Predicting the disease outcome in COVID-19 positive patients through Machine Learning: a retrospective cohort study with Brazilian data

Abstract

The first officially registered case of COVID-19 in Brazil was on February 26, 2020. Since then, the situation has worsened with more than 672,000 confirmed cases and at least 36,000 reported deaths at the time of this writing. Accurate diagnosis of patients with COVID-19 is extremely important to offer adequate treatment, and avoid overloading the healthcare system. Characteristics of patients such as age, comorbidities and varied clinical symptoms can help in classifying the level of infection severity, predict the disease outcome and the need for hospitalization. Here, we present a study to predict a poor prognosis in positive COVID-19 patients and possible outcomes using machine learning. The study dataset comprises information of 13,690 patients concerning closed cases due to cure or death. Our experimental results show the disease outcome can be predicted with a ROC AUC of 0.92, Sensitivity of 0.88 and Specificity of 0.82 for the best prediction model. This is a preliminary retrospective study which can be improved with the inclusion of further data. Conclusion: Machine learning techniques fed with demographic and clinical data along with comorbidities of the patients can assist in the prognostic prediction and physician decision-making, allowing a faster response and contributing to the non-overload of healthcare systems.

On the analysis of mortality risk factors for hospitalized COVID-19 patients: A data-driven study using the major Brazilian database

“…These data are in agreement with a study carried out in the Italian population [37]. In our previous cohort study on outpatients from a Brazilian state (ES) [38], we found a prevalence of 18.55% for cardiac disease and 7.89% for diabetes. Lethality rates in this cohort were 50.05% and 31.82% for cardiac and diabetic patients, respectively.…”

“…Another aspect is the notification data of only hospitalized COVID-19 patients, so that we were unable to trace the profile of outpatients. However, our previous study using a dataset from Espírito Santo state [ 38 ] can be used for comparative purposes, as it involved COVID-19 outpatients.…”

On the Analysis of Mortality Risk Factors for Hospitalized COVID-19 Patients: a Data-driven Study Using the Major Brazilian Database

“…Highly prevalent comorbidities among hospitalized COVID-19 patients were cardiac disease and diabetes, which is in line with observations made in other populations [8], [12]. In our previous cohort study on outpatients from a Brazilian state (ES) [34], we found a prevalence of 18.55% for cardiac disease and 7.89% for diabetes. Mortality rates in this cohort were 50.05% and 31.82% for cardiac and diabetic patients, respectively.…”

“…Another aspect is the notification data of only hospitalized COVID-19 patients, so that we were unable to trace the profile of outpatients. However, our previous study using a dataset from Espírito Santo state [34] can be used for comparative purposes, as it involved COVID-19 outpatients.…”

Applications of artificial intelligence in battling against covid-19: A literature review

“…In [107] , multivariate logistic regression and a deep learning algorithm is used to predict the probability of a patient with mild symptoms developing malignant infection. A data set of 13,690 patients in Brazil is used in [108] to build a model that predicts the poor prognosis in covid-19 patients. The authors use machine learning to build the model.…”