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

Table 14 Classifier accuracy, precision, recall, specificity, and F₁-score obtained by different trained CNN models and by the COVID19-CNN ensemble model for the testing set

Model	Accuracy	Precision	Recall	Specificity	F₁-score
Resnet-101#3	0.912	0.913	0.913	0.911	0.913
Resnet-101#7	0.934	0.935	0.935	0.933	0.935
DenseNet-201#3	0.945	0.956	0.935	0.956	0.945
DenseNet-201#7	0.912	0.865	0.978	0.844	0.918
Inception-v3#7	0.912	0.896	0.935	0.889	0.915
Inception-ResNet-v2#3	0.956	0.938	0.978	0.933	0.957
Inception-ResNet-v2#7	0.934	0.917	0.957	0.911	0.936
COVID19-CNN	0.967	0.957	0.978	0.956	0.968

ISSN: 1471-2105