An image processing apparatus according to an embodiment includes processing circuitry. The processing circuitry obtains, on the basis of a magnetic resonance image, a plurality of shift images resulting from shifting the position of the pixel sampling grid of the magnetic resonance image by a plurality of mutually-different shift amounts. The processing circuitry generates a plurality of ringing-corrected images, by determining, with respect to each of the pixels included in each of the plurality of shift images, a shift amount from the position of the pixel to a position where ringing artifacts will be reduced and further performing a ringing correction to correct the ringing artifacts occurring in the shift images on the basis of the determined shift amounts. The processing circuitry combines, while interleaving, pixels included in each of the plurality of ringing-corrected images.

An image processing apparatus according to an embodiment includes processing circuitry. The processing circuitry determines, with respect to each of the pixels included in a magnetic resonance image, a shift amount from the position of the pixel to a position where ringing artifacts will be reduced and performs a ringing correction to correct the ringing artifacts occurring in the magnetic resonance image on the basis of the determined shift amounts. The processing circuitry estimates, with respect to each of the pixels, a local amplitude of ringing artifacts, and performs the ringing correction on the magnetic resonance image while determining the shift amount of each of the pixels so as to be approximately continuous with the shift amounts of adjacently positioned pixels, sequentially in descending order starting with pixels having a higher local amplitude on the basis of the estimated local amplitudes of ringing artifacts.

A method for reconstructing spatial information, the method includes: (i) Obtaining an under-sampled frequency domain representation (FDR) of the spatial information. The under-sampled FDR was obtained by sampling an FDR of the spatial information with a sampling mask. (ii) Feeding the under-sampled FDR and the sampling mask to a machine learning process. (iii) Reconstructing the spatial information by the machine learning process. The machine learning process was trained using a training data set that includes training under-sampled FRDs of training spatial information, and one or more training sampling masks that were used to sample FDRs of training spatial information.

Systems and methods for MRI are provided. The systems obtain a reference MR image of a target subject. The reference MR image is acquired by performing a reference MR scan on a first anatomical structure of the target subject before a target MR scan. The target MR scan is to be performed on a second anatomical structure of the target subject. The second anatomical structure is smaller than the first anatomical structure. The systems determine geometrical information of the second anatomical structure based on the reference MR image. The systems determine a target scanning protocol with respect to the target MR scan based on the geometrical information. The systems acquire a target MR image of the second anatomical structure by performing the target scan based on the target scanning protocol. In the target MR image, the second anatomical structure is not overlapped with other anatomical structures of the target subject.

An image generating apparatus according to the embodiment includes processing circuitry. The processing circuitry acquires MR data acquired in read-out directions including a first read-out direction and a second read-out direction intersecting the first read-out direction, filter sensitivity distributions corresponding to the read-out directions and indicating distributions of sensitivity of a low-pass filter, and coil sensitivity distributions corresponding to coil elements used to acquire the MR data. The processing circuitry generates synthesis sensitivity distributions for the respective read-out directions by synthesizing the filter sensitivity distributions and the coil sensitivity distributions for the respective read-out directions. The processing circuitry generates an MR image based on the synthesis sensitivity distributions and MR data.

Systems and methods of image reconstruction are provided. A system may have a memory and a processor to receive data corresponding to magnetic resonance imaging coils and data corresponding to a region of interest within a field of view of the magnetic resonance imaging machine. By determining different weights to associate with virtualized magnetic resonance imaging coils, images may be reconstructed to favor signals associated with a region of interest and to disfavor interference associated with areas outside the region of interest.

A system and method for magnetic resonance imaging is provided. The method includes dividing k-space into a plurality of regions along a dividing direction; scanning an object using a plurality of sampling sequences; acquiring a plurality of groups of data lines; filling the plurality of groups of data lines into the plurality of regions of the k-space; and reconstructing an image based on the filled k-space.

A system and computerized method for generating magnetic resonance imaging (MRI) images is provided that includes accessing data acquired from a subject using an MRI system that includes Nyquist ghosts and processing the data using a cost function that exploits a cosine and sine modulation in a ghosted image component of the data. An image of the subject is produced from the data after processing the data using the cost function.

Various methods and systems are provided for a radio frequency coil array comprising a plurality of coil elements for magnetic resonance imaging. In one embodiment, a method includes grouping the plurality of coil elements into receive elements groups (REGs) according to REGs information, generating coil element sensitivity maps for the plurality of coil elements, generating REG sensitivity maps based on the REGs information and the coil element sensitivity maps, determining, for each REG, signals within a region of interest (ROI) and signals outside of the ROI based on the REG sensitivity maps, selecting one or more REGs based on the signals within the ROI and the signals outside of the ROI of each REG, and scanning the ROI with the coil elements in the selected REGs being activated and the coil elements not in any selected REGs being deactivated. In this way, phase wrap artifacts may be reduced.

Systems and methods for accelerated diffusion-weighted magnetic resonance imaging using a tilted reconstruction kernel to synthesize unsampled k-space data in phase encoded and point spread function (PSF) encoded k-space data are provided. Images reconstructed from the data have reduced B.sub.0-related distortions and reduced T.sub.2* blurring. In general, data are acquired with systematically optimized undersampling of the PSF and phase encoding subspace. Parallel imaging reconstruction is implemented with a B.sub.0 inhomogeneity informed approach to achieve greater than twenty-fold acceleration of the PSF encoding dimension. A tilted reconstruction kernel is used to exploit the correlations in the phase encoding-PSF encoding subspace. Self-navigated phase corrections are computed from the acquired data and used to synthesize the unsampled k-space data.