Skip to main content
. 2020 Nov 18;10(11):969. doi: 10.3390/diagnostics10110969

Table 3.

Classification metrics of HAM10000 low- and high-resolution images with aggregated clinical data using eight different DNNs.

Low-Resolution Images 1 High-Resolution Images
DNN 2 Accuracy 3 Error Rate Accuracy Error Rate
ResNet34 77.43 22.57 78.84 21.16
ResNet50 79.31 20.69 80.72 19.28
ResNet101 77.55 22.45 78.96 21.04
SEResNet50 80.01 19.99 80.72 19.28
VGG16 80.25 19.75 81.65 18.35
VGG19 79.43 20.57 79.02 20.98
EfficientNetB5 75.73 24.27 77.14 22.86
MobileNet 81.24 18.76 84.73 15.27

1 low-resolution image: 300 × 224 RGB; high-resolution image: 300 × 224 RGB. 2 Deep neural network; 3 accuracy and error rates are expressed as percentages.