The invention relates to a method of MR imaging of an object (10). It is an object of the invention to enable MR imaging in the presence of motion of the imaged object, wherein full use is made of the acquired MR signal and a high-quality MR image essentially free from motion artefacts is obtained. The method of the invention comprises the steps of: generating MR signals by subjecting the object (10) to an imaging sequence comprising RF pulses and switched magnetic field gradients; acquiring the MR signals as signal data over a given period of time (T); subdividing the period of time into a number of successive time segments (SO, S1, S2, . . . Sn); deriving a geometric transformation (DVF1, DVF2, . . . DVFn) in image space for each pair of consecutive time segments (S0, S1, S2, . . . Sn), which geometric transformation (DVF1, DVF2, . . . DVFn) reflects motion occurring between the two time segments of the respective pair; and reconstructing an MR image from the signal data, wherein a motion compensation is applied according to the derived geometric transformations (DVF1, DVF2, . . . DVFn). Moreover, the invention relates to an MR device (1) and to a computer program for an MR device (1).

Disclosed herein is a method of training a neural network (214) to perform a SENSE magnetic resonance imaging reconstruction. The method comprises receiving (100) initial training data, wherein the initial training data comprises sets of initial training complex channel images each paired with a predetermined number of initial ground truth images. The method further comprises generating (102) additional training data by performing data augmentation on the initial training data such that the data augmentation comprises adding a distinct phase offset to each of the set of initial training complex channel images during generation of the sets of additional training complex channel images. The method further comprises inputting (104) the sets of additional training complex channel images into the neural network and receiving in response a predetermined number of output training images and performing deep learning using the output training images.

The disclosure relates to techniques for determining an acquisition order identified with an acquired magnetic resonance data set, which comprises a total number of slices, using a simultaneous multislice technique.

The present disclosure is related to systems and methods for magnetic resonance imaging (MRI). The method includes obtaining a plurality of target sets of k-space data by filling target MR signals acquired by a plurality of coils of an MRI device into k-space along a corkscrew trajectory. The method includes obtaining a coil sensitivity of each of the plurality of coils. The method includes obtaining a point spread function corresponding to the corkscrew trajectory. The method includes generating a target image based on an objective function.

A method may include obtaining a plurality of imaging signals collected by applying a wave encoding gradient to a region of interest (ROI) of a subject. The method may also include obtaining a plurality of auxiliary signals associated with the ROI. The method may also include obtaining a point spread function corresponding to the wave encoding gradient. The method may also include determining, based on the plurality of auxiliary signals, temporal information relating to at least one temporal dimension of the ROI. The method may also include determining, based on the plurality of auxiliary signals, the plurality of imaging signals, and the point spread function, spatial information relating to at least one spatial dimension of the ROI. The method may also include generating at least one target image of the ROI based on the temporal information and the spatial information.

Method for operating an MR device to acquire MR data slices, wherein in a sequence section of an MR sequence, MR signals of at least two slices are measured simultaneously, and an acquisition order having an association of slices to respective sequence sections of a repetition sequence covering all slices of an associated concatenation is determined using an ordering rule. A crosstalk criterion is evaluated for the acquisition order by checking whether a first slice acquired in a last sequence section of the repetition sequence is directly adjacent to a second slice acquired in a first sequence section of the same repetition sequence. If the crosstalk criterion is fulfilled, the acquisition order is adapted according to an adaptation rule such that a larger temporal acquisition distance between the acquisition of the first and the second slices is provided.

A technology is provided for multi-component and/or multi-configuration imaging with coding, signal composition, signal model, structure model, structure model learning, decoding, reconstruction, performance prediction and performance enhancement. A magnetic resonance imaging example comprises acquiring signal samples in accordance with a coding scheme and a k-space sampling scheme, identifying a structure model in a data assembly formed using an extraction operation, and generating a result consistent with both the acquired signal samples and the identified structure model.

According to one embodiment, a magnetic resonance imaging apparatus includes sequence control circuitry and processing circuitry. The sequence control circuitry performs under-sampled data acquisition whose sample points are located at an equal interval in k-space and acquires k-space frames. The processing circuitry generates a plurality of k-space frames related to a plurality of time resolutions based on the k-space frames. In each of the plurality of k-space frames, the sample points are located at an equal interval, and the interval differs for each of the plurality of k-space frames. The processing circuitry generates a time-series image based on the plurality of k-space frames.

A computer-implemented method for generating an artifact corrected reconstructed contrast image from magnetic resonance imaging (MRI) data is provided. The method includes inputting into a trained deep neural network both a synthesized contrast image derived from multi-delay multi-echo (MDME) scan data or the MDME scan data acquired during a first scan of an object of interest utilizing a MDME sequence and a composite image, wherein the composite image is derived from both the MDME scan data and contrast scan data acquired during a second scan of the object of interest utilizing a contrast MRI sequence. The method also includes utilizing the trained deep neural network to generate the artifact corrected reconstructed contrast image based on both the synthesized contrast image or the MDME scan data and the composite image. The method further includes outputting from the trained deep neural network the artifact corrected reconstructed contrast image.

The present invention is to perform appropriate noise reduction processing on an image having different signal levels or noise levels depending on an imaging condition or a reconstruction condition. A magnetic resonance imaging apparatus according to the invention includes: a measurement unit that receives a nuclear magnetic resonance signal generated in a subject by a receiving coil; an image reconstruction unit that processes the nuclear magnetic resonance signal received by the receiving coil and reconstructs an image of the subject; an SNR spatial distribution calculation unit that calculates spatial distribution of a signal-to-noise ratio of the image using spatial distribution of a noise level and spatial distribution of the signal of the image; and a noise reduction unit that reduces noise from the image based on the spatial distribution of the signal-to-noise ratio.