Skip to main content

Table 8 Average correct rates and SDs in classifying chest CT images as COVID-19 positive/negative when Resnet-101 and each algorithm hyperparameter combination in Table 6 were used in five independent experimental runs

From: Classifying chest CT images as COVID-19 positive/negative using a convolutional neural network ensemble model and uniform experimental design method

Model# experiment number

Dataset

Experimental runs

 

1

2

3

4

5

Average

SD

Resnet-101#1

Training set

0.7124

0.7059

0.7075

0.7042

0.7026

0.7065

0.00376

Validation set

0.6061

0.596

0.6263

0.6162

0.6263

0.6142

0.01317

Resnet-101#2

Training set

0.7876

0.781

0.7876

0.7876

0.7892

0.7866

0.00321

Validation set

0.6566

0.6566

0.6566

0.6566

0.6566

0.6566

0

Resnet-101#3

Training set

0.9804

0.9755

0.982

0.9853

0.982

0.9810

0.00357

Validation set

0.8788

0.8788

0.8788

0.8788

0.8788

0.8788

0

Resnet-101#4

Training set

0.4951

0.5065

0.4951

0.4951

0.4951

0.4974

0.0051

Validation set

0.5051

0.4848

0.5051

0.5051

0.5051

0.5010

0.00908

Resnet-101#5

Training set

0.9085

0.8987

0.9101

0.9101

0.9101

0.9075

0.00497

Validation set

0.7879

0.7273

0.7879

0.798

0.798

0.7798

0.02979

Resnet-101#6

Training set

0.8366

0.8758

0.8317

0.835

0.835

0.8428

0.01852

Validation set

0.7475

0.7677

0.7475

0.7475

0.7475

0.7515

0.00903

Resnet-101#7

Training set

0.9918

0.9869

0.9853

0.9804

0.9869

0.9863

0.00409

Validation set

0.8687

0.899

0.8889

0.8586

0.899

0.8828

0.01835