Exploiting Quasiperiodicity in Motion Correction of Free-Breathing Myocardial Perfusion MRI

Abstract

Free-breathing image acquisition is desirable in first-pass gadolinium- enhanced magnetic resonance imaging (MRI), but the breathing movements hinder the direct automatic analysis of the myocardial perfusion and qualitative readout by visual tracking. Nonrigid registration can be used to compensate for these movements but needs to deal with local contrast and intensity changes with time. We propose an automatic registration scheme that exploits the quasiperiodicity of free breathing to decouple movement from intensity change. First, we identify and register a subset of the images corresponding to the same phase of the breathing cycle. This registration step deals with small differences caused by movement but maintains the full range of intensity change. The remaining images are then registered to synthetic references that are created as a linear combination of images belonging to the already registered subset. Because of the quasiperiodic respiratory movement, the subset images are distributed evenly over time and, therefore, the synthetic references exhibit intensities similar to their corresponding unregistered images. Thus, this second registration step needs to account only for the movement. Validation experiments were performed on data obtained from six patients, three slices per patient, and the automatically obtained perfusion profiles were compared with profiles obtained by manually segmenting the myocardium. The results show that our automatic approach is well suited to compensate for the free-breathing movement and that it achieves a significant improvement in the average Pearson correlation coefficient between manually and automatically obtained perfusion profiles before (0.87 +/- 0.18) and after (0.96 +/- 0.09) registration.

Non-rigid motion compensation in free-breathing myocardial perfusion magnetic resonance imaging

“…The average correlation between the manual and automatically obtained time-intensity curves before registration 0.89 ± 0.11 was increased to 0.98 ± 0.02 (Fig. 3), an improvement over the value of 0.96 ± 0.05 that was achieved by employing the method published in [5].…”

“…by employing more complex registration criterions that can be directly applied to images with varying intensity distributions [3,4]. In [5] a two step procedure was proposed that exploits the quasi-periodicy of the free breathing motion that reduces the need to register images from different breathing phases, but it does not eliminate it completely.…”

“…In addition, they used the ICA and prior knowledge to extract a bounding box around the left heart ventricle (LV) making rigid registration possible. However, in [5] it was discussed that the method does not perform reliably for data acquired with a free breathing protocol: The synthetic references generally contained a lot of residual movement that made it impossible to achieve complete registration, and, in some cases it was impossible to identify the RV and LV cavities, and, hence, to create a proper bounding box that is required for rigid registration.…”

On Post-Acquisition Motion Compensation for Prostate Perfusion Analysis

“…The approach of Wollny et al [12] can be seen as a hybrid to the use of complex registration criteria and the creation of synthetic references. First, they selected a subset of images distributed over the whole series that had a high similarity pointing to images that are already well aligned, and these images were then registered to one global reference optimizing a image similarity criterion based on normalized gradient fields (NGF).…”

Registration of Free-Breathing Abdominal 3D Contrast-Enhanced CT

“…The current study focuses on tumors in liver or kidney using dynamic enhanced CT. Methods for registration of images acquired with free-breathing are being developed essentially for contrast-enhanced MRI. Wollny et al [7] exploit the quasi periodicity within myocardial perfusion and Li et al [2] explicitly estimate the respiratory cycle within abdominal perfusion. In both cases, an important assumption is a high time resolution.…”