From: Convolutional neural networks for detection of transthyretin amyloidosis in 2D scintigraphy images
AUC | Accuracy | Grade 0–2 | Grade 3 | |||
---|---|---|---|---|---|---|
Precision | Recall | Precision | Recall | |||
Linear | 0.94 | 0.99 | 1.00 | 0.99 | 0.73 | 0.89 |
Residual | 0.94 | 0.99 | 1.00 | 1.00 | 0.83 | 0.89 |
VGG16 | 0.92 | 0.99 | 1.00 | 0.99 | 0.62 | 0.85 |
ResNet50 | 0.82 | 0.99 | 0.99 | 1.00 | 0.94 | 0.63 |
InceptionV3 | 0.80 | 0.99 | 0.99 | 1.00 | 0.80 | 0.59 |
MobileNet | 0.72 | 0.99 | 0.99 | 1.00 | 0.92 | 0.44 |