COVID-19 Research Group, Center for Food Analysis (NAL), Technological Development Support Laboratory (LADETEC), Cidade Universitária, Rio de Janeiro RJ 21941-598, Brazil.
Laboratory of Advanced Analysis in Biochemistry and Molecular Biology (LAABBM), Department of Biochemistry, Federal University of Rio de Janeiro (UFRJ), Cidade Universitária, Rio de Janeiro RJ 21941-909, Brazil.
Int J Environ Res Public Health. 2020 Nov 30;17(23):8921. doi: 10.3390/ijerph17238921.
Infinite factors can influence the spread of COVID-19. Evaluating factors related to the spread of the disease is essential to point out measures that take effect. In this study, the influence of 14 variables was assessed together by Artificial Neural Networks (ANN) of the type Self-Organizing Maps (SOM), to verify the relationship between numbers of cases and deaths from COVID-19 in Brazilian states for 110 days. The SOM analysis showed that the variables that presented a more significant relationship with the numbers of cases and deaths by COVID-19 were influenza vaccine applied, Intensive Care Unit (ICU), ventilators, physicians, nurses, and the Human Development Index (HDI). In general, Brazilian states with the highest rates of influenza vaccine applied, ICU beds, ventilators, physicians, and nurses, per 100,000 inhabitants, had the lowest number of cases and deaths from COVID-19, while the states with the lowest rates were most affected by the disease. According to the SOM analysis, other variables such as Personal Protective Equipment (PPE), tests, drugs, and Federal funds, did not have as significant effect as expected.
无穷多的因素可能影响新冠病毒的传播。评估与疾病传播相关的因素对于指出有效的措施至关重要。在这项研究中，通过自组织映射（SOM）类型的人工神经网络（ANN）综合评估了 14 个变量，以验证巴西各州 110 天内新冠病毒病例和死亡人数之间的关系。SOM 分析表明，与新冠病毒病例和死亡人数关系最密切的变量是流感疫苗接种、重症监护病房（ICU）、呼吸机、医生、护士和人类发展指数（HDI）。一般来说，巴西各州每 10 万人中流感疫苗接种、ICU 床位、呼吸机、医生和护士的比例越高，新冠病毒病例和死亡人数就越低，而接种比例最低的州受到的影响最大。根据 SOM 分析，个人防护设备（PPE）、检测、药物和联邦资金等其他变量的影响并不像预期的那样显著。
