Skip to main content

Table 3 Importance values (VIP) of 21 most differentiation among 108 variables used in the primary model

From: Machine-learning-based COVID-19 mortality prediction model and identification of patients at low and high risk of dying

 

Variables

VIP

1

CAD

2.1045

2

Diabetes

1.9098

3

Age > 65

1.7433

4

AMS

1.6922

5

Dementia

1.6309

6

Nursing home

1.5545

7

Q2 saturation < 88

1.5252

8

yno2

1.4903

9

Consolidation

1.4654

10

Hypertension

1.4226

11

Atrial fibrillation

1.3789

12

Alcohol

1.2563

13

Chest pain

1.1566

14

Peripheral vascular disease

1.1133

15

Prothrombin

1.0855

16

Stroke

1.0665

17

Headache

1.0412

18

Dyspnea

1.0212

19

CRP

1.0125

20

Lactate

1.0012

21

Smoking

1.0011

Back to article page

Critical Care

ISSN: 1364-8535