Table 4 Difference in the Accuracy between the CNN and Physicians (McNemar test)
From: Automated detection of anterior cruciate ligament tears using a deep convolutional neural network
| Â | Accuracy(%) | P-value |
|---|---|---|
CNN | 88.5 | - |
K1 | 89.0 | 0.8788 |
K2 | 92.5 | 0.1824 |
K3 | 88.5 | 1.0000 |
K4 | 87.0 | 0.6617 |
K5 | 67.5 |  < .0001* |
K6 | 89.5 | 0.7456 |
K7 | 85.0 | 0.3173 |
K8 | 88.0 | 0.8759 |
K9 | 73.5 | 0.0001* |
K10 | 85.0 | 0.2858 |
R1 | 84.8 | 0.6717 |
R2 | 85.5 | 0.3657 |