System for motion corrected MR diffusion imaging
Inventors
Bhat, Himanshu • Van Der Kouwe, Andre Jan Willem • Tisdall, Matthew Dylan • Heberlein, Keith Aaron
Assignees
Siemens Healthcare GmbH • General Hospital Corp • National Institutes of Health NIH
Publication Number
US-9687172-B2
Publication Date
2017-06-27
Expiration Date
2032-06-20
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
A system determines motion correction data for use in diffusion MR imaging using an RF signal generator and magnetic field gradient generator which sequentially acquire in a single first direction through a volume, first and second slice sets individually comprising multiple individual diffusion image slices. The first set of slices and the second set of slices are spatially interleaved within the volume, by providing in acquiring the second slice set, a low flip angle RF pulse successively followed by a non-diffusion image data readout magnetic field gradient for acquisition of data representing a two dimensional (2D) non-diffusion image used for motion detection of the first slice set successively followed by, a first diffusion imaging RF pulse followed by a first diffusion imaging phase encoding magnetic field gradient for preparation for acquiring data representing a diffusion image slice of the second slice set.
Core Innovation
This invention provides a system for determining motion correction data for use in diffusion magnetic resonance (MR) imaging of an anatomical volume. It employs an RF signal generator and a magnetic field gradient generator that sequentially acquire, in a single direction through the volume, two spatially interleaved slice sets comprising multiple diffusion image slices. The acquisition of the second slice set includes a low flip angle RF pulse followed immediately by a non-diffusion image data readout magnetic field gradient to acquire two-dimensional (2D) non-diffusion images that facilitate motion detection of the first slice set. This is then followed by diffusion imaging RF pulses and phase encoding gradients to acquire diffusion image slices of the second set.
The system also has an integrated embodiment where an individual diffusion image slice is acquired by a sequence comprising a first diffusion imaging RF pulse, a non-diffusion image readout gradient for 2D non-diffusion image acquisition for motion detection, and a diffusion imaging phase encoding gradient preparing for acquiring the diffusion image slice. This approach provides prospective motion correction by using non-diffusion encoded low-resolution single-shot echo planar imaging (EPI) images as navigators during diffusion scans, allowing detection and compensation for subject motion during image acquisition.
The underlying problem addressed is the inherent sensitivity of diffusion MR imaging to patient motion, especially during long scan times which can be from one to thirty minutes. Motion causes misalignment of images acquired in different diffusion directions and can result in signal dropouts. Existing retrospective and prospective motion correction methods have limited success in addressing these issues. This invention provides a comprehensive prospective motion correction technique that improves image quality and accuracy in diffusion MR neuroimaging by incorporating real-time motion detection and correction with interleaved or integrated low flip angle navigator images.
Claims Coverage
The patent discloses two main inventive systems and methods related to motion correction in diffusion MR imaging, set forth in the independent claims. Key inventive features relate to sequential acquisition of interleaved slice sets with navigator acquisitions and integrated acquisition sequences combining diffusion imaging with motion detection.
Interleaved acquisition of spatially interleaved slice sets with navigators for motion correction
A system comprising an RF signal generator and magnetic field gradient generator configured to sequentially acquire first and second spatially interleaved slice sets in a single direction through a volume. The second slice set acquisition includes a low flip angle RF pulse followed by a non-diffusion image data readout gradient enabling acquisition of a 2D non-diffusion image for motion detection of the first slice set, followed by diffusion imaging RF pulses and phase encoding gradients to acquire diffusion image slices of the second slice set.
Complementary interleaved acquisition of the first slice set including navigators
The system further acquires the first slice set using a low flip angle RF pulse followed by a non-diffusion image readout gradient for motion detection of the second slice set, followed by diffusion imaging RF and phase encoding gradients. This acquisition completes interleaving of diffusion image slices and navigator images for both slice sets to enable motion correction.
Diffusion imaging pulse sequence parameters and timing
Features include use of a low flip angle ranging from 5-30 degrees for navigator excitation, a 90 degree first diffusion imaging RF pulse, and a 180 degree second diffusion imaging RF pulse. Diffusion imaging phase encoding gradients are timed to occur within half the Echo Time (TE) used in acquiring diffusion image slices, preserving timing symmetry. The diffusion image data readout uses echo planar imaging (EPI) gradients.
Use of multiple diffusion acquisition techniques
The system supports acquiring diffusion image slices using multiple diffusion acquisition methods including Stejskal-Tanner, twice refocused, stimulated echo, q-space, diffusion spectrum imaging, and diffusion tensor imaging techniques.
Motion detection and correction based on multi-directional 2D non-diffusion images
The system acquires 2D non-diffusion images in a single second direction different from the first. An image data processor compares multiple non-diffusion images acquired in the first and second directions to detect object motion. It corrects three-dimensional spatial coordinates of diffusion image slices acquired in the second direction relative to slices acquired in the first direction, compensating for detected motion.
Together, these inventive features provide a system and method for prospective motion correction in diffusion MR imaging by interleaving or integrating low flip angle non-diffusion navigator acquisitions with diffusion imaging slices. This arrangement enables accurate motion detection and correction during diffusion image acquisition, accommodating multiple diffusion encoding methods without signal degradation.
Stated Advantages
Prospective motion correction improves alignment of diffusion images acquired in multiple diffusion directions, reducing errors in diffusion parameter calculations.
The system minimizes MR signal dropouts and image blurring caused by patient motion during long diffusion scans.
The interleaved method maintains the original Echo Time (TE) for diffusion images, resulting in negligible signal-to-noise ratio (SNR) loss, whereas the integrated method provides slice-specific motion estimates.
The method works independently of b-value used and does not require retrospective adjustment of diffusion gradient encoding (b-matrix).
Using low flip angle navigator images minimizes signal attenuation in diffusion acquisitions and reduces interaction between navigator and diffusion scans.
Documented Applications
Prospective motion correction in 2D multi-slice diffusion MR neuroimaging to improve image quality and parameter accuracy.
Motion detection and compensation during diffusion imaging in brain scans of subjects who may move during scanning.
Use in diffusion acquisitions employing various diffusion encoding methods such as Stejskal-Tanner, twice refocused spin echo, stimulated echo, q-space, diffusion spectrum imaging, and diffusion tensor imaging.
Implementation in clinical or research MR scanners equipped with RF coils, magnetic gradient systems, and real-time image processing to reduce motion artifacts during diffusion imaging.
Interested in licensing this patent?