Building dynamic population graph for accurate correspondence detection

Shaoyi Du, Yanrong Guo, Gerard Sanroma, Dong Ni, Guorong Wu, Dinggang Shen

Research output: Contribution to journalArticle

32 Citations (Scopus)

Abstract

In medical imaging studies, there is an increasing trend for discovering the intrinsic anatomical difference across individual subjects in a dataset, such as hand images for skeletal bone age estimation. Pair-wise matching is often used to detect correspondences between each individual subject and a pre-selected model image with manually-placed landmarks. However, the large anatomical variability across individual subjects can easily compromise such pair-wise matching step. In this paper, we present a new framework to simultaneously detect correspondences among a population of individual subjects, by propagating all manually-placed landmarks from a small set of model images through a dynamically constructed image graph. Specifically, we first establish graph links between models and individual subjects according to pair-wise shape similarity (called as forward step). Next, we detect correspondences for the individual subjects with direct links to any of model images, which is achieved by a new multi-model correspondence detection approach based on our recently-published sparse point matching method. To correct those inaccurate correspondences, we further apply an error detection mechanism to automatically detect wrong correspondences and then update the image graph accordingly (called as backward step). After that, all subject images with detected correspondences are included into the set of model images, and the above two steps of graph expansion and error correction are repeated until accurate correspondences for all subject images are established. Evaluations on real hand X-ray images demonstrate that our proposed method using a dynamic graph construction approach can achieve much higher accuracy and robustness, when compared with the state-of-the-art pair-wise correspondence detection methods as well as a similar method but using static population graph.

Original languageEnglish
Pages (from-to)256-267
Number of pages12
JournalMedical Image Analysis
Volume26
Issue number1
DOIs
Publication statusPublished - 2015 Dec 1

Fingerprint

Population dynamics
Population Dynamics
Hand
Diagnostic Imaging
Individuality
Population
Error detection
Medical imaging
X-Rays
Error correction
Bone and Bones
Bone
X rays

Keywords

  • Correspondence detection
  • Dynamic population graph
  • Multi-models
  • Pair-wise matching

ASJC Scopus subject areas

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

Cite this

Building dynamic population graph for accurate correspondence detection. / Du, Shaoyi; Guo, Yanrong; Sanroma, Gerard; Ni, Dong; Wu, Guorong; Shen, Dinggang.

In: Medical Image Analysis, Vol. 26, No. 1, 01.12.2015, p. 256-267.

Research output: Contribution to journalArticle

Du, Shaoyi ; Guo, Yanrong ; Sanroma, Gerard ; Ni, Dong ; Wu, Guorong ; Shen, Dinggang. / Building dynamic population graph for accurate correspondence detection. In: Medical Image Analysis. 2015 ; Vol. 26, No. 1. pp. 256-267.
@article{3fcf78a0cb024f8ab539589f55381a17,
title = "Building dynamic population graph for accurate correspondence detection",
abstract = "In medical imaging studies, there is an increasing trend for discovering the intrinsic anatomical difference across individual subjects in a dataset, such as hand images for skeletal bone age estimation. Pair-wise matching is often used to detect correspondences between each individual subject and a pre-selected model image with manually-placed landmarks. However, the large anatomical variability across individual subjects can easily compromise such pair-wise matching step. In this paper, we present a new framework to simultaneously detect correspondences among a population of individual subjects, by propagating all manually-placed landmarks from a small set of model images through a dynamically constructed image graph. Specifically, we first establish graph links between models and individual subjects according to pair-wise shape similarity (called as forward step). Next, we detect correspondences for the individual subjects with direct links to any of model images, which is achieved by a new multi-model correspondence detection approach based on our recently-published sparse point matching method. To correct those inaccurate correspondences, we further apply an error detection mechanism to automatically detect wrong correspondences and then update the image graph accordingly (called as backward step). After that, all subject images with detected correspondences are included into the set of model images, and the above two steps of graph expansion and error correction are repeated until accurate correspondences for all subject images are established. Evaluations on real hand X-ray images demonstrate that our proposed method using a dynamic graph construction approach can achieve much higher accuracy and robustness, when compared with the state-of-the-art pair-wise correspondence detection methods as well as a similar method but using static population graph.",
keywords = "Correspondence detection, Dynamic population graph, Multi-models, Pair-wise matching",
author = "Shaoyi Du and Yanrong Guo and Gerard Sanroma and Dong Ni and Guorong Wu and Dinggang Shen",
year = "2015",
month = "12",
day = "1",
doi = "10.1016/j.media.2015.10.001",
language = "English",
volume = "26",
pages = "256--267",
journal = "Medical Image Analysis",
issn = "1361-8415",
publisher = "Elsevier",
number = "1",

