Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 3 de 3
Research and Innovation Forum, Rii Forum 2021 ; : 13-24, 2021.
Article in English | Scopus | ID: covidwho-1469594


This article presents a model that aims to identify with Machine Learning (ML) technics the main symptoms and risk factors affected in patients with Coronavirus Covid-19, registered in the database of epidemiological surveillance of state and municipal information in Brazil. The concept behind ML is the ability to learn and reason. Its application can optimize and make the treatment and care process more accurate for the cases diagnosed with the Covid-19, also known as SARS-CoV-2, adjusting the medical data recorded concerning the disease and reducing the number of symptoms and risk factors, denoting an efficient form of attribute engineering, providing those involved with the clinical observation of a minor sign. We propose an approach structured in the composition of Machine Learning algorithms, aiming to discover knowledge and concepts followed by the refinement of the results. In this article, the proposed model is presented, and a shorter trail of symptomatic observations from Covid-19 are provided. © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Epidemiol Infect ; 149: e60, 2021 02 25.
Article in English | MEDLINE | ID: covidwho-1101607


The objective of this study was to analyse the dynamics of spatial dispersion of the coronavirus disease 2019 (COVID-19) in Brazil by correlating them to socioeconomic indicators. This is an ecological study of COVID-19 cases and deaths between 26 February and 31 July 2020. All Brazilian counties were used as units of analysis. The incidence, mortality, Bayesian incidence and mortality rates, global and local Moran indices were calculated. A geographic weighted regression analysis was conducted to assess the relationship between incidence and mortality due to COVID-19 and socioeconomic indicators (independent variables). There were confirmed 2 662 485 cases of COVID-19 reported in Brazil from February to July 2020 with higher rates of incidence in the north and northeast. The Moran global index of incidence rate (0.50, P = 0.01) and mortality (0.45 with P = 0.01) indicate a positive spatial autocorrelation with high standards in the north, northeast and in the largest urban centres between cities in the southeast region. In the same period, there were 92 475 deaths from COVID-19, with higher mortality rates in the northern states of Brazil, mainly Amazonas, Pará and Amapá. The results show that there is a geospatial correlation of COVID-19 in large urban centres and regions with the lowest human development index in the country. In the geographic weighted regression, it was possible to identify that the percentage of people living in residences with density higher than 2 per dormitory, the municipality human development index (MHDI) and the social vulnerability index were the indicators that most contributed to explaining incidence, social development index and the municipality human development index contributed the most to the mortality model. We hope that the findings will contribute to reorienting public health responses to combat COVID-19 in Brazil, the new epicentre of the disease in South America, as well as in other countries that have similar epidemiological and health characteristics to those in Brazil.

COVID-19/diagnosis , Pandemics/statistics & numerical data , Bayes Theorem , Brazil/epidemiology , COVID-19/epidemiology , Cities/epidemiology , Humans , Incidence , Linear Models , Pandemics/prevention & control , Risk Factors , Socioeconomic Factors , Spatial Analysis
Epidemiol Infect ; 148: e123, 2020 06 25.
Article in English | MEDLINE | ID: covidwho-615328


This study aims to identify the risk factors associated with mortality and survival of COVID-19 cases in a state of the Brazilian Northeast. It is a historical cohort with a secondary database of 2070 people that presented flu-like symptoms, sought health assistance in the state and tested positive to COVID-19 until 14 April 2020, only moderate and severe cases were hospitalised. The main outcome was death as a binary variable (yes/no). It also investigated the main factors related to mortality and survival of the disease. Time since the beginning of symptoms until death/end of the survey (14 April 2020) was the time variable of this study. Mortality was analysed by robust Poisson regression, and survival by Kaplan-Meier and Cox regression. From the 2070 people that tested positive to COVID-19, 131 (6.3%) died and 1939 (93.7%) survived, the overall survival probability was 87.7% from the 24th day of infection. Mortality was enhanced by the variables: elderly (HR 3.6; 95% CI 2.3-5.8; P < 0.001), neurological diseases (HR 3.9; 95% CI 1.9-7.8; P < 0.001), pneumopathies (HR 2.6; 95% CI 1.4-4.7; P < 0.001) and cardiovascular diseases (HR 8.9; 95% CI 5.4-14.5; P < 0.001). In conclusion, mortality by COVID-19 in Ceará is similar to countries with a large number of cases of the disease, although deaths occur later. Elderly people and comorbidities presented a greater risk of death.

Coronavirus Infections/mortality , Pneumonia, Viral/mortality , Adult , Age Factors , Aged , Brazil/epidemiology , COVID-19 , Cardiovascular Diseases/complications , Cohort Studies , Comorbidity , Coronavirus Infections/complications , Diabetes Complications/complications , Female , Hospitalization , Humans , Intensive Care Units , Kaplan-Meier Estimate , Kidney Diseases/complications , Lung Diseases/complications , Male , Middle Aged , Nervous System Diseases/complications , Pandemics , Pneumonia, Viral/complications , Poisson Distribution , Proportional Hazards Models , Risk Factors , Sex Factors , Time Factors