ABSTRACT
The 2019 novel coronavirus disease (COVID-19) epidemic was officially announced by the World Health Organization (WHO) as an international public health emergency. The medical research world is responding to the COVID-19 pandemic at breathtaking speed. Most of the studies related to this outbreak identify the epidemiology and clinical characteristics of infected patients and focus on its short-term effects. However, there are many studies with inappropriate study design, data mining, and statistical analysis. Proper design and reliability assessment of COVID-19 diagnosis systems (e.g., proper feature selection, classification, and performance assessment) must be performed. Also, advanced statistical methods (e.g., multistate and competing risk models) are required to avoid the risk of bias in prognosis systems. Moreover, many studies may be too small and poorly designed to be helpful, merely adding to the COVID-19 noise. Additionally, trials without a control group, non-randomized and imbalanced trials are common problems of experimental designs. Thus, in this chapter, we aim to address the critical methods to use in the diagnosis and prognosis of COVID-19.