}

TY - JOUR

T1 - Building dynamic population graph for accurate correspondence detection

AU - Du, Shaoyi

AU - Guo, Yanrong

AU - Sanroma, Gerard

AU - Ni, Dong

AU - Wu, Guorong

AU - Shen, Dinggang

PY - 2015/12/1

Y1 - 2015/12/1

N2 - In medical imaging studies, there is an increasing trend for discovering the intrinsic anatomical difference across individual subjects in a dataset, such as hand images for skeletal bone age estimation. Pair-wise matching is often used to detect correspondences between each individual subject and a pre-selected model image with manually-placed landmarks. However, the large anatomical variability across individual subjects can easily compromise such pair-wise matching step. In this paper, we present a new framework to simultaneously detect correspondences among a population of individual subjects, by propagating all manually-placed landmarks from a small set of model images through a dynamically constructed image graph. Specifically, we first establish graph links between models and individual subjects according to pair-wise shape similarity (called as forward step). Next, we detect correspondences for the individual subjects with direct links to any of model images, which is achieved by a new multi-model correspondence detection approach based on our recently-published sparse point matching method. To correct those inaccurate correspondences, we further apply an error detection mechanism to automatically detect wrong correspondences and then update the image graph accordingly (called as backward step). After that, all subject images with detected correspondences are included into the set of model images, and the above two steps of graph expansion and error correction are repeated until accurate correspondences for all subject images are established. Evaluations on real hand X-ray images demonstrate that our proposed method using a dynamic graph construction approach can achieve much higher accuracy and robustness, when compared with the state-of-the-art pair-wise correspondence detection methods as well as a similar method but using static population graph.

AB - In medical imaging studies, there is an increasing trend for discovering the intrinsic anatomical difference across individual subjects in a dataset, such as hand images for skeletal bone age estimation. Pair-wise matching is often used to detect correspondences between each individual subject and a pre-selected model image with manually-placed landmarks. However, the large anatomical variability across individual subjects can easily compromise such pair-wise matching step. In this paper, we present a new framework to simultaneously detect correspondences among a population of individual subjects, by propagating all manually-placed landmarks from a small set of model images through a dynamically constructed image graph. Specifically, we first establish graph links between models and individual subjects according to pair-wise shape similarity (called as forward step). Next, we detect correspondences for the individual subjects with direct links to any of model images, which is achieved by a new multi-model correspondence detection approach based on our recently-published sparse point matching method. To correct those inaccurate correspondences, we further apply an error detection mechanism to automatically detect wrong correspondences and then update the image graph accordingly (called as backward step). After that, all subject images with detected correspondences are included into the set of model images, and the above two steps of graph expansion and error correction are repeated until accurate correspondences for all subject images are established. Evaluations on real hand X-ray images demonstrate that our proposed method using a dynamic graph construction approach can achieve much higher accuracy and robustness, when compared with the state-of-the-art pair-wise correspondence detection methods as well as a similar method but using static population graph.

KW - Correspondence detection

KW - Dynamic population graph

KW - Multi-models

KW - Pair-wise matching

UR - http://www.scopus.com/inward/record.url?scp=84945395526&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84945395526&partnerID=8YFLogxK

U2 - 10.1016/j.media.2015.10.001

DO - 10.1016/j.media.2015.10.001

M3 - Article

C2 - 26519794

AN - SCOPUS:84945395526

VL - 26

SP - 256

EP - 267

JO - Medical Image Analysis

JF - Medical Image Analysis

SN - 1361-8415

IS - 1

ER -