Towards Automatic Bone Age Estimation From MRI: Localization of 3D Anatomical Landmarks: localization of 3D anatomical landmarks

Thomas Ebner, Darko Stern, Rene Donner, Horst Bischof, Martin Urschler

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

Bone age estimation (BAE) is an important procedure in forensic practice which recently has seen a shift in attention from X-ray to MRI based imaging. To automate BAE from MRI, localization of the joints between hand bones is a crucial first step, which is challenging due to anatomical variations, different poses and repeating structures within the hand. We propose a landmark localization algorithm using multiple random regression forests, first analyzing the shape of the hand from information of the whole image, thus implicitly modeling the global landmark configuration, followed by a refinement based on more local information to increase prediction accuracy. We are able to clearly outperform related approaches on our dataset of 60 T1-weighted MR images, achieving a mean landmark localization error of 1.4 ± 1.5mm, while having only 0.25% outliers with an error greater than 10mm.

Original languageEnglish
Title of host publicationMedical Image Computing and Computer-Assisted Intervention – MICCAI 2014
Subtitle of host publication17th International Conference, Boston, MA, USA, September 14-18, 2014, Proceedings, Part II
EditorsPolina Golland, Nobuhiko Hata, Christian Barillot, Joachim Hornegger, Robert Howe
PublisherSpringer International Publishing AG
Pages421-428
Number of pages8
Volume8674
ISBN (Electronic)978-3-319-10470-6
ISBN (Print)978-3-319-10469-0
DOIs
Publication statusPublished - 2014
EventMICCAI 2014 : International Conference on Medical Image Computing and Computer-Assisted Intervention - Cambridge, United States
Duration: 14 Sep 201418 Oct 2014

Publication series

NameLecture Notes in Computer Science
Volume8674

Conference

ConferenceMICCAI 2014
CountryUnited States
CityCambridge
Period14/09/1418/10/14

Fingerprint

Hand
Hand Bones
Bone and Bones
Joints
X-Rays
Forests
Datasets

Fields of Expertise

  • Information, Communication & Computing

Cooperations

  • BioTechMed-Graz

Cite this

Ebner, T., Stern, D., Donner, R., Bischof, H., & Urschler, M. (2014). Towards Automatic Bone Age Estimation From MRI: Localization of 3D Anatomical Landmarks: localization of 3D anatomical landmarks. In P. Golland, N. Hata, C. Barillot, J. Hornegger, & R. Howe (Eds.), Medical Image Computing and Computer-Assisted Intervention – MICCAI 2014: 17th International Conference, Boston, MA, USA, September 14-18, 2014, Proceedings, Part II (Vol. 8674, pp. 421-428). (Lecture Notes in Computer Science; Vol. 8674). Springer International Publishing AG . https://doi.org/10.1007/978-3-319-10470-6_53, https://doi.org/10.1007/978-3-319-10470-6_53

Towards Automatic Bone Age Estimation From MRI: Localization of 3D Anatomical Landmarks : localization of 3D anatomical landmarks. / Ebner, Thomas; Stern, Darko; Donner, Rene; Bischof, Horst; Urschler, Martin.

Medical Image Computing and Computer-Assisted Intervention – MICCAI 2014: 17th International Conference, Boston, MA, USA, September 14-18, 2014, Proceedings, Part II. ed. / Polina Golland; Nobuhiko Hata; Christian Barillot; Joachim Hornegger; Robert Howe. Vol. 8674 Springer International Publishing AG , 2014. p. 421-428 (Lecture Notes in Computer Science; Vol. 8674).

