For the Segmentation Contest, the clinical metrics are those that are the most widely used in cardiac clinical practice and the geometrical metrics are the classic ones used in the segmentation evaluation.
For the Classification Contest, the metrics is the classification accuracy (normal vs pathological cases).
The python code of the metrics is available on GitHub.
The clinical metrics include the average errors for the volume of the myocardium of the left ventricle (in mm3), the volume (in mm3) and the percentage of MI and no-reflow.
The geometrical metrics are the average DICE index for the different areas and Hausdorff distance (in 3D) for the myocardium. The Dice index gives an overall information about the quality of the segmentation, the Hausdorff distance highlights the outliers.
For each geometrical and clinical metric, a ranking will be done, and the final ranking consists of the sum of the ranking for each metric.
As it is a binary classification, only the classification accuracy (normal vs pathological cases) is mandatory.