Premium
Pulmonary CT image registration and warping for tracking tissue deformation during the respiratory cycle through 3D consistent image registration
Author(s) -
Li Baojun,
Christensen Gary E.,
Hoffman Eric A.,
McLennan Geoffrey,
Reinhardt Joseph M.
Publication year - 2008
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.3005633
Subject(s) - image warping , image registration , computer vision , artificial intelligence , computer science , template matching , tracking (education) , breathing , airway , radiology , medicine , image (mathematics) , surgery , anatomy , psychology , pedagogy
Tracking lung tissues during the respiratory cycle has been a challenging task for diagnostic CT and CT‐guided radiotherapy. We propose an intensity‐ and landmark‐based image registration algorithm to perform image registration and warping of 3D pulmonary CT image data sets, based on consistency constraints and matching corresponding airway branchpoints. In this paper, we demonstrate the effectivenss and accuracy of this algorithm in tracking lung tissues by both animal and human data sets. In the animal study, the result showed a tracking accuracy of 1.9 mm between 50% functional residual capacity (FRC) and 85% total lung capacity (TLC) for 12 metal seeds implanted in the lungs of a breathing sheep under precise volume control using a pulmonary ventilator. Visual inspection of the human subject results revealed the algorithm's potential not only in matching the global shapes, but also in registering the internal structures (e.g., oblique lobe fissures, pulmonary artery branches, etc.). These results suggest that our algorithm has significant potential for warping and tracking lung tissue deformation with applications in diagnostic CT, CT‐guided radiotherapy treatment planning, and therapeutic effect evaluation.