ABSTRACT
Uzbekistan as well as the rest of the world faces the third wave of COVID-19 and uses machine learning algorithms to predict the adverse outcome during the admission of new patients. We collected the dataset of 1145 patients admitted to the Republican Center for Emergency Medicine. We use different machine learning models to predict the COVID-19 severe course. This study uses feature selection procedures based on statistical tests and the elimination of linearly dependent features. The resulting multilayer perceptron yields a ROC AUC of 86.9% on the test set outperforming other machine learning algorithms and several competing works. The model relies on easily collected features without blood laboratory testing. It increases an availability of the reliable risk prediction to developing countries. © 2022, Springer Nature Switzerland AG.
ABSTRACT
A retrospective analysis of 561 patients with confirmed COVID-19 was performed to determine the risk factors for severity and mortality which could predict the disease outcome in early stages. Patients were divided into 4 groups in accordance with disease severity: mild, moderate, severe and critical. And initial clinical and laboratory parameters of patients at admission were studied. The age of severe and deceased patients was significantly higher than patients with mild and moderate course (р=0.003). Patiens with severe disease and fatal outcome had higher incidence of concomitant diseases compared to patients with mild and moderate course (p=0.01). The time passed from onset of first symptoms and hospital admission was shorter in patients with mild and moderate disease than patients with severe and critical disease (р=0.0001). The leukocytosis, significant lymphopenia (р=0.0001), high D-dimer and ferritin levels were associated with severe disease. Male gender, old age, presence of concomitant diseases should be considered as risk factors for severe course and death at COVID-19. Copyright (c) 2021: Author(s).