Objective To study the causal relationship between R&D investment and enterprise performance of domestic pharmaceutical enterprises.Methods Panel data model was adopted for empirical analysis.Results and Conclusio...Objective To study the causal relationship between R&D investment and enterprise performance of domestic pharmaceutical enterprises.Methods Panel data model was adopted for empirical analysis.Results and Conclusion Increasing the R&D investment intensity of pharmaceutical enterprises in the Yangtze River Delta and Zhejiang by 1%will increase their profit margins by 0.79%and 0.46%.On the contrary,if the profit margin increases by 1%,the R&D investment intensity will increase by 0.25%and 0.19%.If the profit margin of pharmaceutical enterprises in Beijing,Tianjin,Hebei,Chengdu,Chongqing and other regions increases by 1%,the R&D investment intensity will increase by 0.14%,0.07%and 0.1%,respectively,which are lower than those in the Yangtze River Delta and Zhejiang.The relationship between R&D investment and enterprise performance of pharmaceutical enterprises in the Yangtze River Delta and Zhejiang Province is Granger causality,showing a two-way positive effect.Profits and R&D investment of pharmaceutical enterprises in Beijing,Tianjin,Hebei,Chengdu,Chongqing and other regions are also Granger causality.But in the Pearl River Delta,profits and R&D investment have not passed the stability test,it is impossible to determine the causality between them.展开更多
Background:Adolescent depression and school refusal(SR)are prevalent and important global concerns that need to be understood and addressed.Cross-sectional associations have been reported but prospective relationships...Background:Adolescent depression and school refusal(SR)are prevalent and important global concerns that need to be understood and addressed.Cross-sectional associations have been reported but prospective relationships between them remain unclear.This longitudinal study investigated the bidirectional relationships between these two problems among Chinese adolescents.Methods:A longitudinal study was conducted in Taizhou,China,surveying students of three junior high schools,three senior high schools,and one vocational high school.A total of 3882 students completed the questionnaire at baseline(T1);3167 of them completed an identical follow-up questionnaire after 6 months(T2).Depression was assessed via the Patient Health Questionnaire(PHQ-9)and SR via the modified Chinese version of The School Refusal Assessment Scale-Revised(SRAS-R).Cross-lagged panel modeling(CLPM)analysis was conducted to test the reciprocal relationships,adjusting for socio-demographic factors.Multiple group analysis was conducted to test whether the CLPM differed by gender and grade.Results:Statistically significant bidirectional relationships were found.A higher level of SR assessed at T1 is prospectively associated with a higher level of depression at T2(β=0.07,p=0.006);a higher level of depression at T1 also is prospectively associated with a higher level of SR at T2(β=0.14,p<0.001).Such models differed significantly by neither gender nor grade.Conclusion:SR and depression should be seen as each other’s mutually reinforcing association.The bidirectional relationships potentially result in a vicious cycle.Early interventions may target both problems concurrently.Future studies may involve more time points and test some mediators.展开更多
