Non-cartesian retrospective reconstruction of cardiac motion in patients with severe arrhythmia
A retrospective reconstruction method uses cardiac self-gating for patients with severe arrhythmias. Self-gated myocardial systolic and diastolic motion is determined from low-spatial and high-temporal resolution images and then the MRI-dataset is retrospectively reconstructed to obtain high quality images. The method uses undersampled image reconstruction to obtain the low-spatial and high-temporal resolution images, including those of different beat morphologies. Processing of these images is utilized to generate a cardiac phase signal. This signal allows for arrhythmia detection and cardiac phase sorting. The cardiac phase signal allows for detection of end-systolic and diastolic events which allows for improved sampling efficiency. In the case of frequent and severe arrhythmia, the method utilizes data from the normal and interrupted beats to improve sampling and image quality.
1. A method of retrospectively reconstructing moving cardiac images with high resolution, comprising:
obtaining cardiac magnetic resonance data from a subject which allows for reconstruction of low spatial resolution and high temporal resolution cardiac images;
applying image processing to the obtained low spatial resolution cardiac images to create a cardiac phase signal;
applying arrhythmia detection and cardiac phase sorting to the cardiac phase signal to estimate the types of heart beats and estimate the cardiac phase for each data sample; and
applying the cardiac phase information to perform a retrospectively gated reconstruction.
2. The method according to claim 1 , further comprising obtaining cardiac data using a k-space sampling trajectory that allows for the reconstruction of low-spatial and high-temporal resolution images from a subset of the data.
3. The method according to claim 1 , further comprising obtaining cardiac data using a golden angle radial k-space sampling trajectory by the magnetic resonance imaging device.
4. The method according to claim 1 , further comprising obtaining cardiac images by reconstruction of undersampled datasets resulting in low spatial and high temporal resolution images.
5. The method according to claim 1 , wherein the low spatial resolution and high temporal resolution cardiac images comprise images reconstructed from a non-Cartesian sampling trajectory ink-space.
6. The method according to claim 5 , wherein the low spatial resolution and high temporal resolution cardiac images comprise images reconstructed from a spiral or radial sampling trajectory ink-space.
7. The method according to claim 1 , further comprising obtaining cardiac phase information by image segmentation of the low-spatial and high-temporal resolution images to estimate cardiac volume.
8. The method according to claim 1 , further comprising obtaining cardiac phase information by image-based metrics of the low-spatial and high-temporal resolution images to estimate cardiac phase such as image cross-correlation.
9. The method according to claim 1 , further comprising applying arrhythmia detection by measuring changes in duration of beats calculated from the cardiac phase signal.
10. The method according to claim 1 , further comprising applying arrhythmia detection by measuring changes in measured maximum and minimum values calculated from the cardiac phase signal.
11. The method according to claim 1 , further comprising applying arrhythmia detection by using MM-derived physiologic markers including the opening or closing of the aortic or mitral valves.
12. The method according to claim 1 wherein said cardiac magnetic resonance data is obtained with respect to at least two different beat morphologies from said subject, and wherein said reconstruction is performed with respect to each of said at least two different beat morphologies.