Figure 2 | Scientific Reports

Figure 2

From: The preliminary in vitro study and application of deep learning algorithm in cone beam computed tomography image implant recognition

Figure 2

Confusion matrix results for 9 algorithms. Note: The nine algorithmic models are GoogleNet, InceptionResNetV2, InceptionV3, ResNet50, ResNet50V2, ResNet101, ResNet101V2, ResNet152 and ResNet152V2, respectively; the horizontal coordinates represent the predicted category, the vertical coordinates represent the real category, and the numbers 0–26 indicate the implant classification information as in Table 1.

Back to article page