目的建立院内感染风险预测模型,并分析其可行性,助力医院院内感染管理。方法选取76564例住院患者作为研究对象,按照7∶3的比例划分为训练集和测试集。根据住院期间是否发生感染分为感染组和非感染组,分析训练集两组临床相关资料,采用二...目的建立院内感染风险预测模型,并分析其可行性,助力医院院内感染管理。方法选取76564例住院患者作为研究对象,按照7∶3的比例划分为训练集和测试集。根据住院期间是否发生感染分为感染组和非感染组,分析训练集两组临床相关资料,采用二元logistic回归模型和深度神经网络(DNN)模型分析院内感染发生风险的相关影响因素并构建预测模型。结果训练集中626例发生感染,感染率为1.17%,测试集中289例发生感染,感染率为1.26%。受试者特征曲线(ROC)显示,DNN模型预测发生院内感染风险的曲线下面积(AUC)显著高于二元logistic回归模型(0.978 vs 0.686)。结论基于DNN的院内感染风险预测模型具有较高的准确性,有助于早期识别院内感染高危患者。展开更多
Under strong seismic excitation, a rigid block will uplift from its support and undergo rocking oscillations which may lead to （complete） overturning. Numerical and analytical solutions to this highly nonlinear vibr...Under strong seismic excitation, a rigid block will uplift from its support and undergo rocking oscillations which may lead to （complete） overturning. Numerical and analytical solutions to this highly nonlinear vibration problem are first highlighted in the paper and then utilized to demonstrate how sensitive the overturning behavior is not only to the intensity and frequency content of the base motion, but also to thc presence of strong pulses, to their detailed sequence, and even to their asymnletry. Five idealised pulses capable of representing ＂rupture-directivity＂ and ＂fling＂ affected ground motions near the fault, are utilized to this end ： the one-cycle sinus, the one-cycle cosinus, the Ricker wavelet, the truncated （T）-Ricker wavelet, and the rectangular pulse ＂Overturning-Acceleration Amplification＂ and ＂Rotation＂ spectra are introduced and presented. Artificial neural network modeling is then developed as an alternative numerical solution. The neural network analysis leads to closed-form expressions for predicting the overturning failure or survival of a rigid block, as a function of its geometric properties and the characteristics of the excitation time history. The capability of the developed neural network modeling is validated through comparisons with the numerical solution. The derived analytical expressions could also serve as a tool for assessing the destructiveness of near-fault ground motions, for structures sensitive to rocking with foundation uplift.展开更多
The National Strong Motion Observation Network System （NSMONS） of China is briefly introduced in this paper. The NSMONS consists of permanent free-field stations, special observation arrays, mobile observatories and...The National Strong Motion Observation Network System （NSMONS） of China is briefly introduced in this paper. The NSMONS consists of permanent free-field stations, special observation arrays, mobile observatories and a network management system. During the Wenchuan Earthquake, over 1,400 components of acceleration records were obtained from 460 permanent free-field stations and three arrays for topographical effect and structural response observation in the network system from the main shock, and over 20,000 components of acceleration records from strong aftershocks occurred before August 1, 2008 were also obtained by permanent free-field stations of the NSMONS and 59 mobile instruments quickly deployed after the main shock. The strong motion recordings from the main shock and strong aftershocks are summarized in this paper. In the ground motion recordings, there are over 560 components with peak ground acceleration （PGA） over 10 Gal, the largest being 957.7 Gal. The largest PGA recorded during the aftershock exceeds 300 Gal.展开更多
Background:Liver cirrhosis is a complex and heterogeneous disease,with a mortality rate of up to 57%,resulting in 1.03 million deaths per year.The prevalence of liver cirrhosis is on the rise.Patients with liver cirrh...Background:Liver cirrhosis is a complex and heterogeneous disease,with a mortality rate of up to 57%,resulting in 1.03 million deaths per year.The prevalence of liver cirrhosis is on the rise.Patients with liver cirrhosis have a variety of clinical phenotypes and are prone to various complications related to liver cirrhosis.Therefore,there is an urgent need to improve the early prevention and clinical management of cirrhosis and its complications.Methods:We use a precise medical approach to analyze and characterize the complex manifestations of cirrhotic patient populations,and we propose a Heterogeneous Medical Record Network(HEMnet)that includes electronic medical records,molecular interaction networks,and domain knowledge.We train the network embedding vector on HEMnet to obtain the low-dimensional vector representation of each node.With these vectors,we enriched the original medical record and identified six subtypes of cirrhosis.Results:Subtype 1 is characterized by heart disease,and subtype 2 has the strongest association with metabolic-related diseases.Subtype 3 was characterized by Chronic gastritis diseases.Subtype 4 was characterized by Liver cirrhosis-related complications-serous effusion.Subtype 5 had the strongest association with hepatitis-cirrhosis-related complications diseases and gallbladder disease.Subtype 6 was most strongly associated with Liver cirrhosis-related complications and hepatic carcinoma.By assessing the human disease-gene association of each subtype,the rich phenotype and biological functions of each subtype at the gene level were matched to the disease comorbidities and clinical differences we identified through EHR.Conclusion:Our approach demonstrates the utility of applying a precision medicine paradigm to cirrhosis and the prospect of extending this approach to other complexes,multifactorial diseases.展开更多
