Learning statistical correlation of prostate deformations for fast registration

Yonghong Shi, Shu Liao, Dinggang Shen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper presents a novel fast registration method for aligning the planning image onto each treatment image of a patient for adaptive radiation therapy of the prostate cancer. Specifically, an online correspondence interpolation method is presented to learn the statistical correlation of the deformations between prostate boundary and non-boundary regions from a population of training patients, as well as from the online-collected treatment images of the same patient. With this learned statistical correlation, the estimated boundary deformations can be used to rapidly predict regional deformations between prostates in the planning and treatment images. In particular, the population-based correlation can be initially used to interpolate the dense correspondences when the number of available treatment images from the current patient is small. With the acquisition of more treatment images from the current patient, the patient-specific information gradually plays a more important role to reflect the prostate shape changes of the current patient during the treatment. Eventually, only the patient-specific correlation is used to guide the regional correspondence prediction, once a sufficient number of treatment images have been acquired and segmented from the current patient. Experimental results show that the proposed method can achieve much faster registration speed yet with comparable registration accuracy compared with the thin plate spline (TPS) based interpolation approach.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages1-9
Number of pages9
Volume7009 LNCS
DOIs
Publication statusPublished - 2011 Oct 17
Externally publishedYes
Event2nd International Workshop on Machine Learning in Medical Imaging, MLMI 2011, in Conjunction with the 14th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2011 - Toronto, ON, Canada
Duration: 2011 Sep 182011 Sep 18

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7009 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other2nd International Workshop on Machine Learning in Medical Imaging, MLMI 2011, in Conjunction with the 14th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2011
CountryCanada
CityToronto, ON
Period11/9/1811/9/18

Fingerprint

Statistical Learning
Registration
Interpolation
Planning
Radiotherapy
Correspondence
Splines
Interpolate
Radiation Therapy
Thin-plate Spline
Prostate Cancer
Interpolation Method
Sufficient
Predict
Prediction
Experimental Results

Keywords

  • Adaptive radiation therapy
  • Canonical correlation analysis
  • Fast registration
  • Patient-specific statistical correlation

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Shi, Y., Liao, S., & Shen, D. (2011). Learning statistical correlation of prostate deformations for fast registration. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7009 LNCS, pp. 1-9). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7009 LNCS). https://doi.org/10.1007/978-3-642-24319-6_1

Learning statistical correlation of prostate deformations for fast registration. / Shi, Yonghong; Liao, Shu; Shen, Dinggang.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7009 LNCS 2011. p. 1-9 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7009 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Shi, Y, Liao, S & Shen, D 2011, Learning statistical correlation of prostate deformations for fast registration. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 7009 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7009 LNCS, pp. 1-9, 2nd International Workshop on Machine Learning in Medical Imaging, MLMI 2011, in Conjunction with the 14th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2011, Toronto, ON, Canada, 11/9/18. https://doi.org/10.1007/978-3-642-24319-6_1
Shi Y, Liao S, Shen D. Learning statistical correlation of prostate deformations for fast registration. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7009 LNCS. 2011. p. 1-9. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-24319-6_1
Shi, Yonghong ; Liao, Shu ; Shen, Dinggang. / Learning statistical correlation of prostate deformations for fast registration. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7009 LNCS 2011. pp. 1-9 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{d6345802f14d4f068d6599be2fdb35b0,
title = "Learning statistical correlation of prostate deformations for fast registration",
abstract = "This paper presents a novel fast registration method for aligning the planning image onto each treatment image of a patient for adaptive radiation therapy of the prostate cancer. Specifically, an online correspondence interpolation method is presented to learn the statistical correlation of the deformations between prostate boundary and non-boundary regions from a population of training patients, as well as from the online-collected treatment images of the same patient. With this learned statistical correlation, the estimated boundary deformations can be used to rapidly predict regional deformations between prostates in the planning and treatment images. In particular, the population-based correlation can be initially used to interpolate the dense correspondences when the number of available treatment images from the current patient is small. With the acquisition of more treatment images from the current patient, the patient-specific information gradually plays a more important role to reflect the prostate shape changes of the current patient during the treatment. Eventually, only the patient-specific correlation is used to guide the regional correspondence prediction, once a sufficient number of treatment images have been acquired and segmented from the current patient. Experimental results show that the proposed method can achieve much faster registration speed yet with comparable registration accuracy compared with the thin plate spline (TPS) based interpolation approach.",
keywords = "Adaptive radiation therapy, Canonical correlation analysis, Fast registration, Patient-specific statistical correlation",
author = "Yonghong Shi and Shu Liao and Dinggang Shen",
year = "2011",
month = "10",
day = "17",
doi = "10.1007/978-3-642-24319-6_1",
language = "English",
isbn = "9783642243189",
volume = "7009 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "1--9",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Learning statistical correlation of prostate deformations for fast registration

AU - Shi, Yonghong

AU - Liao, Shu

AU - Shen, Dinggang

PY - 2011/10/17

Y1 - 2011/10/17

N2 - This paper presents a novel fast registration method for aligning the planning image onto each treatment image of a patient for adaptive radiation therapy of the prostate cancer. Specifically, an online correspondence interpolation method is presented to learn the statistical correlation of the deformations between prostate boundary and non-boundary regions from a population of training patients, as well as from the online-collected treatment images of the same patient. With this learned statistical correlation, the estimated boundary deformations can be used to rapidly predict regional deformations between prostates in the planning and treatment images. In particular, the population-based correlation can be initially used to interpolate the dense correspondences when the number of available treatment images from the current patient is small. With the acquisition of more treatment images from the current patient, the patient-specific information gradually plays a more important role to reflect the prostate shape changes of the current patient during the treatment. Eventually, only the patient-specific correlation is used to guide the regional correspondence prediction, once a sufficient number of treatment images have been acquired and segmented from the current patient. Experimental results show that the proposed method can achieve much faster registration speed yet with comparable registration accuracy compared with the thin plate spline (TPS) based interpolation approach.

AB - This paper presents a novel fast registration method for aligning the planning image onto each treatment image of a patient for adaptive radiation therapy of the prostate cancer. Specifically, an online correspondence interpolation method is presented to learn the statistical correlation of the deformations between prostate boundary and non-boundary regions from a population of training patients, as well as from the online-collected treatment images of the same patient. With this learned statistical correlation, the estimated boundary deformations can be used to rapidly predict regional deformations between prostates in the planning and treatment images. In particular, the population-based correlation can be initially used to interpolate the dense correspondences when the number of available treatment images from the current patient is small. With the acquisition of more treatment images from the current patient, the patient-specific information gradually plays a more important role to reflect the prostate shape changes of the current patient during the treatment. Eventually, only the patient-specific correlation is used to guide the regional correspondence prediction, once a sufficient number of treatment images have been acquired and segmented from the current patient. Experimental results show that the proposed method can achieve much faster registration speed yet with comparable registration accuracy compared with the thin plate spline (TPS) based interpolation approach.

KW - Adaptive radiation therapy

KW - Canonical correlation analysis

KW - Fast registration

KW - Patient-specific statistical correlation

UR - http://www.scopus.com/inward/record.url?scp=80054003265&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80054003265&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-24319-6_1

DO - 10.1007/978-3-642-24319-6_1

M3 - Conference contribution

AN - SCOPUS:80054003265

SN - 9783642243189

VL - 7009 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 1

EP - 9

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -