Predicting peri-operative outcomes using machine learning (ML)-based processes has been shown to improve decision making for subsequent steps. For quite some time now, ML models have been utilized in many different application domains to identify and prioritize risk indicators. Commonly employed to deal with forecasting issues are a variety of prediction approaches. This research shows that ML models can accurately predict the future number of individuals infected with COVID-19, a virus that is now being treated as a possible threat to humankind. This research has employed four common forecasting models to predict the risk variables of COVID-19: linear regression (LR), least absolute shrinkage and selection operator (LASSO), support vector machine (SVM), and exponential smoothing (ES). Each model forecasts the rate of infection, the likelihood of mortality, and the likelihood of recovery during the following 10 days. The study's findings indicate that these techniques hold promise for addressing the current COVID-19 pandemic scenario. The results show that, given the provided dataset, ES performs the best in all prediction scenarios, followed by LR and LASSO, which perform well in forecasting the new confirmed cases, death rate, and recovery rate, respectively. SVM performs badly in all prediction situations.

Predicting peri-operative outcomes using machine learning (ML)-based processes has been shown to improve decision making for subsequent steps.