TY - GEN
T1 - Voxel-wise analysis for spatial characterisation of pseudo-ct errors in MRI-only radiotherapy planning
AU - Chourak, Hilda
AU - Barateau, Anais
AU - Mylona, Eugenia
AU - Cadin, Capucine
AU - Lafond, Caroline
AU - Greer, Peter
AU - Dowling, Jason
AU - Jean-Claude-Nunes, null
AU - Crevoisier, Renaud De
AU - Acosta, Oscar
N1 - Funding Information:
This work was partially funded by region Bretagne (France) through ARED scholarship program, the University of Rennes 1 ”Defis Scientifiques Emergents” grant (France), and a PhD scholarship Grant from e-health Research Centre-CSIRO (Australia). The authors have no relevant financial or nonfinancial interest to disclose.
Publisher Copyright:
© 2021 IEEE.
PY - 2021/4/13
Y1 - 2021/4/13
N2 - Several approaches have been proposed to generate pseudo computed tomography (pCT) from MR images for radiotherapy dose calculation. Quantification of errors in pCT has been reported using global scores disregarding spatial heterogeneity. The aim of this work was to propose a population voxel-based workflow allowing the local assessment of errors in the generation of pCTs from MRI. For the voxel-wise analysis to be anatomically meaningful, a robust customized inter-patient non-rigid registration method brought the population images to the same coordinate system. To illustrate the use of this methodology, four pCT generation methods were compared: atlas-based, patch-based, and two deep learning methods. Considering global and local scores, deep learning appeared widely superior. Main source of errors were found in the cortical bones. The proposed workflow paves the way for quality control procedures within the clinical workflow.
AB - Several approaches have been proposed to generate pseudo computed tomography (pCT) from MR images for radiotherapy dose calculation. Quantification of errors in pCT has been reported using global scores disregarding spatial heterogeneity. The aim of this work was to propose a population voxel-based workflow allowing the local assessment of errors in the generation of pCTs from MRI. For the voxel-wise analysis to be anatomically meaningful, a robust customized inter-patient non-rigid registration method brought the population images to the same coordinate system. To illustrate the use of this methodology, four pCT generation methods were compared: atlas-based, patch-based, and two deep learning methods. Considering global and local scores, deep learning appeared widely superior. Main source of errors were found in the cortical bones. The proposed workflow paves the way for quality control procedures within the clinical workflow.
KW - MRI-Radiotherapy Treatment Planning
KW - Pseudo CT generation
KW - Voxel-based analysis
UR - http://www.scopus.com/inward/record.url?scp=85107196386&partnerID=8YFLogxK
UR - https://biomedicalimaging.org/2021/
U2 - 10.1109/ISBI48211.2021.9433800
DO - 10.1109/ISBI48211.2021.9433800
M3 - Conference contribution
AN - SCOPUS:85107196386
SN - 9781665429474
T3 - Proceedings - International Symposium on Biomedical Imaging
SP - 395
EP - 399
BT - 2021 IEEE 18th International Symposium on Biomedical Imaging, ISBI 2021
PB - IEEE, Institute of Electrical and Electronics Engineers
CY - United States
T2 - 18th IEEE International Symposium on Biomedical Imaging, ISBI 2021
Y2 - 13 April 2021 through 16 April 2021
ER -