Skip to main content

Table 3 The top 20 variables are screened after combining "%IncMSE" and "IncNodePurity"

From: Development and validation of a machine learning-based nomogram for predicting HLA-B27 expression

Type

%IncMSE

IncNodePurity

Type

%IncMSE

IncNodePurity

GLB

29.528091

24.689556

WBC

10.300806

5.874292

ALB/GLB

15.510579

11.479635

BPC

9.368241

5.840838

ALP

12.608679

8.768030

MCHC

10.315155

5.801214

TP

13.707016

7.469052

DBil

11.072646

5.122757

CRP

11.578667

6.706310

BUN

6.273594

4.754315

MPV

9.930096

6.357899

MCV

7.115748

4.583968

HCO

9.131109

6.168996

NEUT

7.071780

4.488072

RBC

9.802478

6.075653

ESR

8.378661

4.398344

NEUT#

9.951013

5.949444

Cys-C

9.704187

4.253772

PAB

11.458047

5.889966

MCH

4.973804

4.240079