Fig. 6
From: Automatic segmentation of the great arteries for computational hemodynamic assessment
Segmentation metrics. The model’s segmentation (ML) is compared with the an observer’s segmentation (SO) for two different datasets: our original data (ML vs SO) and an external test set from a different site and vendor (Ext-ML vs Ext-SO). A, B: Confusion-based similarity metrics: dice score and intersection-over-union (IoU). C, D Distance-based similarity metrics: Hausdorff distance (HD) and average surface distance (ASD), measured in pixels