Examinando por Autor "Figueroa Arias, José Julián"
Mostrando 1 - 1 de 1
Resultados por página
Opciones de ordenación
- ÍtemModelo de clasificación machine learning para pronosticar secuelas físicas en pacientes postcovid.(Fundación Universitaria Los Libertadores. Sede Bogotá., ) Figueroa Arias, José Julián; González Veloza, José John FredyCovid 19 is an infectious virus that produces a severe acute respiratory syndrome, and among the most frequent symptoms are respiratory symptoms, fever and also gastrointestinal symptoms. One of the characteristics of this virus is that after the recovery period, in some cases there are physical sequelae such as coughing, loss of smell, muscle pain, headache, etc., sequelae that have caused fatalities throughout the world. Therefore, in this study, a machine learning classification model was carried out to predict physical sequelae in postcovid patients, as a sample, information was obtained from 1436 observations of patients from the Nariño Departmental University Hospital who were positive for covid 19, after recovery. Information was obtained from these patients on the variable of interest for this study, which was the presentation of post-COVID physical sequelae. It was found that the model with the best performance metrics was the classification tree with auc of 0.73. It is concluded that the classification model is useful to identify possible cases of individuals with post-covid sequelae and thus manage hospital actions to reduce complications and fatalities after the recovery period caused by the Covid-19 virus.