EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Automated Craniofacial Artery Segmentation with Vessel Enhancement-Guided Deep Learning

Hyeonju Park iD, Young Chul Kim, Kyoyeong Koo iD, Sangyun Kang, Jong Woo Choi, Chan-Ung Park iD

Bioengineering · 24.06.2026 · 13 (7) · 728

DOI: 10.3390/bioengineering13070728 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioengineering, online erschienen am 24.06.2026. Verfasst von Hyeonju Park und 5 weiteren Autor:innen. Beteiligte Einrichtung: SKIA Inc., 1502, 288 Digital-ro, Guro-gu, Seoul 08390, Republic of Korea. Die Arbeit nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Korea Health Industry Development Institute.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Computed tomography angiography (CTA)-based segmentation of the superficial temporal arteries (STAs) and facial vessels (FVs) is important for neurosurgical and reconstructive planning. Nevertheless, segmentation of STAs and FVs remains challenging because of their small caliber, tortuous courses, and proximity to high-intensity bony structures. This study aims to develop a deep learning framework for accurate automated segmentation of these craniofacial vessels. A single-input 3D nnU-Net v2 model was trained using raw CTA 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. Instead of being used as an additional input channel, the FVM was incorporated into the loss function as a spatial prior to guide the network toward vessel boundaries and distal branches. 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. For the STAs, the Average Symmetric Surface Distance decreased from 6.543 mm to 2.941 mm. Qualitative evaluation further showed reduced segmentation noise and fewer false positives near bone and distal branches. These findings suggest that integrating classical vessel enhancement into deep learning supervision can improve morphologically consistent craniofacial vessel segmentation and support preoperative surgical planning.

Schlagwörter

Stichwörter der Autor:innen

arteriescomputed tomography angiographydeep learningimage processing

Autor:innen und Institutionen

  • SKIA Inc., 1502, 288 Digital-ro, Guro-gu, Seoul 08390, Republic of Korea.
  • Department of Plastic and Reconstructive Surgery, Asan Medical Center, Ulsan University College of Medicine, 88 Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Republic of Korea.

ORCID-Profile: Hyeonju Park, Kyoyeong Koo, Chan-Ung Park

Förderung

  • Korea Health Industry Development Institute

Zitieren

APA

Park, H., Kim, Y. C., Koo, K., Kang, S., Choi, J. W., & Park, C. U. (2026). Automated Craniofacial Artery Segmentation with Vessel Enhancement-Guided Deep Learning. Bioengineering, 13(7), 728. https://doi.org/10.3390/bioengineering13070728

Vancouver

Park H, Kim YC, Koo K, Kang S, Choi JW, Park CU. Automated Craniofacial Artery Segmentation with Vessel Enhancement-Guided Deep Learning. Bioengineering. 2026;13(7):728. doi:10.3390/bioengineering13070728

BibTeX
@article{Park2026,
  author = {Park, Hyeonju and Kim, Young Chul and Koo, Kyoyeong and Kang, Sangyun and Choi, Jong Woo and Park, Chan-Ung},
  title = {Automated Craniofacial Artery Segmentation with Vessel Enhancement-Guided Deep Learning},
  journal = {Bioengineering},
  year = {2026},
  volume = {13},
  number = {7},
  pages = {728},
  doi = {10.3390/bioengineering13070728}
}
Kostenlos starten →