目的使用logistic回归分析构建多个临床指标联合预测重症脂源性胰腺炎的预测模型,并评价其预测效果。方法纳入首都医科大学宣武医院2012年11月至2019年12月收住院的207例高甘油三酯血症性胰腺炎(hypertriglyceridemic pancreatitis,HTGP...目的使用logistic回归分析构建多个临床指标联合预测重症脂源性胰腺炎的预测模型,并评价其预测效果。方法纳入首都医科大学宣武医院2012年11月至2019年12月收住院的207例高甘油三酯血症性胰腺炎(hypertriglyceridemic pancreatitis,HTGP)患者作为研究对象,根据胰腺炎的严重程度将患者分为重症急性胰腺炎(severe acute pancreatitis,SAP)组(44例)和中重度胰腺炎(moderately severe acute pancreatitis,MSAP)+轻度胰腺炎(mild acute pancreatitis,MAP)组(163例)。比较两组患者一般临床资料,入院第1天中性粒细胞/淋巴细胞比值(neutrophil to lymphocyte ratio,NLR)、淀粉酶、脂肪酶、甘油三酯、C反应蛋白/白蛋白比值(C-reactive protein to albumin ratio,CAR)、降钙素原(procalcitonin,PCT)水平,CT严重指数(CT severity index,CTSI)评分、急性生理与慢性健康Ⅱ(acute physiology and chronic health evaluation-Ⅱ,APACHE-Ⅱ)评分、RANSON评分,住院天数、预后情况。应用logistic回归分析纳入差异有显著性的变量作为协变量,将是否发生SAP作为二分类结局变量,得出SAP的相关独立变量并拟合多个独立变量构建新的联合预测模型。构建受试者工作特征曲线(receiver operating characteristic curve,ROC)分析并评估联合预测模型及各独立变量对SAP的预测性能,得到曲线下面积(area under the curve,AUC)、最佳临界值、敏感度、特异度等参数;通过Z检验比较联合预测模型及各独立变量的AUC,从而判断联合预测模型较单一指标的工作性能。结果SAP组入院第1天的NLR、淀粉酶、脂肪酶、甘油三酯、CAR、PCT、CTSI评分、RANSON评分、APACHE-Ⅱ评分均高于MAP+MASP组,差异有显著性(P<0.01)。Logistic回归分析显示,CAR(a)、NLR(b)、甘油三酯(c)是SAP发生的独立危险因素,据此构建的logistic模型表达式为:logit(P)=-4.873+0.279a+0.056b+0.075c。对logistic回归方程进行等式变换,联合预测模型(Pre)表达式为L=0.279a+0.056b+0.075c。建立预测SAP的ROC曲线,Pre最佳临界值为3.24。Z检验显示,联合预测模型Pre预测SAP的AUC显著大于各单一指标,差异有显著性。结论使用logistic回归分析构建的联合预测模型对重症脂源性胰腺炎的预测效能优于仅使用CAR、NLR、甘油三酯等单一指标。展开更多
An innovative monochromator shielding is designed and implemented for the cold neutron spectrometers XINGZHI and BOYA operated by Renmin University of China at China Advanced Research Reactor.Via Monte Carlo simulatio...An innovative monochromator shielding is designed and implemented for the cold neutron spectrometers XINGZHI and BOYA operated by Renmin University of China at China Advanced Research Reactor.Via Monte Carlo simulations and careful mechanical designs,a shielding configuration has been successfully developed to satisfy safety requirements of below 3μSv/h dose rate at its exterior,meanwhile fulfilling space,floor load and nonmagnetic requirements.Composite materials are utilized to form the sandwich-type shielding walls:the inner layer of boron carbide rubber,the middle layer of steel-encased lead and the outer layer of borated polyethylene.Special-shaped liftable shielding blocks are incorporated to facilitate a continuous adjustment of the neutron energy while preventing radiation leakage.Our work has demonstrated that by utilizing composite shielding materials,along with the sandwich structure and liftable shielding blocks,a compact and lightweight shielding solution can be achieved.This enables the realization of advanced neutron scattering instruments that provide expanded space of measurement,larger energy and momentum coverage,and higher flux on the sample.This shielding represents the first of its kind in neutron scattering instruments in China.Following its successful operation,it has been subsequently employed by other neutron instruments across the country.展开更多