随着大型语言模型(LLMs)在超大规模语料库上开展预训练,数据污染问题逐渐凸显,这对模型性能评估的准确性构成了直接威胁。提出了一种创新的动态数据评估方法EdEval(equal distribution dynamic evaluation),旨在降低数据污染对评估准确...随着大型语言模型(LLMs)在超大规模语料库上开展预训练,数据污染问题逐渐凸显,这对模型性能评估的准确性构成了直接威胁。提出了一种创新的动态数据评估方法EdEval(equal distribution dynamic evaluation),旨在降低数据污染对评估准确性的影响。EdEval通过提取核心知识点与主旨,确保生成的评估问题在本质上与静态数据一致,并结合联网检索对知识点进行深入阐述,生成具有高质量知识支撑的评估样本。此外,EdEval通过控制问题数量和复杂度,实现动态对齐与灵活调节,以匹配静态数据的难度水平并满足不同复杂度的需求。采用布鲁姆分类法,EdEval从记忆、理解、应用、分析、评价和创造六个维度对LLMs进行综合评估。实验结果表明,EdEval在多个数据集上有效减轻了数据污染的影响,显著提高了评估的公正性和准确性。展开更多
目的采用数据挖掘方法总结肺纤维化动物模型的特点及建立药效指标评价体系。方法通过中国知网(CNKI)、万方数据库(Wangfang)、维普中文科技期刊全文数据库(VIP)、PubMed、Web of Science、Embase数据库检索与肺纤维化动物药效研究相关...目的采用数据挖掘方法总结肺纤维化动物模型的特点及建立药效指标评价体系。方法通过中国知网(CNKI)、万方数据库(Wangfang)、维普中文科技期刊全文数据库(VIP)、PubMed、Web of Science、Embase数据库检索与肺纤维化动物药效研究相关的文献,归纳整理、分析肺纤维化动物模型的造模方法、干预药物等,并统计检测指标类型、方法等,构建肺纤维化动物药效指标体系。结果共纳入1174篇文献,动物造模常见的诱导因素有肿瘤药物、环境/职业暴露颗粒、物理因素等,以C57 BL/6小鼠、SD大鼠为主要研究对象,其中博来霉素以无创性气管滴注诱导肺纤维化动物模型最常见。肺纤维化动物药效研究中常见的干预药物有化学药、抑制剂/激动剂、天然药物、中药复方等。肺纤维化动物药效研究中检测指标包括一般情况、肺功能、肺组织病理、细胞外基质、上皮间质转化、细胞因子、氧化应激等七类,其中一般情况以体质量、肺系数、生存分析检测为主;肺功能指标主要包括用力肺活量、动态肺顺应性、潮气量等;常见的肺组织病理染色方法有HE染色、Masson染色及天狼猩红染色等;细胞外基质检测指标以Ⅰ型胶原蛋白、羟脯氨酸、纤维连接蛋白等为主;上皮间质转化指标有α-平滑肌肌动蛋白、E-钙黏蛋白、波形蛋白等;细胞因子检测指标主要有转化生长因子β1、肿瘤坏死因子α、白细胞介素6等;氧化应激检测指标主要包括丙二醛、超氧化物歧化酶、谷胱甘肽等。根据检测指标频次≥200次作为肺纤维化动物药效研究检测的强推荐指标,将一般情况(体质量、肺系数)、肺病理(HE染色、Masson染色等)、细胞外基质(羟脯氨酸、Ⅰ型胶原蛋白、Ⅲ型胶原蛋白、纤维连接蛋白)、上皮间质化(α-平滑肌肌动蛋白)、细胞因子(肿瘤坏死因子α、白细胞介素6/1β、转化生长因子β1)作为肺纤维化动物药效研究检测的强推荐指标。结论本研究为肺纤维化动物模型构建及药效指标评价体系的建立提供了更多参考。展开更多
基金Shenyang Pharmaceutical University Young and Middle aged Teacher Career Development Support PlanPublic Welfare Research Fund for Scientific Undertakings of Liaoning Province in 2022(Soft Science Research Plan)(No.2022JH4/10100040).
文摘Objective To study the causal relationship between R&D investment and enterprise performance of domestic pharmaceutical enterprises.Methods Panel data model was adopted for empirical analysis.Results and Conclusion Increasing the R&D investment intensity of pharmaceutical enterprises in the Yangtze River Delta and Zhejiang by 1%will increase their profit margins by 0.79%and 0.46%.On the contrary,if the profit margin increases by 1%,the R&D investment intensity will increase by 0.25%and 0.19%.If the profit margin of pharmaceutical enterprises in Beijing,Tianjin,Hebei,Chengdu,Chongqing and other regions increases by 1%,the R&D investment intensity will increase by 0.14%,0.07%and 0.1%,respectively,which are lower than those in the Yangtze River Delta and Zhejiang.The relationship between R&D investment and enterprise performance of pharmaceutical enterprises in the Yangtze River Delta and Zhejiang Province is Granger causality,showing a two-way positive effect.Profits and R&D investment of pharmaceutical enterprises in Beijing,Tianjin,Hebei,Chengdu,Chongqing and other regions are also Granger causality.But in the Pearl River Delta,profits and R&D investment have not passed the stability test,it is impossible to determine the causality between them.
基金funded by Science and Technology Program of Wenzhou(Y20220843).
