Estimation of shape model parameters for 3D surfaces

Søren Gylling Hemmingsen Erbou; Sune Darkner; Jurgen Fripp; Sébastien Ourselin; Bjarne Kjær Ersbøll

doi:10.1109/ISBI.2008.4541073

Estimation of shape model parameters for 3D surfaces

Søren Gylling Hemmingsen Erbou, Sune Darkner, Jurgen Fripp, Sébastien Ourselin, Bjarne Kjær Ersbøll

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

310 Downloads (Pure)

Abstract

Statistical shape models are widely used as a compact way of representing shape variation. Fitting a shape model to unseen data enables characterizing the data in terms of the model parameters. In this paper a Gauss-Newton optimization scheme is proposed to estimate shape model parameters of 3D surfaces using distance maps, which enables the estimation of model parameters without the requirement of point correspondence. For applications with acquisition limitations such as speed and cost, this formulation enables the fitting of a statistical shape model to arbitrarily sampled data. The method is applied to a database of 3D surfaces from a section of the porcine pelvic bone extracted from 33 CT scans. A leave-one-out validation shows that the parameters of the first 3 modes of the shape model can be predicted with a mean difference within [-0.01,0.02] from the true mean, with a standard deviation less than 0.34.

Original language	English
Title of host publication	5th IEEE International Symposium on Biomedical Imaging
Publisher	IEEE
Publication date	2008
Pages	624-627
ISBN (Print)	14-24-42002-4
DOIs	https://doi.org/10.1109/ISBI.2008.4541073
Publication status	Published - 2008
Event	2008 IEEE International Symposium on Biomedical Imaging: From Nano to Macro - Paris, France Duration: 14 May 2008 → 17 May 2008 Conference number: 5 http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=4534844

Conference

Conference	2008 IEEE International Symposium on Biomedical Imaging
Number	5
Country/Territory	France
City	Paris
Period	14/05/2008 → 17/05/2008
Internet address	http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=4534844

Bibliographical note

Copyright: 2008 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE

Keywords

Optimization methods
X-ray tomography
Image shape analysis
Image registration
Biomedical image processing

Access to Document

10.1109/ISBI.2008.4541073

ErbouFinal published version, 303 KB

OpenUrl availability

Full text

Cite this

@inproceedings{5b9ca9a58aea4b56923dbe7bfd7da3db,

title = "Estimation of shape model parameters for 3D surfaces",

abstract = "Statistical shape models are widely used as a compact way of representing shape variation. Fitting a shape model to unseen data enables characterizing the data in terms of the model parameters. In this paper a Gauss-Newton optimization scheme is proposed to estimate shape model parameters of 3D surfaces using distance maps, which enables the estimation of model parameters without the requirement of point correspondence. For applications with acquisition limitations such as speed and cost, this formulation enables the fitting of a statistical shape model to arbitrarily sampled data. The method is applied to a database of 3D surfaces from a section of the porcine pelvic bone extracted from 33 CT scans. A leave-one-out validation shows that the parameters of the first 3 modes of the shape model can be predicted with a mean difference within [-0.01,0.02] from the true mean, with a standard deviation less than 0.34.",

keywords = "Optimization methods, X-ray tomography, Image shape analysis, Image registration, Biomedical image processing",

author = "Erbou, {S{\o}ren Gylling Hemmingsen} and Sune Darkner and Jurgen Fripp and S{\'e}bastien Ourselin and Ersb{\o}ll, {Bjarne Kj{\ae}r}",

note = "Copyright: 2008 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE; 2008 IEEE International Symposium on Biomedical Imaging : From Nano to Macro, ISBI 2008 ; Conference date: 14-05-2008 Through 17-05-2008",

year = "2008",

doi = "10.1109/ISBI.2008.4541073",

language = "English",

isbn = "14-24-42002-4",

pages = "624--627",

booktitle = "5th IEEE International Symposium on Biomedical Imaging",

publisher = "IEEE",

address = "United States",

url = "http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=4534844",

}

TY - GEN

T1 - Estimation of shape model parameters for 3D surfaces

AU - Erbou, Søren Gylling Hemmingsen

AU - Darkner, Sune

AU - Fripp, Jurgen

AU - Ourselin, Sébastien

AU - Ersbøll, Bjarne Kjær

N1 - Conference code: 5

PY - 2008

Y1 - 2008

N2 - Statistical shape models are widely used as a compact way of representing shape variation. Fitting a shape model to unseen data enables characterizing the data in terms of the model parameters. In this paper a Gauss-Newton optimization scheme is proposed to estimate shape model parameters of 3D surfaces using distance maps, which enables the estimation of model parameters without the requirement of point correspondence. For applications with acquisition limitations such as speed and cost, this formulation enables the fitting of a statistical shape model to arbitrarily sampled data. The method is applied to a database of 3D surfaces from a section of the porcine pelvic bone extracted from 33 CT scans. A leave-one-out validation shows that the parameters of the first 3 modes of the shape model can be predicted with a mean difference within [-0.01,0.02] from the true mean, with a standard deviation less than 0.34.

AB - Statistical shape models are widely used as a compact way of representing shape variation. Fitting a shape model to unseen data enables characterizing the data in terms of the model parameters. In this paper a Gauss-Newton optimization scheme is proposed to estimate shape model parameters of 3D surfaces using distance maps, which enables the estimation of model parameters without the requirement of point correspondence. For applications with acquisition limitations such as speed and cost, this formulation enables the fitting of a statistical shape model to arbitrarily sampled data. The method is applied to a database of 3D surfaces from a section of the porcine pelvic bone extracted from 33 CT scans. A leave-one-out validation shows that the parameters of the first 3 modes of the shape model can be predicted with a mean difference within [-0.01,0.02] from the true mean, with a standard deviation less than 0.34.

KW - Optimization methods

KW - X-ray tomography

KW - Image shape analysis

KW - Image registration

KW - Biomedical image processing

U2 - 10.1109/ISBI.2008.4541073

DO - 10.1109/ISBI.2008.4541073

M3 - Article in proceedings

SN - 14-24-42002-4

SP - 624

EP - 627

BT - 5th IEEE International Symposium on Biomedical Imaging

PB - IEEE

T2 - 2008 IEEE International Symposium on Biomedical Imaging

Y2 - 14 May 2008 through 17 May 2008

ER -

Estimation of shape model parameters for 3D surfaces

Abstract

Conference

Bibliographical note

Keywords

Access to Document

OpenUrl availability

Fingerprint

Cite this