Classifying microscopic images as acute lymphoblastic leukemia by Resnet ensemble model and Taguchi method

BMC Bioinformatics

Table 14 Classification accuracy, precision, recall, specificity, and F₁-score obtained by trained Resnet-101 individual models and by the Resnet101-7 ensemble model for the preliminary test set

Model	Accuracy	Precision	Recall	Specificity	F₁-score
Resnet-101-8249	0.8249	0.8436	0.8983	0.6867	0.8701
Resnet-101-8184	0.8184	0.8624	0.8589	0.7423	0.8607
Resnet-101-8452	0.8452	0.8582	0.9139	0.716	0.8852
Resnet-101-8125	0.8125	0.8376	0.8843	0.6775	0.8603
Resnet-101-8061	0.8061	0.8562	0.845	0.733	0.8505
Resnet-101-8281	0.8281	0.8376	0.9139	0.6667	0.8741
Resnet-101-8307	0.8307	0.8536	0.8942	0.7114	0.8734
Resnet101-7 ensemble	0.8506	0.8638	0.9155	0.7284	0.8889

ISSN: 1471-2105