Learning coupled prior shape and appearance models for segmentation

Xiaolei Huang, Zhiguo Li, Dimitris Metaxas

Research output: Contribution to journalConference article

21 Citations (Scopus)

Abstract

We present a novel framework for learning a joint shape and appearance model from a large set of un-labelled training examples in arbitrary positions and orientations. The shape and intensity spaces are unified by implicitly representing shapes as "images" in the space of distance transforms. A stochastic chord-based matching algorithm is developed to align photo-realistic training examples under a common reference frame. Then dense local deformation fields, represented using the cubic B-spline based Free Form Deformations (FFD), are recovered to register the training examples in both shape and intensity spaces. Principal Component Analysis (PCA) is applied on the FFD control lattices to capture the variations in shape as well as on registered object interior textures. We show examples where we have built coupled shape and appearance prior models for the left ventricle and whole heart in short-axis cardiac tagged MR images, and used them to delineate the heart chambers in noisy, cluttered images. We also show quantitative validation on the automatic segmentation results by comparing to expert solutions.

Original languageEnglish (US)
Pages (from-to)60-69
Number of pages10
JournalLecture Notes in Computer Science
Volume3216
Issue numberPART 1
StatePublished - Dec 1 2004
EventMedical Image Computing and Computer-Assisted Intervention, MICCAI 2004 - 7th International Conference, Proceedings - Saint-Malo, France
Duration: Sep 26 2004Sep 29 2004

Fingerprint

Segmentation
Free-form Deformation
Splines
Principal component analysis
Textures
Model
Cubic B-spline
Distance Transform
Left Ventricle
Chord or secant line
Matching Algorithm
Large Set
Cardiac
Principal Component Analysis
Learning
Texture
Interior
Arbitrary
Training
Heart

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Huang, Xiaolei ; Li, Zhiguo ; Metaxas, Dimitris. / Learning coupled prior shape and appearance models for segmentation. In: Lecture Notes in Computer Science. 2004 ; Vol. 3216, No. PART 1. pp. 60-69.
@article{32c3ad3c59f34a63a9a6faf664bf9474,
title = "Learning coupled prior shape and appearance models for segmentation",
abstract = "We present a novel framework for learning a joint shape and appearance model from a large set of un-labelled training examples in arbitrary positions and orientations. The shape and intensity spaces are unified by implicitly representing shapes as {"}images{"} in the space of distance transforms. A stochastic chord-based matching algorithm is developed to align photo-realistic training examples under a common reference frame. Then dense local deformation fields, represented using the cubic B-spline based Free Form Deformations (FFD), are recovered to register the training examples in both shape and intensity spaces. Principal Component Analysis (PCA) is applied on the FFD control lattices to capture the variations in shape as well as on registered object interior textures. We show examples where we have built coupled shape and appearance prior models for the left ventricle and whole heart in short-axis cardiac tagged MR images, and used them to delineate the heart chambers in noisy, cluttered images. We also show quantitative validation on the automatic segmentation results by comparing to expert solutions.",
author = "Xiaolei Huang and Zhiguo Li and Dimitris Metaxas",
year = "2004",
month = "12",
day = "1",
language = "English (US)",
volume = "3216",
pages = "60--69",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer Verlag",
number = "PART 1",

}

Learning coupled prior shape and appearance models for segmentation. / Huang, Xiaolei; Li, Zhiguo; Metaxas, Dimitris.

In: Lecture Notes in Computer Science, Vol. 3216, No. PART 1, 01.12.2004, p. 60-69.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Learning coupled prior shape and appearance models for segmentation

AU - Huang, Xiaolei

AU - Li, Zhiguo

AU - Metaxas, Dimitris

PY - 2004/12/1

Y1 - 2004/12/1

N2 - We present a novel framework for learning a joint shape and appearance model from a large set of un-labelled training examples in arbitrary positions and orientations. The shape and intensity spaces are unified by implicitly representing shapes as "images" in the space of distance transforms. A stochastic chord-based matching algorithm is developed to align photo-realistic training examples under a common reference frame. Then dense local deformation fields, represented using the cubic B-spline based Free Form Deformations (FFD), are recovered to register the training examples in both shape and intensity spaces. Principal Component Analysis (PCA) is applied on the FFD control lattices to capture the variations in shape as well as on registered object interior textures. We show examples where we have built coupled shape and appearance prior models for the left ventricle and whole heart in short-axis cardiac tagged MR images, and used them to delineate the heart chambers in noisy, cluttered images. We also show quantitative validation on the automatic segmentation results by comparing to expert solutions.

AB - We present a novel framework for learning a joint shape and appearance model from a large set of un-labelled training examples in arbitrary positions and orientations. The shape and intensity spaces are unified by implicitly representing shapes as "images" in the space of distance transforms. A stochastic chord-based matching algorithm is developed to align photo-realistic training examples under a common reference frame. Then dense local deformation fields, represented using the cubic B-spline based Free Form Deformations (FFD), are recovered to register the training examples in both shape and intensity spaces. Principal Component Analysis (PCA) is applied on the FFD control lattices to capture the variations in shape as well as on registered object interior textures. We show examples where we have built coupled shape and appearance prior models for the left ventricle and whole heart in short-axis cardiac tagged MR images, and used them to delineate the heart chambers in noisy, cluttered images. We also show quantitative validation on the automatic segmentation results by comparing to expert solutions.

UR - http://www.scopus.com/inward/record.url?scp=20344395345&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=20344395345&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:20344395345

VL - 3216

SP - 60

EP - 69

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

IS - PART 1

ER -