Table 1.
Previous studies on automated leukemia classification.
Fig 1.
The outline of the proposed methodology.
Fig 2.
Sample images of the datasets.
(a) ALL, (b) ALL-IDB2, (c) C-NMC, and (d) Mixture-Leukemia.
Table 2.
The properties of the used datasets.
Table 3.
Splitting train-validation-test of data.
Fig 3.
The 5-fold cross-validation procedure.
Fig 4.
The architecture of the proposed networks.
Fig 5.
The architecture of the PCAB.
Fig 6.
The architecture of the SCAB.
Table 4.
Experimental hardware and software environments.
Table 5.
Hyperparameter configuration of the networks.
Fig 7.
The accuracy and loss curves of the ResNet18 models.
Fig 8.
The accuracy and loss curves of the MobileNetv4 models.
Fig 9.
The confusion matrix of the ResNet18 models.
Fig 10.
The confusion matrix of the MobileNetv4 models.
Table 6.
Performance metrics equations for leukemia classification.
Table 7.
The performance metrics of the ResNet18 models.
Table 8.
The performance metrics of the MobileNetv4 models.
Table 9.
The accuracy of the ResNet18 models.
Table 10.
The accuracy of the MobileNetv4 models.
Table 11.
Comparison of the time and computational complexities.
Table 12.
Ablation experiment data for different pooling layers.
Table 13.
Ablation experiment data for different convolution layers.
Table 14.
Overall comparison of the model with existing models.
Fig 11.
The ANOVA results for models.
Fig 12.
Feature maps from the last convolution layers of the models on the datasets.
() are the images and their feature maps from the ALL, ALL-IDB2, C-NMC, and Mixture-Leukemia datasets, whereas (1) include input images, (
) are the output feature maps of the last convolution layers of each model.
Fig 13.
Grad-CAM images from the last convolution layers of the models on the datasets.
() are the images and their activation maps from the ALL, ALL-IDB2, C-NMC, and Mixture-Leukemia datasets, whereas (1) include input images, (
) are the Grad-CAM images extracted from the last convolution layers of the ResNet18, ResNet18PCAB, ResNet18SCAB, MobileNetv4, MobileNetv4PCAB, and MobileNetv4SCAB, respectively.
Fig 14.
Samples of images of correct and incorrect prediction of the datasets.
() are correctly classified while (
) are misclassified.