Skip to main content

Table 9 Average correct rates and SDs in classifying chest CT images as COVID-19 positive/negative when DenseNet-201 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

DenseNet-201#1

Training set

0.7484

0.7484

0.7484

0.7533

0.7402

0.7477

0.00472

Validation set

0.6263

0.6263

0.6465

0.6364

0.6465

0.6364

0.0101

DenseNet-201#2

Training set

0.7745

0.7745

0.781

0.7859

0.7859

0.7804

0.00571

Validation set

0.6667

0.6667

0.6667

0.6566

0.6768

0.6667

0.00714

DenseNet-201#3

Training set

0.9902

0.9853

0.9886

0.9967

0.9837

0.9889

0.00507

Validation set

0.8788

0.8889

0.8788

0.8485

0.8687

0.8727

0.01532

DenseNet-201#4

Training set

0.598

0.598

0.6062

0.6078

0.6013

0.6023

0.00457

Validation set

0.4949

0.4949

0.4848

0.4949

0.5051

0.4949

0.00718

DenseNet-201#5

Training set

0.9265

0.9281

0.9281

0.9248

0.9167

0.9248

0.00475

Validation set

0.7677

0.7677

0.8182

0.7778

0.7879

0.7839

0.02094

DenseNet-201#6

Training set

0.9183

0.9183

0.9134

0.9118

0.9118

0.9147

0.00333

Validation set

0.7677

0.7677

0.7677

0.8182

0.7576

0.7758

0.02411

DenseNet-201#7

Training set

0.9951

0.9967

0.9869

0.9886

0.9967

0.9928

0.00469

Validation set

0.899

0.9293

0.8687

0.8586

0.8687

0.8849

0.0291

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us