A system for image reconstruction in magnetic resonance imaging (MRI) is provided. The system may obtain undersampled k-space data associated with an object, wherein the undersampled K-space data may be generated based on magnetic resonance (MR) signals collected by an MR scanner that scans the object. The system may construct an ordinary differential equation (ODE) that formulates a reconstruction of an MR image based on the undersampled k-space data. The system may further generate the MR image of the object by solving the ODE based on the undersampled k-space data using an ODE solver.

A method of magnetic resonance (MR) imaging of an object includes: generating MR signals by subjecting the object to a number N of shots of a multi-echo imaging sequence comprising multi-slice RF pulses for simultaneously exciting two or more spatially separate image slices, with a phase offset in the slice direction being imparted to the MR signals; acquiring the MR signals that are received in parallel via a set of at least two RF coils having different spatial sensitivity profiles; and reconstructing a MR image for each image slice from the acquired MR signals using a parallel reconstruction algorithm, wherein the MR signal contributions from the different image slices are separated on the basis of the spatial encodings of the MR signals according to the spatial sensitivity profiles of the RF coils and of the phase offsets attributed to the respective image slices and shots.

In a method for activating a magnetic resonance imaging (MRI) system for generating MRI data relating to an examination subject, in which system raw magnetic resonance (MR) data is captured, having at least one spin echo or turbo spin echo pulse sequence: a radio frequency (RF) excitation pulse is emitted to excite a region that contains a region to be imaged, the excited region being defined by a selection gradient; an RF refocusing pulse is emitted to influence a refocusing region, the refocusing region at least partly including the region to be imaged, and being defined by at least one selection gradient; and high-frequency (HF) signals are received to acquire raw MR data. A spatial extent of the excitation region is selected to be different (e.g. significantly different) from a spatial extent of the refocusing region.

In a method and magnetic resonance (MR) apparatus for reducing artifacts in an image dataset reconstructed from MR raw data that were acquired by radial sampling using different coil elements, for each of at least some of the coil elements, exclusion information is determined that identify MR data from that coil element that are responsible for at least one artifact, by a comparison of a sensitivity map, which defines a spatial reception capability of that coil element, with at least one comparison dataset obtained from at least a portion of the MR data from that coil element. At least the MR data identified from the exclusion information are excluded from the reconstruction of the image dataset.

A magnetic resonance imaging system includes an array radiofrequency coil and processing circuitry operatively linked to the array radiofrequency coil and configured to receive output signals from the array radiofrequency coil commensurate with a simultaneous multi-slice magnetic imaging characterized by simultaneous multi-slice parameters, estimate distorted regions of the image volume using either data obtained via a pre-scan or a pre-computed model, minimize overlap of the distorted regions with image voxels representing tissue to obtain optimized values of the simultaneous multi-slice parameters, configuring and executing the simultaneous multi-slice imaging sequence based on the optimized values of the simultaneous multi-slice parameters, and reconstruct simultaneous multi-slice images with minimized artifacts.

A method for magnetic resonance imaging reconstructs images that have reduced under-sampling artifacts from highly accelerated multi-spectral imaging acquisitions. The method includes performing by a magnetic resonance imaging (MRI) apparatus an accelerated multi-spectral imaging (MSI) acquisition within a field of view of the MRI apparatus, where the sampling trajectories of different spectral bins in the acquisition are different; and reconstructing bin images using neural network priors learned from training data as regularization to reduce under-sampling artifacts.

A method is disclosed for phase contrast magnetic resonance imaging (MRI) comprising: acquiring phase contrast 3D spatiotemporal MRI image data; inputing the 3D spatiotemporal MRI image data to a three-dimensional spatiotemporal convolutional neural network to produce a phase unwrapping estimate; generating from the phase unwrapping estimate an integer number of wraps per pixel; and combining the integer number of wraps per pixel with the phase contrast 3D spatiotemporal MRI image data to produce final output.

A method of cardiac strain analysis uses displacement encoded magnetic resonance image (MRI) data of a heart of the subject and includes generating a phase image for each frame of the displacement encoded MRI data. Phase images include potentially phase-wrapped measured phase values corresponding to pixels of the frame. A convolutional neural network CNN computes a wrapping label map for the phase image, and the wrapping label map includes a respective number of phase wrap cycles present at each pixel in the phase image. Computing an unwrapped phase image includes adding a respective phase correction to each of the potentially-wrapped measured phase values of the phase image, and the phase correction is based on the number of phase wrap cycles present at each pixel. Computing myocardial strain follows by using the unwrapped phase image for strain analysis of the subject.

Systems and methods for suppressing Nyquist ghost for diffusion weighted magnetic resonance imaging are disclosed. An exemplary method includes acquiring multiple k-space data sets using multiple sets of diffusion weighted imaging pulse sequences, reconstructing a magnetic resonance image from each of the multiple k-space data sets respectively, and averaging magnitudes of the magnetic resonance images to generate an average magnitude magnetic resonance image.

A method for accelerating diffusion magnetic resonance imaging (MRI) acquisition via slice interleaved diffusion encoding (SIDE) includes conducting a plurality of simultaneous multislice (SMS) excitations for each of a plurality of SIDE diffusion-weighted volumes to obtain SMS images of an MRI subject at different diffusion orientations, regrouping the images into slice groups with different orientations, generating a plurality of slice-undersampled diffusion weighted volumetric images of the subject, wherein each of the plurality of slice-undersampled diffusion weighted volumetric images is produced by cyclically interleaving the slice groups, such that each slice group is associated with a different diffusion wavevector, and reconstructing a full diffusion-weighted volumetric image of the subject by providing the plurality of slice-undersampled diffusion weighted volumetric images to a neural network trained to produce full diffusion-weighted volumetric versions of diffusion magnetic resonance images from undersampled versions of the diffusion magnetic resonance images.