Estimating lung respiratory motion using combined global and local statistical models

Zhong Xue, Ramiro Pino, Bin Teh

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

1 Scopus citations

Abstract

In image-guided therapy of lung cancer, discrepancies between pre-procedural 3D or 4D-CT and patient’s current lung shape could reduce the accuracy for guidance. While real-time imaging is often not available, it is necessary to estimate the lung motion from real-time measurable signals such as on-chest/abdominal fiducials or surface. Traditional regression models take the entire lung motion deformation as a whole and perform the estimation in a global manner. Given high dimensionality and complexity of the lung motion patterns, the correlation of lung motion in different local areas with the surface motion could be different. Therefore, we propose a combined global and local statistics-based estimation to improve the estimation accuracy because local deformations have similar patterns and could have higher correlation with the surface motion. Results with 37 4D-CT datasets suggest local motion estimation further improves the performance for lung respiratory motion modeling.

Original languageEnglish (US)
Title of host publicationPatch-Based Techniques in Medical Imaging - 2nd International Workshop, Patch-MI 2016 held in conjunction with MICCAI 2016, Proceedings
EditorsPierrick Coupe, Brent C. Munsell, Daniel Rueckert, Yiqiang Zhan, Guorong Wu
PublisherSpringer-Verlag
Pages133-140
Number of pages8
ISBN (Print)9783319471174
DOIs
StatePublished - 2016
Event2nd International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2016 - Athens, Greece
Duration: Oct 17 2016Oct 17 2016

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9993 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other2nd International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2016
CountryGreece
CityAthens
Period10/17/1610/17/16

Keywords

  • 4D-CT
  • High-dimensional data regression
  • Respiratory motion estimation
  • Statistical model

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint

Dive into the research topics of 'Estimating lung respiratory motion using combined global and local statistical models'. Together they form a unique fingerprint.

Cite this