Prediction and Categorization Of COVID-19 Related Dermatological Manifestations Using Machine Learning

Abstract

COVID-19 is global epidemic instigated because of “severe acute respiratory syndrome corona virus 2 “. Fever, cough, tiredness, dyspnea, and hypogeusia/ hyposmia are all common signs. Dermatological indications have become more common in recent months among the extrapulmonary indicators associated with COVID-19. Our group proposed a taxonomy based on the polymorphic character of COVID-19-related cutaneous symptoms, which includes the following six primary clinical patterns:Urticarial rash, confluent erythematous/maculopapular/ morbilliform rash, papulovesicular exanthem, chilblain-like acral, livedo reticularis / racemosa-like, purpuric “vasculitic” patterns. To offer an evaluation of possible pathophysiological routes of COVID19- related cutaneous symptoms, this research focuses upon that clinical features ampersand therapeutic treatment of every category. Machine learning algorithms such as SVM, RF, DT, KNN, LR, and NB are used in the analysis.