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Ebner, T, Stern, D, Donner, R, Bischof, H & Urschler, M 2014, Towards Automatic Bone Age Estimation From MRI: Localization of 3D Anatomical Landmarks: localization of 3D anatomical landmarks. in P Golland, N Hata, C Barillot, J Hornegger & R Howe (eds), Medical Image Computing and Computer-Assisted Intervention – MICCAI 2014: 17th International Conference, Boston, MA, USA, September 14-18, 2014, Proceedings, Part II. vol. 8674, Lecture Notes in Computer Science, vol. 8674, Springer International Publishing AG , pp. 421-428, MICCAI 2014 , Cambridge, United States, 14/09/14. https://doi.org/10.1007/978-3-319-10470-6_53, https://doi.org/10.1007/978-3-319-10470-6_53
Ebner T, Stern D, Donner R, Bischof H, Urschler M. Towards Automatic Bone Age Estimation From MRI: Localization of 3D Anatomical Landmarks: localization of 3D anatomical landmarks. In Golland P, Hata N, Barillot C, Hornegger J, Howe R, editors, Medical Image Computing and Computer-Assisted Intervention – MICCAI 2014: 17th International Conference, Boston, MA, USA, September 14-18, 2014, Proceedings, Part II. Vol. 8674. Springer International Publishing AG . 2014. p. 421-428. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-319-10470-6_53, https://doi.org/10.1007/978-3-319-10470-6_53
Ebner, Thomas ; Stern, Darko ; Donner, Rene ; Bischof, Horst ; Urschler, Martin. / Towards Automatic Bone Age Estimation From MRI: Localization of 3D Anatomical Landmarks : localization of 3D anatomical landmarks. Medical Image Computing and Computer-Assisted Intervention – MICCAI 2014: 17th International Conference, Boston, MA, USA, September 14-18, 2014, Proceedings, Part II. editor / Polina Golland ; Nobuhiko Hata ; Christian Barillot ; Joachim Hornegger ; Robert Howe. Vol. 8674 Springer International Publishing AG , 2014. pp. 421-428 (Lecture Notes in Computer Science).
@inproceedings{8bdf152fe9814fa994dd9eae117b6522,
title = "Towards Automatic Bone Age Estimation From MRI: Localization of 3D Anatomical Landmarks: localization of 3D anatomical landmarks",
abstract = "Bone age estimation (BAE) is an important procedure in forensic practice which recently has seen a shift in attention from X-ray to MRI based imaging. To automate BAE from MRI, localization of the joints between hand bones is a crucial first step, which is challenging due to anatomical variations, different poses and repeating structures within the hand. We propose a landmark localization algorithm using multiple random regression forests, first analyzing the shape of the hand from information of the whole image, thus implicitly modeling the global landmark configuration, followed by a refinement based on more local information to increase prediction accuracy. We are able to clearly outperform related approaches on our dataset of 60 T1-weighted MR images, achieving a mean landmark localization error of 1.4 ± 1.5mm, while having only 0.25{\%} outliers with an error greater than 10mm.",
author = "Thomas Ebner and Darko Stern and Rene Donner and Horst Bischof and Martin Urschler",
year = "2014",
doi = "10.1007/978-3-319-10470-6_53",
language = "English",
isbn = "978-3-319-10469-0",
volume = "8674",
series = "Lecture Notes in Computer Science",
publisher = "Springer International Publishing AG",
pages = "421--428",
editor = "Polina Golland and Nobuhiko Hata and Christian Barillot and Joachim Hornegger and Robert Howe",
booktitle = "Medical Image Computing and Computer-Assisted Intervention – MICCAI 2014",
address = "Switzerland",

}

TY - GEN

T1 - Towards Automatic Bone Age Estimation From MRI: Localization of 3D Anatomical Landmarks

T2 - localization of 3D anatomical landmarks

AU - Ebner, Thomas

AU - Stern, Darko

AU - Donner, Rene

AU - Bischof, Horst

AU - Urschler, Martin

PY - 2014

Y1 - 2014

N2 - Bone age estimation (BAE) is an important procedure in forensic practice which recently has seen a shift in attention from X-ray to MRI based imaging. To automate BAE from MRI, localization of the joints between hand bones is a crucial first step, which is challenging due to anatomical variations, different poses and repeating structures within the hand. We propose a landmark localization algorithm using multiple random regression forests, first analyzing the shape of the hand from information of the whole image, thus implicitly modeling the global landmark configuration, followed by a refinement based on more local information to increase prediction accuracy. We are able to clearly outperform related approaches on our dataset of 60 T1-weighted MR images, achieving a mean landmark localization error of 1.4 ± 1.5mm, while having only 0.25% outliers with an error greater than 10mm.

AB - Bone age estimation (BAE) is an important procedure in forensic practice which recently has seen a shift in attention from X-ray to MRI based imaging. To automate BAE from MRI, localization of the joints between hand bones is a crucial first step, which is challenging due to anatomical variations, different poses and repeating structures within the hand. We propose a landmark localization algorithm using multiple random regression forests, first analyzing the shape of the hand from information of the whole image, thus implicitly modeling the global landmark configuration, followed by a refinement based on more local information to increase prediction accuracy. We are able to clearly outperform related approaches on our dataset of 60 T1-weighted MR images, achieving a mean landmark localization error of 1.4 ± 1.5mm, while having only 0.25% outliers with an error greater than 10mm.

UR - http://www.scopus.com/inward/record.url?scp=84922328575&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-10470-6_53

DO - 10.1007/978-3-319-10470-6_53

M3 - Conference contribution

SN - 978-3-319-10469-0

VL - 8674

T3 - Lecture Notes in Computer Science

SP - 421

EP - 428

BT - Medical Image Computing and Computer-Assisted Intervention – MICCAI 2014

A2 - Golland, Polina

A2 - Hata, Nobuhiko

A2 - Barillot, Christian

A2 - Hornegger, Joachim

A2 - Howe, Robert

PB - Springer International Publishing AG

ER -