TY - GEN
T1 - A generic respiratory motion model based on 4D MRI imaging and 2D image navigators
AU - Fayad, Hadi J.
AU - Buerger, Christian
AU - Tsoumpas, Charalampos
AU - Cheze-Le-Rest, Catherine
AU - Visvikis, Dimitris
PY - 2012
Y1 - 2012
N2 - Respiratory motion modeling is a key method improving the accuracy of both diagnostic and therapeutic applications of multi-modality imaging. Most of the respiratory motion models presented to date are patient specific, based on 4D CT or MRI datasets. They require a 4D anatomical image acquisition for every patient, implying in the case of CT an associated increased dose not justified for all patients or require a 4D MRI acquisition which is associated with a compromise between image quality and longer acquisitions not compatible with clinical practice. The objective of this work was to create a global respiratory motion model based on principal component analysis using 4D T1 MRI volunteer studies covering the thoracic region. This model can then be adapted on a specific patient using only two 'breath hold' 3D MRI volumes in addition to 2D MRI images acquired using a 2D image navigator. Our generic model relates these navigators to the internal organ motion described by motion fields. Five volunteers were included in this study. Four of them were used to create the motion model which was subsequently tested on a fifth subject using a leave-one-out strategy. The global model accuracy was assessed by generating 4D MRI series and comparing them to the volunteer specific acquired 4D MRI images. The obtained accuracy and reproducibility results of the proposed generic respiratory motion model may allow robust and accurate respiratory motion correction in PET/MRI imaging without the need for patient specific 4D MRI acquisitions.
AB - Respiratory motion modeling is a key method improving the accuracy of both diagnostic and therapeutic applications of multi-modality imaging. Most of the respiratory motion models presented to date are patient specific, based on 4D CT or MRI datasets. They require a 4D anatomical image acquisition for every patient, implying in the case of CT an associated increased dose not justified for all patients or require a 4D MRI acquisition which is associated with a compromise between image quality and longer acquisitions not compatible with clinical practice. The objective of this work was to create a global respiratory motion model based on principal component analysis using 4D T1 MRI volunteer studies covering the thoracic region. This model can then be adapted on a specific patient using only two 'breath hold' 3D MRI volumes in addition to 2D MRI images acquired using a 2D image navigator. Our generic model relates these navigators to the internal organ motion described by motion fields. Five volunteers were included in this study. Four of them were used to create the motion model which was subsequently tested on a fifth subject using a leave-one-out strategy. The global model accuracy was assessed by generating 4D MRI series and comparing them to the volunteer specific acquired 4D MRI images. The obtained accuracy and reproducibility results of the proposed generic respiratory motion model may allow robust and accurate respiratory motion correction in PET/MRI imaging without the need for patient specific 4D MRI acquisitions.
U2 - 10.1109/NSSMIC.2012.6551927
DO - 10.1109/NSSMIC.2012.6551927
M3 - Conference contribution
AN - SCOPUS:84881574976
SN - 9781467320306
T3 - IEEE Nuclear Science Symposium Conference Record
SP - 4058
EP - 4061
BT - 2012 IEEE Nuclear Science Symposium and Medical Imaging Conference Record, NSS/MIC 2012
T2 - 2012 IEEE Nuclear Science Symposium and Medical Imaging Conference Record, NSS/MIC 2012
Y2 - 29 October 2012 through 3 November 2012
ER -