Automated Craniofacial Artery Segmentation with Vessel Enhancement-Guided Deep Learning.

A single-input 3D nnU-Net v2 model was trained using raw computed tomography angiography volumes, while a Fusion-based Vesselness Map (FVM) was constructed from multiscale vessel-enhancement filters to emphasize small vascular structures and suppress irrelevant regions such as the skull and skin.

In 72 clinical cases, the FVM-guided model improved segmentation accuracy compared with a baseline model trained with Dice Focal Loss, particularly in boundary delineation.

Qualitative evaluation further showed reduced segmentation noise and fewer false positives near bone and distal branches.