Table 3.

The most important markers of COVID-19 severity identified by recursive feature elimination (RFE) in conjunction with machine learning algorithms such as Logistic Regression (LR), Support Vector Machine (SVM) and Random Forest (RF).

RFE + LRaccuracy = 83%	RFE + SVMaccuracy = 87%	RFE + RFaccuracy = 85%
MDC	MDC	Creatinine
Fibrinogen	Glucose	CRP
Creatinine	Creatinine	MIG
Glucose	IL-6	Monocytes
MIG	Fibrinogen	Fibrinogen
Monocytes	MIG	MDC
CRP	CRP	TNFa
IL-6	LDH	IL-6
LDH	AHTV	IL-18
TNFa	Monocytes	Glucose
		ALT
		D-dimer

In each column the markers are arranged in descending order of their importance, determined by the corresponding algorithm.