Table 15 Evaluation metrics for the proposed ensemble per each class of ISIC 2019 dataset.

From: Multimodal deep learning ensemble framework for skin cancer detection

Class

TP

FP

FN

TN

Precision

Recall

F1 Score

Specificity

AUC

MEL

403

122

26

2545

0.768

0.944

0.847

0.942

0.953

NV

1165

11

129

1791

0.991

0.900

0.943

0.991

0.974

BCC

319

18

32

2727

0.947

0.909

0.927

0.992

0.936

AK

87

14

14

2981

0.861

0.861

0.861

0.994

0.923

BKL

234

12

24

2826

0.951

0.907

0.929

0.995

0.954

DF

21

3

0

3072

0.429

1.000

0.600

0.989

0.951

VASC

27

14

9

3046

0.659

0.964

0.783

0.994

0.984

SCC

52

7

2

3035

0.881

0.963

0.920

0.997

0.969

Back to article page