利用专利文献数据识别技术领域的技术主题演化发展路径并分析其发展趋势,对于科技界、企业界进行专利技术创新具有重要的意义。首先,使用Open IE 5.1进行SAO(subject-action-object)三元组抽取,基于LDA(latent Dirichlet allocation)模...利用专利文献数据识别技术领域的技术主题演化发展路径并分析其发展趋势,对于科技界、企业界进行专利技术创新具有重要的意义。首先,使用Open IE 5.1进行SAO(subject-action-object)三元组抽取,基于LDA(latent Dirichlet allocation)模型进行主题识别,根据TRIZ技术创新思想,基于action语义词典将技术主题划分到四个维度;然后,通过计算SAO三元组之间的相似度来测度技术主题之间的语义关联构建技术主题创新演化路径,并利用可视化技术构建技术主题创新演化路径可视化图谱,利用该图谱辅助分析技术主题演化脉络及其发展趋势。最后,通过石墨烯超级电容器(集流体)领域的实证,对该领域的技术问题(problem to problem,P-P)主题、技术功能(solution to solution,S-S)主题、解决方案(problem to solution,P-S)主题和技术效果(solution to problem,S-P)主题的创新演化路径进行解读分析,验证了本研究提出方法的可行性和有效性。展开更多
Rationale: In the literature, some risk factors for severity and mortality from COVID-19 have been indicated. However, these factors can change, depending on the characteristics of the population and health services. ...Rationale: In the literature, some risk factors for severity and mortality from COVID-19 have been indicated. However, these factors can change, depending on the characteristics of the population and health services. In this sense, longitudinal studies can be useful for understanding local realities and subsidizing health actions based on these realities. Objective: To analyze the risk factors for severity and death in hospitalized patients with COVID-19. Methods: A retrospective cohort of patients with COVID-19 hospitalized from August 1 to October 16, 2021 (3<sup>rd</sup> wave of the pandemic), notified by the Department of Epidemiological Surveillance of Sao Tome and Principe. We employed measures of strength of associations for the analysis of exposure risk factors. Results: We analyzed 110 hospitalized patients (31.8% severe-critical and 68.2% non-severe). The risk factors for severe forms of COVID-19 were: being aged ≥60 years (RR = 3.3), being male (RR = 2), having comorbidities (RR = 2) and the risk increases to 10-fold for multicomorbidities, with emphasis on obesity, neoplasia, skin-muscle-surgical infection, dementia and to some degree CVD. 62.9% of patients with severe forms of the disease were not vaccinated. Risk factors for death among hospitalized and severe/critical cases, respectively, were having comorbidities (RR = 8 and 2.4) multicomorbidities (RR = 10 and 2.8 for those with 2 comorbidities and RR = 33.3 and 4 for those with 3 or 4 comorbidities), especially diabetes, dementia, neoplasia, cutaneous-muscular infection, and obesity. Although CVD was not associated with risk factors for death, these were the most frequently found among the severely hospitalized and deaths. In addition, important risk factors associated with death were not using corticoids (RR = 3.3, 230-fold risk) and not using anticoagulants-heparin (RR = 1.3, 30% risk) more compared to the severe cases that did use them. Most of the patients who died (63.2%) were not vaccinated. Moreover, having only 1 dose of the vaccine was a risk factor 1.9 times more for death among all hospitalized patients, but in the severe cases, there was no association between the variable vaccination and death. Among those hospitalized with 2 doses, it was a 0.5-fold protective factor among those hospitalized. The Delta variant of Sarscov-2 was the one found among severe cases and deaths investigated by genetic sequencing, with more exuberant clinical features compared to the other 2 previous vaccinations. Conclusion: Being elderly, male and presenting comorbidities, mainly multicomorbidities were the main characteristics associated with severity of COVID-19. On the other hand, comorbidities, and even worse, multicomorbidities, hospitalization for respiratory failure, lowered level of consciousness, no use of corticoid and no use of anticoagulation in critically ill patients, and not having at least 2 doses of vaccine for covid-19, were characteristics associated with death by COVID-19. These results will help inform healthcare providers so that the best interventions can be implemented to improve outcomes for patients with COVID-19. Public health interventions must be carefully tailored and implemented in these susceptible groups to reduce the risk of mortality in patients with COVID-19 and then the risk of major complications. Intensive and regular follow-up is needed to detect early occurrences of clinical conditions.展开更多
