GDL-FIRE4D: Deep Learning-based fast 4D CT image registration
Standard
GDL-FIRE4D: Deep Learning-based fast 4D CT image registration. / Sentker, Thilo; Madesta, Frederic; Werner, Rene.
Medical Image Computing and Computer Assisted Intervention – MICCAI 2018. Hrsg. / Alejandro F. Frangi; Julia A. Schnabel; Christos Davatzikos; Carlos Alberola-López; Gabor Fichtinger. Band 11070 1. Aufl. Springer, 2018. S. 765 - 773 (Lecture Notes in Computer Science).Publikationen: SCORING: Beitrag in Buch/Sammelwerk › SCORING: Beitrag in Sammelwerk › Forschung › Begutachtung
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - CHAP
T1 - GDL-FIRE4D: Deep Learning-based fast 4D CT image registration
AU - Sentker, Thilo
AU - Madesta, Frederic
AU - Werner, Rene
PY - 2018/9
Y1 - 2018/9
N2 - Deformable image registration (DIR) in thoracic 4D CT image data is integral for, e.g., radiotherapy treatment planning, but time consuming. Deep learning (DL)-based DIR promises speed-up, but present solutions are limited to small image sizes. In this paper, we propose a General Deep Learning-based Fast Image Registration framework suitable for application to clinical 4D CT data (GDL-FIRE4D). Open source DIR frameworks are selected to build GDL-FIRE4D variants. In-house acquired 4D CT images serve as training and open 4D CT data repositories as external evaluation cohorts. Taking up current attempts to DIR uncertainty estimation, dropout-based uncertainty maps for GDLFIRE4D variants are analyzed. We show that (1) registration accuracy of GDL-FIRE4D and standard DIR are in the same order; (2) computationtime is reduced to a few seconds (here: 60-fold speed-up); and (3) dropout-based uncertainty maps do not correlate to across-DIR vector field differences, raising doubts about applicability in the given context.
AB - Deformable image registration (DIR) in thoracic 4D CT image data is integral for, e.g., radiotherapy treatment planning, but time consuming. Deep learning (DL)-based DIR promises speed-up, but present solutions are limited to small image sizes. In this paper, we propose a General Deep Learning-based Fast Image Registration framework suitable for application to clinical 4D CT data (GDL-FIRE4D). Open source DIR frameworks are selected to build GDL-FIRE4D variants. In-house acquired 4D CT images serve as training and open 4D CT data repositories as external evaluation cohorts. Taking up current attempts to DIR uncertainty estimation, dropout-based uncertainty maps for GDLFIRE4D variants are analyzed. We show that (1) registration accuracy of GDL-FIRE4D and standard DIR are in the same order; (2) computationtime is reduced to a few seconds (here: 60-fold speed-up); and (3) dropout-based uncertainty maps do not correlate to across-DIR vector field differences, raising doubts about applicability in the given context.
UR - https://link.springer.com/content/pdf/10.1007%2F978-3-030-00928-1_86.pdf
M3 - SCORING: Contribution to collected editions/anthologies
VL - 11070
T3 - Lecture Notes in Computer Science
SP - 765
EP - 773
BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2018
A2 - Frangi, Alejandro F.
A2 - Schnabel, Julia A.
A2 - Davatzikos, Christos
A2 - Alberola-López, Carlos
A2 - Fichtinger, Gabor
PB - Springer
ER -
