Additionally,F.nucleatum enhanced the antitumor effects of PD-L1 blockade on CRC in mice and prolonged survival.Combining F.nucleatum supplementation with immunotherapy rescued the therapeutic effects of PD-L1 blockade.Furthermore,F.nucleatum induced PD-L1 expression by activating STING signaling and increased the accumulation of interferon-gamma(IFN-γ)^(+)CD8^(+)tumor-infiltrating lymphocytes(TILs)during treatment with PD-L1 blockade,thereby augmenting tumor sensitivity to PD-L1 blockade.Finally,patient-derived organoid models demonstrated that increased F.nucleatum levels correlated with an improved therapeutic response to PD-L1 blockade.These findings suggest that F.nucleatum may modulate immune checkpoint therapy for CRC.展开更多
The ChatGPT,a lite and conversational variant of Generative Pretrained Transformer 4(GPT-4)developed by OpenAI,is one of the milestone Large Language Models(LLMs)with billions of parameters.LLMs have stirred up much i...The ChatGPT,a lite and conversational variant of Generative Pretrained Transformer 4(GPT-4)developed by OpenAI,is one of the milestone Large Language Models(LLMs)with billions of parameters.LLMs have stirred up much interest among researchers and practitioners in their impressive skills in natural language processing tasks,which profoundly impact various fields.This paper mainly discusses the future applications of LLMs in dentistry.We introduce two primary LLM deployment methods in dentistry,including automated dental diagnosis and cross-modal dental diagnosis,and examine their potential applications.Especially,equipped with a cross-modal encoder,a single LLM can manage multi-source data and conduct advanced natural language reasoning to perform complex clinical operations.We also present cases to demonstrate the potential of a fully automatic Multi-Modal LLM AI system for dentistry clinical application.While LLMs offer significant potential benefits,the challenges,such as data privacy,data quality,and model bias,need further study.Overall,LLMs have the potential to revolutionize dental diagnosis and treatment,which indicates a promising avenue for clinical application and research in dentistry.展开更多
Drug discovery and development affects various aspects of human health and dramatically impacts the pharmaceutical market.However,investments in a new drug often go unrewarded due to the long and complex process of dr...Drug discovery and development affects various aspects of human health and dramatically impacts the pharmaceutical market.However,investments in a new drug often go unrewarded due to the long and complex process of drug research and development(R&D).With the advancement of experimental technology and computer hardware,artificial intelligence(AI)has recently emerged as a leading tool in analyzing abundant and high-dimensional data.Explosive growth in the size of biomedical data provides advantages in applying AI in all stages of drug R&D.Driven by big data in biomedicine,AI has led to a revolution in drug R&D,due to its ability to discover new drugs more efficiently and at lower cost.This review begins with a brief overview of common AI models in the field of drug discovery;then,it summarizes and discusses in depth their specific applications in various stages of drug R&D,such as target discovery,drug discovery and design,preclinical research,automated drug synthesis,and influences in the pharmaceutical market.Finally,the major limitations of AI in drug R&D are fully discussed and possible solutions are proposed.展开更多
文摘针对输入数据特征多时负荷预测模型精度提升难的问题,文章提出一种并行多模型融合的混合神经网络超短期负荷预测方法。将卷积神经网络(convolutional neural network,CNN)与门控循环单元神经网络(gated recurrent unit neural network,GRU-NN)并行,分别提取局部特征与时序特征,将2个网络结构的输出拼接并输入深度神经网络(deep neural network,DNN),由DNN进行超短期负荷预测。最后应用负荷与温度数据进行预测实验,结果表明相比于GRUNN网络结构、长短期记忆(long short term memory,LSTM)网络结构、串行CNN-LSTM网络结构与串行CNN-GRU网络结构,所提方法具有更好的预测性能。
文摘目的:探讨NSCLC患者癌组织中Ki-67与PD-L1表达的相关性及两者对患者预后的影响。方法:选取符合纳入标准的2012年1月至2018年8月在海军军医大学附属长海医院,手术确诊为NSCLC并进行免疫组化PD-L1和Ki-67检测的患者401例,收集其临床病理资料,定期生存随访,应用统计学方法分析Ki-67与PD-L1表达的相关性及两者对患者术后DFS和化疗后PFS的影响。结果:NSCLC组织中PD-L1和Ki-67表达阳性率分别为37.9%(152/401)和96.3%(386/401),单因素分析显示Ki-67为PDL1表达相关的影响因素(OR=0.33,95%CI=0.28～0.39,P<0.0001),曲线拟合分析显示Ki-67与PD-L1表达显著正相关,阈值效应分析、分段多因素Logistic和ROC曲线分析表明14%是Ki-67较适宜与PD-L1联用的阈值。Kaplan-Meier分析显示,术后DFS,Ki-67高表达组显著短于Ki-67低表达组[(21.88±11.25) vs (41.22±16.25)个月,P<0.0001],PD-L1阳性组显著短于PD-L1阴性组[(24.75±14.59) vs (38.27±16.75)个月,P<0.0001],Ki-67高表达/PD-L1阳性组与其余3组相比术后DFS最短[(20.57±11.33) vs(24.11±10.79) vs (36.00±16.79) vs (42.91±15.77)个月,P<0.0001];化疗PFS,Ki-67高表达组显著长于Ki-67低表达组[(7.70±3.01) vs(5.80±2.99)个月,P=0.016],PD-L1阳性组与阴性组相比差异无统计学意义[(7.04±3.21) vs (6.33±3.06)个月,P=0.22],Ki-67与PDL1联合测评,Ki-67高表达两组的PFS显著长于Ki-67低表达两组[(7.74±3.25) vs (7.43±2.38) vs(4.91±1.97) vs (6.02±3.19)个月,P=0.041]。结论:NSCLC组织中Ki-67与PD-L1表达呈正相关,Ki-67 14%是适宜与PD-L1联用的阈值,Ki-67和PD-L1均为患者预后不良的预测因子,两者联合对预后不良的预测有"叠加效应",同时Ki-67高表达患者对化疗的敏感性较好。
基金This work is supported by the National Natural Science Foundation of China(8177100280,81730102,81972221,and 81702037)Science and Technology Commission of Shanghai(20ZR1442800)Clinical research plan of SHDC(No.SHDC2020CR2069B,No.SHDC2020CR5006-002 and No.SHDC12019114).
