Table 1 Comparison between the manual labels (thresholding \(+\) manual refinement), the ground truth (GT) manually fine-edited atlas propagated labels and UNETR outputs: average Dice per vessel ROI
Method | SVC | AO | AD | DAO | MPA | RPA | LPA |
|---|---|---|---|---|---|---|---|
LP+editing | \(0.77\pm 0.03\) | \(0.68\pm 0.04\) | \(0.73\pm 0.03\) | \(0.77\pm 0.4\) | \(0.78\pm 0.05\) | \(0.64\pm 0.01\) | \(0.57\pm 0.05\) |
UNETR | \(0.76\pm 0.05\) | \(0.68\pm 0.04\) | \(0.73\pm 0.03\) | \(0.78\pm 0.04\) | \(0.73\pm 0.06\) | \(0.66\pm 0.05\) | \(0.57\pm 0.07\) |
Method | INV | PVS | LSA | BCA | LCCA | AZV | ALSA |
LP+editing | \(0.68\pm 0.02\) | \(0.48\pm 0.08\) | \(0.68\pm 0.05\) | \(0.58\pm 0.05\) | \(0.47\pm 0.23\) | \(0.64\pm 0.06\) | \(0.68\pm 0.05\) |
UNETR | \(0.72\pm 0.06\) | \(0.48\pm 0.13\) | \(0.71\pm 0.07\) | \(0.66\pm 0.09\) | \(0.58\pm 0.19\) | \(0.70\pm 0.02\) | \(0.76\pm 0.05\) |