文摘Background:Adolescent depression and school refusal(SR)are prevalent and important global concerns that need to be understood and addressed.Cross-sectional associations have been reported but prospective relationships between them remain unclear.This longitudinal study investigated the bidirectional relationships between these two problems among Chinese adolescents.Methods:A longitudinal study was conducted in Taizhou,China,surveying students of three junior high schools,three senior high schools,and one vocational high school.A total of 3882 students completed the questionnaire at baseline(T1);3167 of them completed an identical follow-up questionnaire after 6 months(T2).Depression was assessed via the Patient Health Questionnaire(PHQ-9)and SR via the modified Chinese version of The School Refusal Assessment Scale-Revised(SRAS-R).Cross-lagged panel modeling(CLPM)analysis was conducted to test the reciprocal relationships,adjusting for socio-demographic factors.Multiple group analysis was conducted to test whether the CLPM differed by gender and grade.Results:Statistically significant bidirectional relationships were found.A higher level of SR assessed at T1 is prospectively associated with a higher level of depression at T2(β=0.07,p=0.006);a higher level of depression at T1 also is prospectively associated with a higher level of SR at T2(β=0.14,p<0.001).Such models differed significantly by neither gender nor grade.Conclusion:SR and depression should be seen as each other’s mutually reinforcing association.The bidirectional relationships potentially result in a vicious cycle.Early interventions may target both problems concurrently.Future studies may involve more time points and test some mediators.
文摘随着大型语言模型(LLMs)在超大规模语料库上开展预训练,数据污染问题逐渐凸显,这对模型性能评估的准确性构成了直接威胁。提出了一种创新的动态数据评估方法EdEval(equal distribution dynamic evaluation),旨在降低数据污染对评估准确性的影响。EdEval通过提取核心知识点与主旨,确保生成的评估问题在本质上与静态数据一致,并结合联网检索对知识点进行深入阐述,生成具有高质量知识支撑的评估样本。此外,EdEval通过控制问题数量和复杂度,实现动态对齐与灵活调节,以匹配静态数据的难度水平并满足不同复杂度的需求。采用布鲁姆分类法,EdEval从记忆、理解、应用、分析、评价和创造六个维度对LLMs进行综合评估。实验结果表明,EdEval在多个数据集上有效减轻了数据污染的影响,显著提高了评估的公正性和准确性。
文摘目的采用数据挖掘方法总结肺纤维化动物模型的特点及建立药效指标评价体系。方法通过中国知网(CNKI)、万方数据库(Wangfang)、维普中文科技期刊全文数据库(VIP)、PubMed、Web of Science、Embase数据库检索与肺纤维化动物药效研究相关的文献,归纳整理、分析肺纤维化动物模型的造模方法、干预药物等,并统计检测指标类型、方法等,构建肺纤维化动物药效指标体系。结果共纳入1174篇文献,动物造模常见的诱导因素有肿瘤药物、环境/职业暴露颗粒、物理因素等,以C57 BL/6小鼠、SD大鼠为主要研究对象,其中博来霉素以无创性气管滴注诱导肺纤维化动物模型最常见。肺纤维化动物药效研究中常见的干预药物有化学药、抑制剂/激动剂、天然药物、中药复方等。肺纤维化动物药效研究中检测指标包括一般情况、肺功能、肺组织病理、细胞外基质、上皮间质转化、细胞因子、氧化应激等七类,其中一般情况以体质量、肺系数、生存分析检测为主;肺功能指标主要包括用力肺活量、动态肺顺应性、潮气量等;常见的肺组织病理染色方法有HE染色、Masson染色及天狼猩红染色等;细胞外基质检测指标以Ⅰ型胶原蛋白、羟脯氨酸、纤维连接蛋白等为主;上皮间质转化指标有α-平滑肌肌动蛋白、E-钙黏蛋白、波形蛋白等;细胞因子检测指标主要有转化生长因子β1、肿瘤坏死因子α、白细胞介素6等;氧化应激检测指标主要包括丙二醛、超氧化物歧化酶、谷胱甘肽等。根据检测指标频次≥200次作为肺纤维化动物药效研究检测的强推荐指标,将一般情况(体质量、肺系数)、肺病理(HE染色、Masson染色等)、细胞外基质(羟脯氨酸、Ⅰ型胶原蛋白、Ⅲ型胶原蛋白、纤维连接蛋白)、上皮间质化(α-平滑肌肌动蛋白)、细胞因子(肿瘤坏死因子α、白细胞介素6/1β、转化生长因子β1)作为肺纤维化动物药效研究检测的强推荐指标。结论本研究为肺纤维化动物模型构建及药效指标评价体系的建立提供了更多参考。
