Skip to main content

Table 2 Feature importance results and values

From: Machine learning-based prediction of COVID-19 mortality using immunological and metabolic biomarkers

Description (Feature name)

Feature value

Age

0.11057

Urea [Moles/volume] in serum or plasma

0.0499

Procalcitonin [Mass/volume] in serum or plasma

0.0390

Albumin/Globulin [Mass ratio] in serum or plasma

0.0357

Magnesium [Moles/volume] in serum or plasma

0.0314

Base excess in blood

0.0298

Creatinine [Moles/volume] in serum or plasma

0.0289

Glomerular filtration rate/1.73 m2predicted[VolumeRate/Area]inserumorplasma by Creatinine-based formula (CKD-EPI)

0.02623

Calcium [Moles/volume] in serum or plasma

0.0252

Erythrocytes [/volume] in blood

0.02480

Carbon dioxide [Partial pressure] in blood

0.02071

Albumin [Mass/volume] in serum or plasma

0.01949

Lymphocytes/100 leukocytes in blood

0.01921

Lactate dehydrogenase [Enzymatic activity/volume] in serum or plasma

0.0179

Ferritin [Mass/volume] in serum or plasma

0.01778

Creatine kinase [Enzymatic activity/volume] in serum or plasma

0.01737

PH of blood

0.01696

C reactive protein [Mass/volume] in serum or plasma

0.01472