BMC Bioinformatics

Table 11 Average correct rates and SDs in classifying chest CT images as COVID-19 positive/negative when Inception-ResNet-v2 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
Model# experiment number	Dataset	1	2	3	4	5	Average	SD
Inception-ResNet-v2#1	Training set	0.7092	0.7157	0.7271	0.7157	0.7141	0.7164	0.00657
Inception-ResNet-v2#1	Validation set	0.6061	0.6061	0.6162	0.6061	0.6061	0.6081	0.00452
Inception-ResNet-v2#2	Training set	0.6471	0.6503	0.6454	0.6487	0.6307	0.6444	0.00789
Inception-ResNet-v2#2	Validation set	0.5253	0.5051	0.5253	0.5051	0.5051	0.5132	0.01106
Inception-ResNet-v2#3	Training set	0.9755	0.9837	0.9853	0.9869	0.9788	0.9820	0.00475
Inception-ResNet-v2#3	Validation set	0.899	0.8687	0.899	0.8788	0.8586	0.8808	0.01807
Inception-ResNet-v2#4	Training set	0.5114	0.5049	0.5098	0.5098	0.5033	0.5078	0.00352
Inception-ResNet-v2#4	Validation set	0.4848	0.4848	0.4848	0.4848	0.5051	0.4889	0.00908
Inception-ResNet-v2#5	Training set	0.8971	0.902	0.9069	0.9003	0.9069	0.9026	0.00427
Inception-ResNet-v2#5	Validation set	0.7677	0.7778	0.7475	0.7677	0.7879	0.7697	0.01498
Inception-ResNet-v2#6	Training set	0.7958	0.7876	0.8056	0.7876	0.781	0.7915	0.00946
Inception-ResNet-v2#6	Validation set	0.6768	0.7172	0.697	0.7172	0.6566	0.6930	0.02634
Inception-ResNet-v2#7	Training set	0.9886	0.9755	0.9918	0.9869	0.9902	0.9866	0.00647
Inception-ResNet-v2#7	Validation set	0.9192	0.9192	0.899	0.899	0.9091	0.9091	0.0101

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com