Skip to main content
Fig. 3 | Brain Informatics

Fig. 3

From: Deep learning-based algorithm for postoperative glioblastoma MRI segmentation: a promising new tool for tumor burden assessment

Fig. 3

Positive effect of STAPLE fusion for resection cavity segmentation. Results obtained from the fivefold cross-validation process (fold seg.) are merged by the STAPLE algorithm to obtain a final result (STAPLE seg.). The figure shows, as an example, how the STAPLE convergence is able to recognize oversegmentation of a hypointense region misclassified as resection cavity (blue: cavity, yellow: enhancing, green: whole)

Back to article page