文摘目的建立院内感染风险预测模型,并分析其可行性,助力医院院内感染管理。方法选取76564例住院患者作为研究对象,按照7∶3的比例划分为训练集和测试集。根据住院期间是否发生感染分为感染组和非感染组,分析训练集两组临床相关资料,采用二元logistic回归模型和深度神经网络(DNN)模型分析院内感染发生风险的相关影响因素并构建预测模型。结果训练集中626例发生感染,感染率为1.17%,测试集中289例发生感染,感染率为1.26%。受试者特征曲线(ROC)显示,DNN模型预测发生院内感染风险的曲线下面积(AUC)显著高于二元logistic回归模型(0.978 vs 0.686)。结论基于DNN的院内感染风险预测模型具有较高的准确性,有助于早期识别院内感染高危患者。
文摘Under strong seismic excitation, a rigid block will uplift from its support and undergo rocking oscillations which may lead to （complete） overturning. Numerical and analytical solutions to this highly nonlinear vibration problem are first highlighted in the paper and then utilized to demonstrate how sensitive the overturning behavior is not only to the intensity and frequency content of the base motion, but also to thc presence of strong pulses, to their detailed sequence, and even to their asymnletry. Five idealised pulses capable of representing ＂rupture-directivity＂ and ＂fling＂ affected ground motions near the fault, are utilized to this end ： the one-cycle sinus, the one-cycle cosinus, the Ricker wavelet, the truncated （T）-Ricker wavelet, and the rectangular pulse ＂Overturning-Acceleration Amplification＂ and ＂Rotation＂ spectra are introduced and presented. Artificial neural network modeling is then developed as an alternative numerical solution. The neural network analysis leads to closed-form expressions for predicting the overturning failure or survival of a rigid block, as a function of its geometric properties and the characteristics of the excitation time history. The capability of the developed neural network modeling is validated through comparisons with the numerical solution. The derived analytical expressions could also serve as a tool for assessing the destructiveness of near-fault ground motions, for structures sensitive to rocking with foundation uplift.
基金NSFC Under Grant No. 90715038MOST of China Under Grant No. 2006BAC13B02
文摘The National Strong Motion Observation Network System （NSMONS） of China is briefly introduced in this paper. The NSMONS consists of permanent free-field stations, special observation arrays, mobile observatories and a network management system. During the Wenchuan Earthquake, over 1,400 components of acceleration records were obtained from 460 permanent free-field stations and three arrays for topographical effect and structural response observation in the network system from the main shock, and over 20,000 components of acceleration records from strong aftershocks occurred before August 1, 2008 were also obtained by permanent free-field stations of the NSMONS and 59 mobile instruments quickly deployed after the main shock. The strong motion recordings from the main shock and strong aftershocks are summarized in this paper. In the ground motion recordings, there are over 560 components with peak ground acceleration （PGA） over 10 Gal, the largest being 957.7 Gal. The largest PGA recorded during the aftershock exceeds 300 Gal.
基金This research was supported by the National Administration of Traditional Chinese Medicine(No.Z155080000004)This study was also supported by the National Administration of Traditional Chinese Medicine Department Memorandum(Chinese Medicine Science and Technology Memorandum[2021]36)project.
文摘Background:Liver cirrhosis is a complex and heterogeneous disease,with a mortality rate of up to 57%,resulting in 1.03 million deaths per year.The prevalence of liver cirrhosis is on the rise.Patients with liver cirrhosis have a variety of clinical phenotypes and are prone to various complications related to liver cirrhosis.Therefore,there is an urgent need to improve the early prevention and clinical management of cirrhosis and its complications.Methods:We use a precise medical approach to analyze and characterize the complex manifestations of cirrhotic patient populations,and we propose a Heterogeneous Medical Record Network(HEMnet)that includes electronic medical records,molecular interaction networks,and domain knowledge.We train the network embedding vector on HEMnet to obtain the low-dimensional vector representation of each node.With these vectors,we enriched the original medical record and identified six subtypes of cirrhosis.Results:Subtype 1 is characterized by heart disease,and subtype 2 has the strongest association with metabolic-related diseases.Subtype 3 was characterized by Chronic gastritis diseases.Subtype 4 was characterized by Liver cirrhosis-related complications-serous effusion.Subtype 5 had the strongest association with hepatitis-cirrhosis-related complications diseases and gallbladder disease.Subtype 6 was most strongly associated with Liver cirrhosis-related complications and hepatic carcinoma.By assessing the human disease-gene association of each subtype,the rich phenotype and biological functions of each subtype at the gene level were matched to the disease comorbidities and clinical differences we identified through EHR.Conclusion:Our approach demonstrates the utility of applying a precision medicine paradigm to cirrhosis and the prospect of extending this approach to other complexes,multifactorial diseases.
