With its novel chemical structure, artemisinin is an antimalarial component isolated from the traditional Chinese medicine qinghao(Artemisia annua L.). Nowadays, artemisinin and its derivatives are used compatibly wit...With its novel chemical structure, artemisinin is an antimalarial component isolated from the traditional Chinese medicine qinghao(Artemisia annua L.). Nowadays, artemisinin and its derivatives are used compatibly with new synthesized chemical antimalarial compounds to create artemisinin-based combination therapies(ACTs). These have become the first choice in treating malaria p.f. all over the world, providing an effective solution for the global challenge of curing drug-resistant malaria. Among the five ACTs recommended by the WHO, two were initiated in China and are used as the first-line treatment of falciparum malaria in all malaria endemic areas. As the use of artemisinin-based compound drugs have made such significant contributions to rolling back malaria, regarded as one of the great achievements globally in public health of the early twenty-first century, Tu Youyou, one of the most important researchers in the discovery of artemisinin, was made the first Nobel Prize laureate in Physiology or Medicine from the Chinese mainland. Artemisinin was discovered in a special social and cultural context through a combination of the exploration of traditional Chinese medical literature with the modern research approach of pharmaceutical sciences. This(Project 523) is a typical case of goal-oriented research leading to scientific advance, and the result of scientific research driven by the national needs.展开更多
目的建立预测重症慢性阻塞性肺疾病(简称慢阻肺)患者死亡风险的机器学习模型,探讨与慢阻肺患者死亡风险相关的因素,并加以解释,解决机器学习模型的“黑箱”问题。方法选取美国多中心急诊重症监护病(emergency intensive care unit,eICU...目的建立预测重症慢性阻塞性肺疾病(简称慢阻肺)患者死亡风险的机器学习模型,探讨与慢阻肺患者死亡风险相关的因素,并加以解释,解决机器学习模型的“黑箱”问题。方法选取美国多中心急诊重症监护病(emergency intensive care unit,eICU)数据库中的8088例重症慢阻肺患者为研究对象,提取每次入住重症监护病房的前24 h内的数据并随机分组,70%用于模型训练,30%用于模型验证。采用LASSO回归进行预测变量选择,避免过拟合。采用5种机器学习模型对患者的住院病死率进行预测。通过曲线下面积(area under curve,AUC)比较5种模型和APACHEⅣa评分的预测性能,并采用SHAP(SHapley Additive exPlanations)方法解释随机森林(random forest,RF)模型的预测结果。结果RF模型在5种机器学习模型和APACHEⅣa评分系统中表现出最佳的性能,AUC达到0.830(95%置信区间0.806~0.855)。通过SHAP方法检测最重要的10种预测变量,其中无创收缩压的最小值被认为是最重要的预测变量。结论通过机器学习识别危险因素,并使用SHAP方法解释预测结果,可早期预测患者的死亡风险,有助于临床医生制定准确的治疗计划,合理分配医疗资源。展开更多
It is surprising that,while arsenic trioxide(ATO) is now considered as "the single most active agent in patients with acute promyelocytic leukemia(APL)",the most important discoverer remains obscure and his ...It is surprising that,while arsenic trioxide(ATO) is now considered as "the single most active agent in patients with acute promyelocytic leukemia(APL)",the most important discoverer remains obscure and his original papers have not been cited by a single English paper.The discovery was made during the Cultural Revolution when most Chinese scientists and doctors struggled to survive.Beginning with recipes from a countryside practitioner that were vague in applicable diseases,Zhang TingDong and colleagues proposed in the 1970s that a single chemical in the recipe is most effective and that its target is APL.More than 20 years of work by Zhang and colleagues eliminated the confusions about whether and how ATO can be used effectively.Other researchers,first in China and then in the West,followed his lead.Retrospective analysis of data from his own group proved that APL was indeed the most sensitive target.Removal of a trace amount of mercury chloride from the recipe by another group in his hospital proved that only ATO was required.Publication of Western replication in 1998 made the therapy widely accepted,though neither Western,nor Chinese authors of English papers on ATO cited Zhang's papers in the 1970s.This article focuses on the early papers of Zhang,but also suggests it worth further work to validate Chinese reports of ATO treatment of other cancers,and infers that some findings published in Chinese journals are of considerable value to patients and that doctors from other countries can benefit from the clinical experience of Chinese doctors with the largest population of patients.展开更多
