Table 4 Analysis of the false positives predicted the 3D-CNN-TR model output before and after post-processing, for our private dataset (*: foramen magnum; ICA: internal carotid artery, A1: A1 segment of anterior cerebral artery, A2: A2 segment of anterior cerebral artery, ACA: anterior cerebral artery, PCA: posterior cerebral artery).
All FPs | FPs removed by | |||||
|---|---|---|---|---|---|---|
Method 1 | Method 2 | Method 3 | Method 4 | Method 5 | ||
Vein | 53 | 36 | 49 | 48 | 49 | 49 |
Vein of Galen | 12 | 0 | 12 | 12 | 12 | 12 |
Venous sinus | 0 | 0 | 0 | 0 | 0 | 0 |
Cavernous venous sinus | 4 | 0 | 0 | 0 | 0 | 0 |
Other intracranial veins | 2 | 1* | 2 | 2 | 2 | 2 |
Extracranial veins | 35 | 35 | 35 | 34 | 35 | 35 |
Artery | 97 | 22 | 22 | 7 | 30 | 22 |
Cervical artery | 22 | 22 | 14 | 7 | 22 | 22 |
ICA infundibulum/branch points | 37 | 0 | 1 | 0 | 1 | 0 |
MCA branch points | 10 | 0 | 1 | 0 | 1 | 0 |
Caliber change at A1/A2 or ACA | 5 | 0 | 1 | 0 | 1 | 0 |
Vertebrobasilar confluence | 2 | 0 | 2 | 0 | 2 | 0 |
ICA occlusion | 4 | 0 | 0 | 0 | 0 | 0 |
Ectatic basilar artery | 1 | 0 | 0 | 0 | 0 | 0 |
Basilar tip confluence | 8 | 0 | 1 | 0 | 1 | 0 |
Vertebral artery caliber change | 2 | 0 | 0 | 0 | 0 | 0 |
PCA caliber change | 1 | 0 | 1 | 0 | 1 | 0 |
Daughter aneurysm | 1 | 0 | 0 | 0 | 0 | 0 |
Artifact (e.g. clip, calc) | 4 | 0 | 1 | 0 | 1 | 0 |
Vessel | 15 | 12 | 6 | 5 | 14 | 13 |
Intracranial | 3 | 0 | 2 | 1 | 2 | 1 |
Extracranial | 12 | 12 | 4 | 4 | 12 | 12 |
Tissue | 17 | 8 | 7 | 6 | 10 | 10 |
Choroid plexus | 4 | 0 | 2 | 2 | 2 | 2 |
Pineal gland | 0 | 0 | 0 | 0 | 0 | 0 |
Posterior clinoid process | 3 | 0 | 0 | 0 | 0 | 0 |
Other intracranial | 2 | 0 | 0 | 0 | 0 | 0 |
Extracranial | 8 | 8 | 5 | 4 | 8 | 8 |
Total | 182 | 78 | 84 | 66 | 103 | 94 |
Intracranial | 105 | 1 | 26 | 17 | 26 | 17 |
Extracranial | 77 | 77 | 58 | 49 | 77 | 77 |
