Nonrigid registration with corresponding points constraint for automatic segmentation of cardiac DSCT images
Abstract Background Dual-source computed tomography (DSCT) is a very effective way for diagnosis and treatment of heart disease. The quantitative information of spatiotemporal DSCT images can be important for the evaluation of cardiac function. To avoid the shortcoming of manual delineation, it is i...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
BMC
2017-03-01
|
Series: | BioMedical Engineering OnLine |
Subjects: | |
Online Access: | http://link.springer.com/article/10.1186/s12938-017-0323-1 |
id |
doaj-912234957eaa4fe08ca8295139bf7ad2 |
---|---|
record_format |
Article |
spelling |
doaj-912234957eaa4fe08ca8295139bf7ad22020-11-24T22:13:44ZengBMCBioMedical Engineering OnLine1475-925X2017-03-0116111510.1186/s12938-017-0323-1Nonrigid registration with corresponding points constraint for automatic segmentation of cardiac DSCT imagesXuesong Lu0Rongqian Yang1Qinlan Xie2Shanxing Ou3Yunfei Zha4Defeng Wang5College of Biomedical Engineering, South-Central University for NationalitiesSchool of Materials Science and Engineering, South China University of TechnologyCollege of Biomedical Engineering, South-Central University for NationalitiesRadiology Department, Guangzhou General Hospital of Guangzhou Military Area CommandDepartment of Radiology, Remin Hospital of Wuhan UniversityResearch Center for Medical Image Computing, Department of Imaging and Interventional Radiology, The Chinese University of Hong KongAbstract Background Dual-source computed tomography (DSCT) is a very effective way for diagnosis and treatment of heart disease. The quantitative information of spatiotemporal DSCT images can be important for the evaluation of cardiac function. To avoid the shortcoming of manual delineation, it is imperative to develop an automatic segmentation technique for 4D cardiac images. Methods In this paper, we implement the heart segmentation-propagation framework based on nonrigid registration. The corresponding points of anatomical substructures are extracted by using the extension of n-dimensional scale invariant feature transform method. They are considered as a constraint term of nonrigid registration using the free-form deformation, in order to restrain the large variations and boundary ambiguity between subjects. Results We validate our method on 15 patients at ten time phases. Atlases are constructed by the training dataset from ten patients. On the remaining data the median overlap is shown to improve significantly compared to original mutual information, in particular from 0.4703 to 0.5015 ( $$ p = 5.0 \times 10^{ - 4} $$ p = 5.0 × 10 - 4 ) for left ventricle myocardium and from 0.6307 to 0.6519 ( $$ p = 6.0 \times 10^{ - 4} $$ p = 6.0 × 10 - 4 ) for right atrium. Conclusions The proposed method outperforms standard mutual information of intensity only. The segmentation errors had been significantly reduced at the left ventricle myocardium and the right atrium. The mean surface distance of using our framework is around 1.73 mm for the whole heart.http://link.springer.com/article/10.1186/s12938-017-0323-1Dual-source computed tomography (DSCT)Nonrigid registrationMutual informationCorresponding pointsn-dimensional scale invariant feature transform (n-SIFT)Automatic heart segmentation |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Xuesong Lu Rongqian Yang Qinlan Xie Shanxing Ou Yunfei Zha Defeng Wang |
spellingShingle |
Xuesong Lu Rongqian Yang Qinlan Xie Shanxing Ou Yunfei Zha Defeng Wang Nonrigid registration with corresponding points constraint for automatic segmentation of cardiac DSCT images BioMedical Engineering OnLine Dual-source computed tomography (DSCT) Nonrigid registration Mutual information Corresponding points n-dimensional scale invariant feature transform (n-SIFT) Automatic heart segmentation |
author_facet |
Xuesong Lu Rongqian Yang Qinlan Xie Shanxing Ou Yunfei Zha Defeng Wang |
author_sort |
Xuesong Lu |
title |
Nonrigid registration with corresponding points constraint for automatic segmentation of cardiac DSCT images |
title_short |
Nonrigid registration with corresponding points constraint for automatic segmentation of cardiac DSCT images |
title_full |
Nonrigid registration with corresponding points constraint for automatic segmentation of cardiac DSCT images |
title_fullStr |
Nonrigid registration with corresponding points constraint for automatic segmentation of cardiac DSCT images |
title_full_unstemmed |
Nonrigid registration with corresponding points constraint for automatic segmentation of cardiac DSCT images |
title_sort |
nonrigid registration with corresponding points constraint for automatic segmentation of cardiac dsct images |
publisher |
BMC |
series |
BioMedical Engineering OnLine |
issn |
1475-925X |
publishDate |
2017-03-01 |
description |
Abstract Background Dual-source computed tomography (DSCT) is a very effective way for diagnosis and treatment of heart disease. The quantitative information of spatiotemporal DSCT images can be important for the evaluation of cardiac function. To avoid the shortcoming of manual delineation, it is imperative to develop an automatic segmentation technique for 4D cardiac images. Methods In this paper, we implement the heart segmentation-propagation framework based on nonrigid registration. The corresponding points of anatomical substructures are extracted by using the extension of n-dimensional scale invariant feature transform method. They are considered as a constraint term of nonrigid registration using the free-form deformation, in order to restrain the large variations and boundary ambiguity between subjects. Results We validate our method on 15 patients at ten time phases. Atlases are constructed by the training dataset from ten patients. On the remaining data the median overlap is shown to improve significantly compared to original mutual information, in particular from 0.4703 to 0.5015 ( $$ p = 5.0 \times 10^{ - 4} $$ p = 5.0 × 10 - 4 ) for left ventricle myocardium and from 0.6307 to 0.6519 ( $$ p = 6.0 \times 10^{ - 4} $$ p = 6.0 × 10 - 4 ) for right atrium. Conclusions The proposed method outperforms standard mutual information of intensity only. The segmentation errors had been significantly reduced at the left ventricle myocardium and the right atrium. The mean surface distance of using our framework is around 1.73 mm for the whole heart. |
topic |
Dual-source computed tomography (DSCT) Nonrigid registration Mutual information Corresponding points n-dimensional scale invariant feature transform (n-SIFT) Automatic heart segmentation |
url |
http://link.springer.com/article/10.1186/s12938-017-0323-1 |
work_keys_str_mv |
AT xuesonglu nonrigidregistrationwithcorrespondingpointsconstraintforautomaticsegmentationofcardiacdsctimages AT rongqianyang nonrigidregistrationwithcorrespondingpointsconstraintforautomaticsegmentationofcardiacdsctimages AT qinlanxie nonrigidregistrationwithcorrespondingpointsconstraintforautomaticsegmentationofcardiacdsctimages AT shanxingou nonrigidregistrationwithcorrespondingpointsconstraintforautomaticsegmentationofcardiacdsctimages AT yunfeizha nonrigidregistrationwithcorrespondingpointsconstraintforautomaticsegmentationofcardiacdsctimages AT defengwang nonrigidregistrationwithcorrespondingpointsconstraintforautomaticsegmentationofcardiacdsctimages |
_version_ |
1725799997124902912 |