Fig. 2: The performance of the training and testing process of head and neck CTA postprocessing. | Nature Communications

Fig. 2: The performance of the training and testing process of head and neck CTA postprocessing.

From: Rapid vessel segmentation and reconstruction of head and neck angiograms using 3D convolutional neural network

Fig. 2

Illustration of the changes of the DSC value (a), V-score value (b), and the recall value (c) over the training and validation sets in the vessel segmentation task. d, e The variety of the DSC value and recall value over the training and validation sets in the bone segmentation task. The overall DSC, V-score, recall, and accuracy values in testing set of the pipeline are shown in f.

Back to article page