Robust variational segmentation of 3D bone CT data with thin cartilage interfaces

Tarun Gangwar, Jeffrey W Calder, Takashi Takahashi, Joan E Bechtold, Dominik Schillinger

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

We present a two-stage variational approach for segmenting 3D bone CT data that performs robustly with respect to thin cartilage interfaces. In the first stage, we minimize a flux-augmented Chan–Vese model that accurately segments well-separated regions. In the second stage, we apply a new phase-field fracture inspired model that reliably eliminates spurious bridges across thin cartilage interfaces, resulting in an accurate segmentation topology, from which each bone object can be identified. Its mathematical formulation is based on the phase-field approach to variational fracture, which naturally blends with the variational approach to segmentation. We successfully test and validate our methodology for the segmentation of 3D femur and vertebra bones, which feature thin cartilage regions in the hip joint, the intervertebral disks, and synovial joints of the spinous processes. The major strength of the new methodology is its potential for full automation and seamless integration with downstream predictive bone simulation in a common finite element framework.

Original languageEnglish (US)
Pages (from-to)95-110
Number of pages16
JournalMedical Image Analysis
Volume47
DOIs
StatePublished - Jul 1 2018

Fingerprint

Cartilage
Bone
Bone and Bones
Intervertebral Disc
Automation
Hip Joint
Femur
Spine
Joints
Topology
Fluxes

Keywords

  • 3D bone CT data
  • Femur extraction
  • Flux-augmented Chan–Vese model
  • Phase-field fracture mechanics
  • Thin cartilage interfaces
  • Variational segmentation
  • Vertebra extraction
  • Voxel finite elements

PubMed: MeSH publication types

  • Journal Article
  • Research Support, Non-U.S. Gov't
  • Research Support, U.S. Gov't, Non-P.H.S.

Cite this

Robust variational segmentation of 3D bone CT data with thin cartilage interfaces. / Gangwar, Tarun; Calder, Jeffrey W; Takahashi, Takashi; Bechtold, Joan E; Schillinger, Dominik.

In: Medical Image Analysis, Vol. 47, 01.07.2018, p. 95-110.

Research output: Contribution to journalArticle

@article{05f80f4689e34c7181cdb5bcc065770a,
title = "Robust variational segmentation of 3D bone CT data with thin cartilage interfaces",
abstract = "We present a two-stage variational approach for segmenting 3D bone CT data that performs robustly with respect to thin cartilage interfaces. In the first stage, we minimize a flux-augmented Chan–Vese model that accurately segments well-separated regions. In the second stage, we apply a new phase-field fracture inspired model that reliably eliminates spurious bridges across thin cartilage interfaces, resulting in an accurate segmentation topology, from which each bone object can be identified. Its mathematical formulation is based on the phase-field approach to variational fracture, which naturally blends with the variational approach to segmentation. We successfully test and validate our methodology for the segmentation of 3D femur and vertebra bones, which feature thin cartilage regions in the hip joint, the intervertebral disks, and synovial joints of the spinous processes. The major strength of the new methodology is its potential for full automation and seamless integration with downstream predictive bone simulation in a common finite element framework.",
keywords = "3D bone CT data, Femur extraction, Flux-augmented Chan–Vese model, Phase-field fracture mechanics, Thin cartilage interfaces, Variational segmentation, Vertebra extraction, Voxel finite elements",
author = "Tarun Gangwar and Calder, {Jeffrey W} and Takashi Takahashi and Bechtold, {Joan E} and Dominik Schillinger",
year = "2018",
month = "7",
day = "1",
doi = "10.1016/j.media.2018.04.003",
language = "English (US)",
volume = "47",
pages = "95--110",
journal = "Medical Image Analysis",
issn = "1361-8415",
publisher = "Elsevier",

}

TY - JOUR

T1 - Robust variational segmentation of 3D bone CT data with thin cartilage interfaces

AU - Gangwar, Tarun

AU - Calder, Jeffrey W

AU - Takahashi, Takashi

AU - Bechtold, Joan E

AU - Schillinger, Dominik

PY - 2018/7/1

Y1 - 2018/7/1

N2 - We present a two-stage variational approach for segmenting 3D bone CT data that performs robustly with respect to thin cartilage interfaces. In the first stage, we minimize a flux-augmented Chan–Vese model that accurately segments well-separated regions. In the second stage, we apply a new phase-field fracture inspired model that reliably eliminates spurious bridges across thin cartilage interfaces, resulting in an accurate segmentation topology, from which each bone object can be identified. Its mathematical formulation is based on the phase-field approach to variational fracture, which naturally blends with the variational approach to segmentation. We successfully test and validate our methodology for the segmentation of 3D femur and vertebra bones, which feature thin cartilage regions in the hip joint, the intervertebral disks, and synovial joints of the spinous processes. The major strength of the new methodology is its potential for full automation and seamless integration with downstream predictive bone simulation in a common finite element framework.

AB - We present a two-stage variational approach for segmenting 3D bone CT data that performs robustly with respect to thin cartilage interfaces. In the first stage, we minimize a flux-augmented Chan–Vese model that accurately segments well-separated regions. In the second stage, we apply a new phase-field fracture inspired model that reliably eliminates spurious bridges across thin cartilage interfaces, resulting in an accurate segmentation topology, from which each bone object can be identified. Its mathematical formulation is based on the phase-field approach to variational fracture, which naturally blends with the variational approach to segmentation. We successfully test and validate our methodology for the segmentation of 3D femur and vertebra bones, which feature thin cartilage regions in the hip joint, the intervertebral disks, and synovial joints of the spinous processes. The major strength of the new methodology is its potential for full automation and seamless integration with downstream predictive bone simulation in a common finite element framework.

KW - 3D bone CT data

KW - Femur extraction

KW - Flux-augmented Chan–Vese model

KW - Phase-field fracture mechanics

KW - Thin cartilage interfaces

KW - Variational segmentation

KW - Vertebra extraction

KW - Voxel finite elements

UR - http://www.scopus.com/inward/record.url?scp=85043985955&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85043985955&partnerID=8YFLogxK

U2 - 10.1016/j.media.2018.04.003

DO - 10.1016/j.media.2018.04.003

M3 - Article

C2 - 29702415

AN - SCOPUS:85043985955

VL - 47

SP - 95

EP - 110

JO - Medical Image Analysis

JF - Medical Image Analysis

SN - 1361-8415

ER -