文摘利用专利文献数据识别技术领域的技术主题演化发展路径并分析其发展趋势,对于科技界、企业界进行专利技术创新具有重要的意义。首先,使用Open IE 5.1进行SAO(subject-action-object)三元组抽取,基于LDA(latent Dirichlet allocation)模型进行主题识别,根据TRIZ技术创新思想,基于action语义词典将技术主题划分到四个维度;然后,通过计算SAO三元组之间的相似度来测度技术主题之间的语义关联构建技术主题创新演化路径,并利用可视化技术构建技术主题创新演化路径可视化图谱,利用该图谱辅助分析技术主题演化脉络及其发展趋势。最后,通过石墨烯超级电容器(集流体)领域的实证,对该领域的技术问题(problem to problem,P-P)主题、技术功能(solution to solution,S-S)主题、解决方案(problem to solution,P-S)主题和技术效果(solution to problem,S-P)主题的创新演化路径进行解读分析,验证了本研究提出方法的可行性和有效性。
文摘Rationale: In the literature, some risk factors for severity and mortality from COVID-19 have been indicated. However, these factors can change, depending on the characteristics of the population and health services. In this sense, longitudinal studies can be useful for understanding local realities and subsidizing health actions based on these realities. Objective: To analyze the risk factors for severity and death in hospitalized patients with COVID-19. Methods: A retrospective cohort of patients with COVID-19 hospitalized from August 1 to October 16, 2021 (3<sup>rd</sup> wave of the pandemic), notified by the Department of Epidemiological Surveillance of Sao Tome and Principe. We employed measures of strength of associations for the analysis of exposure risk factors. Results: We analyzed 110 hospitalized patients (31.8% severe-critical and 68.2% non-severe). The risk factors for severe forms of COVID-19 were: being aged ≥60 years (RR = 3.3), being male (RR = 2), having comorbidities (RR = 2) and the risk increases to 10-fold for multicomorbidities, with emphasis on obesity, neoplasia, skin-muscle-surgical infection, dementia and to some degree CVD. 62.9% of patients with severe forms of the disease were not vaccinated. Risk factors for death among hospitalized and severe/critical cases, respectively, were having comorbidities (RR = 8 and 2.4) multicomorbidities (RR = 10 and 2.8 for those with 2 comorbidities and RR = 33.3 and 4 for those with 3 or 4 comorbidities), especially diabetes, dementia, neoplasia, cutaneous-muscular infection, and obesity. Although CVD was not associated with risk factors for death, these were the most frequently found among the severely hospitalized and deaths. In addition, important risk factors associated with death were not using corticoids (RR = 3.3, 230-fold risk) and not using anticoagulants-heparin (RR = 1.3, 30% risk) more compared to the severe cases that did use them. Most of the patients who died (63.2%) were not vaccinated. Moreover, having only 1 dose of the vaccine was a risk factor 1.9 times more for death among all hospitalized patients, but in the severe cases, there was no association between the variable vaccination and death. Among those hospitalized with 2 doses, it was a 0.5-fold protective factor among those hospitalized. The Delta variant of Sarscov-2 was the one found among severe cases and deaths investigated by genetic sequencing, with more exuberant clinical features compared to the other 2 previous vaccinations. Conclusion: Being elderly, male and presenting comorbidities, mainly multicomorbidities were the main characteristics associated with severity of COVID-19. On the other hand, comorbidities, and even worse, multicomorbidities, hospitalization for respiratory failure, lowered level of consciousness, no use of corticoid and no use of anticoagulation in critically ill patients, and not having at least 2 doses of vaccine for covid-19, were characteristics associated with death by COVID-19. These results will help inform healthcare providers so that the best interventions can be implemented to improve outcomes for patients with COVID-19. Public health interventions must be carefully tailored and implemented in these susceptible groups to reduce the risk of mortality in patients with COVID-19 and then the risk of major complications. Intensive and regular follow-up is needed to detect early occurrences of clinical conditions.
