Research Article

Multilevel Strip Pooling-Based Convolutional Neural Network for the Classification of Carotid Plaque Echogenicity

Figure 9

Confusion matrices of the compared networks for the classification of carotid plaques. ER, IM, and EL represent echo-rich plaques, intermediate plaques, and echolucent plaques, respectively.
(a) ResNext50
(b) DRN-d22
(c) MobileNet-v2
(d) DenseNet121
(e) EfficientNet-b7
(f) MSP-based VGGNet