Skip to main content
Fig. 5 | BMC Medical Informatics and Decision Making

Fig. 5

From: Deep learning model for differentiating nasal cavity masses based on nasal endoscopy images

Fig. 5

Grad-CAM results of the best model among five-folds for normal, NPs, and IP on nasal endoscope images. a Positive results for the classification of normal, NPs, and IP in the nasal endoscope images and the corresponding heat maps for our model. b Negative results for the classification of normal, NPs, and IP in the nasal endoscope images and the corresponding heatmaps for our model. Grad-CAM results for each class were extracted independently, with the Grad-CAMs for each class extracted by deep learning and being consistent with specific regions delineated by experts

Back to article page