Objective: This study aims to investigate the effectiveness of non-pharmaceutical interventions in managing the current Coronavirus pandemic and to predict the next wave of infection in Iran. Additionally, it seeks to apply machine learning (ML) to predict the requirement for oxygen-based treatment for hospitalized COVID-19 patients and to leverage ML on hospital data to support hospital managers and practitioners with the treatment of COVID-19 patients by predicting ICU admission, Mortality and length of stay (LOS).
Methods: For the investigation of non-pharmaceutical interventions, the study analyzes the number of cases and deaths before and after the interventions and examines the effective reproduction number of the infection under various scenarios using the SEIR generic model. In addition, the next wave of pandemic predicted using SEIR model. For the prediction of oxygen-based treatment requirement, demographic information, symptoms, and patient's background were extracted from the databases of two local hospitals in Iran. Preprocessing actions were applied, and related features were selected. Five ML models were implemented and compared based on their accuracy and capability. In parallel, the study predicts ICU admission, mortality, and LOS of COVID-19 patients using ML algorithms such as eXtreme Gradient Boosting (XGB), K-Nearest Neighbor (KNN), Random Forest (RF), bagged-CART (b-CART), and LogitBoost (LB). Ensemble stacking approaches were employed to further boost performance.
Results: The study finds that the maximum number of infected individuals in Iran is projected to occur around the end of May and the start of June 2021. It is concluded that a continuation of full lockdown and strict quarantine measures could help mitigate the outbreak. Regarding the prediction of oxygen-based treatment requirement, shortness of breath, cough, age, and fever were identified as the most important variables. The ML models show promising performance scores, with XGBoost