Skip to main content

Table 7 Average correct rates and SDs in classifying chest CT images as COVID-19 positive or negative when VGG-19 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

VGG-19#1

Training set

0.8252

0.8415

0.8448

0.8399

0.8399

0.8383

0.00757

Validation set

0.8182

0.8182

0.8182

0.798

0.798

0.8101

0.01106

VGG-19#2

Training set

0.8578

0.8611

0.8709

0.8513

0.8497

0.8582

0.00851

Validation set

0.8485

0.798

0.8283

0.8182

0.8182

0.8222

0.01835

VGG-19#3

Training set

0.5621

0.5784

0.6405

0.5931

0.4951

0.5738

0.05285

Validation set

0.5859

0.6465

0.6566

0.6061

0.4949

0.5980

0.06447

VGG-19#4

Training set

0.8072

0.8056

0.7974

0.8039

0.8088

0.8046

0.00441

Validation set

0.7879

0.7778

0.7879

0.7677

0.798

0.7839

0.01152

VGG-19#5

Training set

0.915

0.9036

0.915

0.915

0.8938

0.9085

0.00958

Validation set

0.8384

0.8182

0.8081

0.8384

0.8182

0.8243

0.01355

VGG-19#6

Training set

0.9346

0.9265

0.9265

0.9346

0.9314

0.9307

0.00407

Validation set

0.8485

0.8485

0.8485

0.8485

0.8586

0.8505

0.00452

VGG-19#7

Training set

0.8775

0.7026

0.7435

0.8693

0.8415

0.8069

0.07901

Validation set

0.7374

0.7071

0.7475

0.8182

0.8081

0.7637

0.0477

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us