주메뉴 바로가기 본문 바로가기

발표 논문

상세페이지
이달의 논문 2026년 6월
등록일 : 2026.08.21

 

. 2026 Jun 24;13(7):728.
doi: 10.3390/bioengineering13070728.

Automated Craniofacial Artery Segmentation with Vessel Enhancement-Guided Deep Learning

Affiliations<button aria-controls="expanded-authors" aria-expanded="true" class="more-details active" data-alt-text="Collapse" data-fetch-url="/42510395/long-authors/" data-id-prefix="" data-pinger-ignore="" id="toggle-authors" style="box-sizing: inherit; color: rgb(50, 58, 69); font-style: inherit; font-variant: inherit; font-weight: normal; font-stretch: inherit; font-size: 1.6rem; line-height: 2.4rem; font-family: BlinkMacSystemFont, -apple-system, " segoe="" ui",="" roboto,="" oxygen,="" ubuntu,="" cantarell,="" "fira="" sans",="" "droid="" "helvetica="" neue",="" sans-serif;="" font-optical-sizing:="" inherit;="" font-size-adjust:="" font-kerning:="" font-feature-settings:="" font-variation-settings:="" font-language-override:="" margin:="" auto="" 0px="" 0px;="" overflow:="" visible;="" text-transform:="" lowercase;="" appearance:="" none;="" cursor:="" pointer;="" -webkit-font-smoothing:="" antialiased;="" background-color:="" rgb(255,="" 255,="" 255);="" border-width:="" border-style:="" border-color:="" currentcolor;="" border-radius:="" 3px;="" padding:="" 1rem="" 2.5rem;="" width:="" auto;="" transition:="" color="" 0.3s,="" background-color="" 0.3s;="" position:="" relative;="" height:="" 2.4rem;="" outline-offset:="" 0px;"="">Collapse

Affiliations

  • 1SKIA Inc., 1502, 288 Digital-ro, Guro-gu, Seoul 08390, Republic of Korea.
  • 2Department 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.

Abstract

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.

Keywords: arteries; computed tomography angiography; deep learning; image processing.

 
  • 현재 페이지를 인쇄하기
페이지 처음으로 이동
05505 서울특별시 송파구 올림픽로 43길 88 서울아산병원
TEL 1688-7575 webmaster@amc.seoul.kr
Copyright@2014 by Asan Medical Center. All Rights reserved.
  • 바로가기
  • 바로가기
  • 바로가기
  • 바로가기
  • 서울아산병원, 20년 연속 존경받는 병원 1위
  • 서울아산병원, 美 뉴스위크 평가 세계 22위·국내 1위
  • 서울아산병원, 정보보호 관리체계 ISMS 인증 획득