Skip to main content

Table 4 Performance evaluation of the MSFF on the stage data D1–D4 generated by independent test (non-cross-validation method)

From: Exploratory study on classification of chronic obstructive pulmonary disease combining multi-stage feature fusion and machine learning

Phased data set

Model

acc

sn

sp

mcc

Curve

D1

RF

0.750

0.773

0.722

0.495

0.748

SVM

0.775

0.955

0.556

0.568

0.755

KNN

0.575

0.773

0.333

0.118

0.553

PLS

0.775

0.864

0.667

0.545

0.765

D2

RF

0.750

0.773

0.722

0.495

0.748

SVM

0.675

0.818

0.500

0.338

0.659

KNN

0.600

0.818

0.333

0.174

0.576

PLS

0.775

0.864

0.667

0.545

0.765

D3

RF

0.700

0.727

0.667

0.394

0.697

SVM

0.800

0.864

0.722

0.595

0.793

KNN

0.600

0.818

0.333

0.174

0.576

PLS

0.775

0.864

0.667

0.545

0.765

D4

RF

0.775

0.818

0.722

0.544

0.770

SVM

0.800

0.955

0.611

0.614

0.783

KNN

0.600

0.818

0.333

0.174

0.576

PLS

0.775

0.864

0.667

0.545

0.765

  1. D1, D2, D3 and D4 represent the AECOPD patient data of stage 1 to 4. At each stage, there are 408 samples