文摘目的使用logistic回归分析构建多个临床指标联合预测重症脂源性胰腺炎的预测模型,并评价其预测效果。方法纳入首都医科大学宣武医院2012年11月至2019年12月收住院的207例高甘油三酯血症性胰腺炎(hypertriglyceridemic pancreatitis,HTGP)患者作为研究对象,根据胰腺炎的严重程度将患者分为重症急性胰腺炎(severe acute pancreatitis,SAP)组(44例)和中重度胰腺炎(moderately severe acute pancreatitis,MSAP)+轻度胰腺炎(mild acute pancreatitis,MAP)组(163例)。比较两组患者一般临床资料,入院第1天中性粒细胞/淋巴细胞比值(neutrophil to lymphocyte ratio,NLR)、淀粉酶、脂肪酶、甘油三酯、C反应蛋白/白蛋白比值(C-reactive protein to albumin ratio,CAR)、降钙素原(procalcitonin,PCT)水平,CT严重指数(CT severity index,CTSI)评分、急性生理与慢性健康Ⅱ(acute physiology and chronic health evaluation-Ⅱ,APACHE-Ⅱ)评分、RANSON评分,住院天数、预后情况。应用logistic回归分析纳入差异有显著性的变量作为协变量,将是否发生SAP作为二分类结局变量,得出SAP的相关独立变量并拟合多个独立变量构建新的联合预测模型。构建受试者工作特征曲线(receiver operating characteristic curve,ROC)分析并评估联合预测模型及各独立变量对SAP的预测性能,得到曲线下面积(area under the curve,AUC)、最佳临界值、敏感度、特异度等参数;通过Z检验比较联合预测模型及各独立变量的AUC,从而判断联合预测模型较单一指标的工作性能。结果SAP组入院第1天的NLR、淀粉酶、脂肪酶、甘油三酯、CAR、PCT、CTSI评分、RANSON评分、APACHE-Ⅱ评分均高于MAP+MASP组,差异有显著性(P<0.01)。Logistic回归分析显示,CAR(a)、NLR(b)、甘油三酯(c)是SAP发生的独立危险因素,据此构建的logistic模型表达式为:logit(P)=-4.873+0.279a+0.056b+0.075c。对logistic回归方程进行等式变换,联合预测模型(Pre)表达式为L=0.279a+0.056b+0.075c。建立预测SAP的ROC曲线,Pre最佳临界值为3.24。Z检验显示,联合预测模型Pre预测SAP的AUC显著大于各单一指标,差异有显著性。结论使用logistic回归分析构建的联合预测模型对重症脂源性胰腺炎的预测效能优于仅使用CAR、NLR、甘油三酯等单一指标。
基金Project supported by the National Natural Science Foundation of China(Grant Nos.12004426,U2030106,and 12304185)the National Key Scientific Instrument and Equipment Development Project of NSFC(Grant No.11227906)the National Key R&D Program of China(Grant No.2023YFA1406500)。
文摘An innovative monochromator shielding is designed and implemented for the cold neutron spectrometers XINGZHI and BOYA operated by Renmin University of China at China Advanced Research Reactor.Via Monte Carlo simulations and careful mechanical designs,a shielding configuration has been successfully developed to satisfy safety requirements of below 3μSv/h dose rate at its exterior,meanwhile fulfilling space,floor load and nonmagnetic requirements.Composite materials are utilized to form the sandwich-type shielding walls:the inner layer of boron carbide rubber,the middle layer of steel-encased lead and the outer layer of borated polyethylene.Special-shaped liftable shielding blocks are incorporated to facilitate a continuous adjustment of the neutron energy while preventing radiation leakage.Our work has demonstrated that by utilizing composite shielding materials,along with the sandwich structure and liftable shielding blocks,a compact and lightweight shielding solution can be achieved.This enables the realization of advanced neutron scattering instruments that provide expanded space of measurement,larger energy and momentum coverage,and higher flux on the sample.This shielding represents the first of its kind in neutron scattering instruments in China.Following its successful operation,it has been subsequently employed by other neutron instruments across the country.
