Integrating geometric configuration and appearance information into a unified framework for anatomical landmark localization

Martin Urschler, Thomas Ebner, Darko Štern

Research output: Contribution to journalArticle

Abstract

In approaches for automatic localization of multiple anatomical landmarks, disambiguation of locally similar structures as obtained by locally accurate candidate generation is often performed by solely including high level knowledge about geometric landmark configuration. In our novel localization approach, we propose to combine both image appearance information and geometric landmark configuration into a unified random forest framework integrated into an optimization procedure that iteratively refines joint landmark predictions by using the coordinate descent algorithm. Depending on how strong multiple landmarks are correlated in a specific localization task, this integration has the benefit that it remains flexible in deciding whether appearance information or the geometric configuration of multiple landmarks is the stronger cue for solving a localization problem both accurately and robustly. Furthermore, no preliminary choice on how to encode a graphical model describing landmark configuration has to be made. In an extensive evaluation on five challenging datasets involving different 2D and 3D imaging modalities, we show that our proposed method is widely applicable and delivers state-of-the-art results when compared to various other related methods.

LanguageEnglish
Pages23-36
Number of pages14
JournalMedical image analysis
Volume43
DOIs
StatusPublished - 1 Jan 2018

Fingerprint

Imaging techniques
Cues
Joints
Forests
Datasets

Keywords

  • Anatomical landmarks
  • Coordinate descent
  • Localization
  • Random regression forest

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging
  • Computer Vision and Pattern Recognition
  • Health Informatics
  • Computer Graphics and Computer-Aided Design

Fields of Expertise

  • Information, Communication & Computing

Cooperations

  • BioTechMed-Graz

Cite this

Integrating geometric configuration and appearance information into a unified framework for anatomical landmark localization. / Urschler, Martin; Ebner, Thomas; Štern, Darko.

In: Medical image analysis, Vol. 43, 01.01.2018, p. 23-36.

Research output: Contribution to journalArticle

@article{61001f0846504ed292f9dde679dc576f,
title = "Integrating geometric configuration and appearance information into a unified framework for anatomical landmark localization",
abstract = "In approaches for automatic localization of multiple anatomical landmarks, disambiguation of locally similar structures as obtained by locally accurate candidate generation is often performed by solely including high level knowledge about geometric landmark configuration. In our novel localization approach, we propose to combine both image appearance information and geometric landmark configuration into a unified random forest framework integrated into an optimization procedure that iteratively refines joint landmark predictions by using the coordinate descent algorithm. Depending on how strong multiple landmarks are correlated in a specific localization task, this integration has the benefit that it remains flexible in deciding whether appearance information or the geometric configuration of multiple landmarks is the stronger cue for solving a localization problem both accurately and robustly. Furthermore, no preliminary choice on how to encode a graphical model describing landmark configuration has to be made. In an extensive evaluation on five challenging datasets involving different 2D and 3D imaging modalities, we show that our proposed method is widely applicable and delivers state-of-the-art results when compared to various other related methods.",
keywords = "Anatomical landmarks, Coordinate descent, Localization, Random regression forest",
author = "Martin Urschler and Thomas Ebner and Darko Štern",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.media.2017.09.003",
language = "English",
volume = "43",
pages = "23--36",
journal = "Medical image analysis",
issn = "1361-8415",
publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Integrating geometric configuration and appearance information into a unified framework for anatomical landmark localization

AU - Urschler,Martin

AU - Ebner,Thomas

AU - Štern,Darko

PY - 2018/1/1

Y1 - 2018/1/1

N2 - In approaches for automatic localization of multiple anatomical landmarks, disambiguation of locally similar structures as obtained by locally accurate candidate generation is often performed by solely including high level knowledge about geometric landmark configuration. In our novel localization approach, we propose to combine both image appearance information and geometric landmark configuration into a unified random forest framework integrated into an optimization procedure that iteratively refines joint landmark predictions by using the coordinate descent algorithm. Depending on how strong multiple landmarks are correlated in a specific localization task, this integration has the benefit that it remains flexible in deciding whether appearance information or the geometric configuration of multiple landmarks is the stronger cue for solving a localization problem both accurately and robustly. Furthermore, no preliminary choice on how to encode a graphical model describing landmark configuration has to be made. In an extensive evaluation on five challenging datasets involving different 2D and 3D imaging modalities, we show that our proposed method is widely applicable and delivers state-of-the-art results when compared to various other related methods.

AB - In approaches for automatic localization of multiple anatomical landmarks, disambiguation of locally similar structures as obtained by locally accurate candidate generation is often performed by solely including high level knowledge about geometric landmark configuration. In our novel localization approach, we propose to combine both image appearance information and geometric landmark configuration into a unified random forest framework integrated into an optimization procedure that iteratively refines joint landmark predictions by using the coordinate descent algorithm. Depending on how strong multiple landmarks are correlated in a specific localization task, this integration has the benefit that it remains flexible in deciding whether appearance information or the geometric configuration of multiple landmarks is the stronger cue for solving a localization problem both accurately and robustly. Furthermore, no preliminary choice on how to encode a graphical model describing landmark configuration has to be made. In an extensive evaluation on five challenging datasets involving different 2D and 3D imaging modalities, we show that our proposed method is widely applicable and delivers state-of-the-art results when compared to various other related methods.

KW - Anatomical landmarks

KW - Coordinate descent

KW - Localization

KW - Random regression forest

UR - http://www.scopus.com/inward/record.url?scp=85029839917&partnerID=8YFLogxK

U2 - 10.1016/j.media.2017.09.003

DO - 10.1016/j.media.2017.09.003

M3 - Article

VL - 43

SP - 23

EP - 36

JO - Medical image analysis

T2 - Medical image analysis

JF - Medical image analysis

SN - 1361-8415

ER -