文摘Given that only a subset of patients with colorectal cancer(CRC)benefit from immune checkpoint therapy,efforts are ongoing to identify markers that predict immunotherapeutic response.Increasing evidence suggests that microbes influence the efficacy of cancer therapies.Fusobacterium nucleatum induces different immune responses in CRC with different microsatellite-instability(MSI)statuses.Here,we investigated the effect of F.nucleatum on anti-PD-L1 therapy in CRC.We found that high F.nucleatum levels correlate with improved therapeutic responses to PD-1 blockade in patients with CRC.Additionally,F.nucleatum enhanced the antitumor effects of PD-L1 blockade on CRC in mice and prolonged survival.Combining F.nucleatum supplementation with immunotherapy rescued the therapeutic effects of PD-L1 blockade.Furthermore,F.nucleatum induced PD-L1 expression by activating STING signaling and increased the accumulation of interferon-gamma(IFN-γ)^(+)CD8^(+)tumor-infiltrating lymphocytes(TILs)during treatment with PD-L1 blockade,thereby augmenting tumor sensitivity to PD-L1 blockade.Finally,patient-derived organoid models demonstrated that increased F.nucleatum levels correlated with an improved therapeutic response to PD-L1 blockade.These findings suggest that F.nucleatum may modulate immune checkpoint therapy for CRC.
基金supported by the Research and Development Program,West China Hospital of Stomatology,Sichuan University(RD-02-202107)Sichuan Province Science and Technology Support Program(2022NSFSC0743)Sichuan Postdoctoral Science Foundation(TB2022005)grant to H.Huang.
文摘The ChatGPT,a lite and conversational variant of Generative Pretrained Transformer 4(GPT-4)developed by OpenAI,is one of the milestone Large Language Models(LLMs)with billions of parameters.LLMs have stirred up much interest among researchers and practitioners in their impressive skills in natural language processing tasks,which profoundly impact various fields.This paper mainly discusses the future applications of LLMs in dentistry.We introduce two primary LLM deployment methods in dentistry,including automated dental diagnosis and cross-modal dental diagnosis,and examine their potential applications.Especially,equipped with a cross-modal encoder,a single LLM can manage multi-source data and conduct advanced natural language reasoning to perform complex clinical operations.We also present cases to demonstrate the potential of a fully automatic Multi-Modal LLM AI system for dentistry clinical application.While LLMs offer significant potential benefits,the challenges,such as data privacy,data quality,and model bias,need further study.Overall,LLMs have the potential to revolutionize dental diagnosis and treatment,which indicates a promising avenue for clinical application and research in dentistry.
基金funded by the Natural Science Foundation of Zhejiang Province(LR21H300001)National Key R&D Program of China(2022YFC3400501)+4 种基金National Natural Science Foundation of China(22220102001,U1909208,81872798,and 81825020)Leading Talent of the“Ten Thousand Plan”-National High-Level Talents Special Support Plan of ChinaFundamental Research Fund of Central University(2018QNA7023)Key R&D Program of Zhejiang Province(2020C03010)“Double Top-Class”University(181201*194232101)。
文摘Drug discovery and development affects various aspects of human health and dramatically impacts the pharmaceutical market.However,investments in a new drug often go unrewarded due to the long and complex process of drug research and development(R&D).With the advancement of experimental technology and computer hardware,artificial intelligence(AI)has recently emerged as a leading tool in analyzing abundant and high-dimensional data.Explosive growth in the size of biomedical data provides advantages in applying AI in all stages of drug R&D.Driven by big data in biomedicine,AI has led to a revolution in drug R&D,due to its ability to discover new drugs more efficiently and at lower cost.This review begins with a brief overview of common AI models in the field of drug discovery;then,it summarizes and discusses in depth their specific applications in various stages of drug R&D,such as target discovery,drug discovery and design,preclinical research,automated drug synthesis,and influences in the pharmaceutical market.Finally,the major limitations of AI in drug R&D are fully discussed and possible solutions are proposed.