文摘With its novel chemical structure, artemisinin is an antimalarial component isolated from the traditional Chinese medicine qinghao(Artemisia annua L.). Nowadays, artemisinin and its derivatives are used compatibly with new synthesized chemical antimalarial compounds to create artemisinin-based combination therapies(ACTs). These have become the first choice in treating malaria p.f. all over the world, providing an effective solution for the global challenge of curing drug-resistant malaria. Among the five ACTs recommended by the WHO, two were initiated in China and are used as the first-line treatment of falciparum malaria in all malaria endemic areas. As the use of artemisinin-based compound drugs have made such significant contributions to rolling back malaria, regarded as one of the great achievements globally in public health of the early twenty-first century, Tu Youyou, one of the most important researchers in the discovery of artemisinin, was made the first Nobel Prize laureate in Physiology or Medicine from the Chinese mainland. Artemisinin was discovered in a special social and cultural context through a combination of the exploration of traditional Chinese medical literature with the modern research approach of pharmaceutical sciences. This(Project 523) is a typical case of goal-oriented research leading to scientific advance, and the result of scientific research driven by the national needs.
文摘目的建立预测重症慢性阻塞性肺疾病(简称慢阻肺)患者死亡风险的机器学习模型,探讨与慢阻肺患者死亡风险相关的因素,并加以解释,解决机器学习模型的“黑箱”问题。方法选取美国多中心急诊重症监护病(emergency intensive care unit,eICU)数据库中的8088例重症慢阻肺患者为研究对象,提取每次入住重症监护病房的前24 h内的数据并随机分组,70%用于模型训练,30%用于模型验证。采用LASSO回归进行预测变量选择,避免过拟合。采用5种机器学习模型对患者的住院病死率进行预测。通过曲线下面积(area under curve,AUC)比较5种模型和APACHEⅣa评分的预测性能,并采用SHAP(SHapley Additive exPlanations)方法解释随机森林(random forest,RF)模型的预测结果。结果RF模型在5种机器学习模型和APACHEⅣa评分系统中表现出最佳的性能,AUC达到0.830(95%置信区间0.806~0.855)。通过SHAP方法检测最重要的10种预测变量,其中无创收缩压的最小值被认为是最重要的预测变量。结论通过机器学习识别危险因素,并使用SHAP方法解释预测结果,可早期预测患者的死亡风险,有助于临床医生制定准确的治疗计划,合理分配医疗资源。
文摘It is surprising that,while arsenic trioxide(ATO) is now considered as "the single most active agent in patients with acute promyelocytic leukemia(APL)",the most important discoverer remains obscure and his original papers have not been cited by a single English paper.The discovery was made during the Cultural Revolution when most Chinese scientists and doctors struggled to survive.Beginning with recipes from a countryside practitioner that were vague in applicable diseases,Zhang TingDong and colleagues proposed in the 1970s that a single chemical in the recipe is most effective and that its target is APL.More than 20 years of work by Zhang and colleagues eliminated the confusions about whether and how ATO can be used effectively.Other researchers,first in China and then in the West,followed his lead.Retrospective analysis of data from his own group proved that APL was indeed the most sensitive target.Removal of a trace amount of mercury chloride from the recipe by another group in his hospital proved that only ATO was required.Publication of Western replication in 1998 made the therapy widely accepted,though neither Western,nor Chinese authors of English papers on ATO cited Zhang's papers in the 1970s.This article focuses on the early papers of Zhang,but also suggests it worth further work to validate Chinese reports of ATO treatment of other cancers,and infers that some findings published in Chinese journals are of considerable value to patients and that doctors from other countries can benefit from the clinical experience of Chinese doctors with the largest population of patients.
