Respiratory motion in coronary magnetic resonance angiography: A comparison of different motion models

Abstract

Purpose: To assess respiratory motion models for coronary magnetic resonance angiography (CMRA). In this study various motion models that describe the respiration-induced 3D displacements and deformations of the main coronary arteries were compared. Materials and Methods:Multiple high-resolution 3D coronary MR images were acquired in healthy volunteers using navigator-based respiratory gating, each depicting the coronary vessels at different respiratory motion states. In the images representing the different inspiratory states the displacements and deformations of the main coronary vessels with respect to the end-expiratory state were determined, by means of elastic registration. Several correction models (superior-inferior (SI) translation, 3D translation, and 3D affine transformation) were tested and compared with respect to their ability to map a selected inspiratory to the end-expiratory motion state.Results: 3D translation was found to be superior over SI translation, which is commonly used for prospective motion correction in CMRA. The 3D affine transformation was found to be the best correction model considered in this study. Furthermore, a large intersubject variability of the model parameters was observed. Conclusion:The results of this study indicate that a patient-adapted 3D correction model (3D translation or better 3D affine) will considerably improve prospective motion correction in CMRA.

Patient motion correction for multiplanar, multi‐breath‐hold cardiac cine MR imaging

“…It was concluded that the translational correction required is significantly higher than the corrections required for other 3D linear transformations (i.e., rotation, scaling, and shear) (12). However, correction involving a combination of other transformations may be necessary to further reduce the errors (12). Furthermore, respiratory motion is usually combined with bulk patient motion, which is predominantly translational.…”

Respiratory motion models: A review

“…In retrospective correction the acquired k-space data is postprocessed to remove the effects of motion. Such corrections are not limited to linear transformations (Batchelor et al, 2005) but the correction can be computationally demanding (Manke et al, 2002a;Odille et al, 2008b).…”

Adaptive Averaging for Improved SNR in Real-Time Coronary Artery MRI

“…If the set of component images satisfy (2) where is some underlying common structure, and represents a noise term that is white, additive and Gaussian with zero mean and standard deviation , then the resulting composite image and SNR will be given, respectively, by [11] (3) (4) (5) where is the SNR in the individual component images.…”

“…Additionally, unlike motion compensation methods, real-time MR is not dependent on a precise characterization of the complex coronary-diaphragm-electrocardiogram relationships [2], [3]. However, the high temporal resolution of real-time MR constrains the achievable signal-to-noise ratio (SNR) and/or spatial resolution.…”