의학물리학

본문글자크기
  • [Phys Med Biol.] Automatic 3D cephalometric annotation system using shadowed 2D image-based machine learning.2D 영상 기반 머신러닝을 이용한 자동 3D 두부 계측 주석 시스템 연구

    연세대 / 이성민, 서진근*

  • 출처
    Phys Med Biol.
  • 등재일
    2019 Feb 20
  • 저널이슈번호
    64(5):055002. doi: 10.1088/1361-6560/ab00c9.
  • 내용

    바로가기  >

    Abstract
    This paper presents a new approach to automatic three-dimensional (3D) cephalometric annotation for diagnosis, surgical planning, and treatment evaluation. There has long been considerable demand for automated cephalometric landmarking, since manual landmarking requires considerable time and experience as well as objectivity and scrupulous error avoidance. Due to the inherent limitation of two-dimensional (2D) cephalometry and the 3D nature of surgical simulation, there is a trend away from current 2D to 3D cephalometry. Deep learning approaches to cephalometric landmarking seem highly promising, but there exist serious difficulties in handling high dimensional 3D CT data, dimension referring to the number of voxels. To address this issue of dimensionality, this paper proposes a shadowed 2D image-based machine learning method which uses multiple shadowed 2D images with various lighting and view directions to capture 3D geometric cues. The proposed method using VGG-net was trained and tested using 2700 shadowed 2D images and corresponding manual landmarkings. Test data evaluation shows that our method achieved an average point-to-point error of 1.5 mm for the seven major landmarks.

     


    Author information

    Lee SM1, Kim HP, Jeon K, Lee SH, Seo JK.
    1
    Department of Computational Science and Engineering, Yonsei University, Seoul, Republic of Korea.

  • 편집위원

    - 인공지능은 이제 거의 모든 분야에서 각광을 받고 있는 듯 하다. 하지만 각 분야의 특성에 맞게 구체적인 적용방법은 약간 달리해야 좋은 결과를 얻는 것 같다. 몇 장의 3D 이미지로부터 2D 이미지를 추출하고 lighting을 달리한 많은 영상을 만듦으로써 학습데이터의 양을 늘리고 이로 인하여 좋은 결과를 얻었다. 앞으로는 이런 적용방법을 달리 설정해 주는 인공지능도 나오지 않을까?

    2019-04-17 16:05:57

  • 덧글달기
    덧글달기
       IP : 44.210.86.29

    등록