In a method for generating an image dataset and one reference image dataset: a first raw dataset is provided that is recorded using a MR system, the first raw dataset including measurement signals at readout points in k-space that lie on a first k-space trajectory; a second raw dataset is provided that is recorded using the same MR system and on the same examination object, the second raw dataset including measurement signals at readout points in k-space that lie on a second, different k-space trajectory; image datasets are reconstructed from the first raw dataset, where a separate equalization coefficient set is used before the reconstruction of each image dataset; a reference image dataset is reconstructed from the second raw dataset; the reference image data set is compared with each image dataset to generate respective similarity values; and the image dataset is selected with a greatest similarity value.

Techniques are disclosed relating to the generation of a magnetic resonance (MR) image of a predetermined portion of a volume of an examination object. MR data of the portion may be acquired using echo trains in a first step and in a second step, with each of the echo trains acquiring MR data of a plurality of k-space lines. The plurality of k-space lines extend parallel to one another and perpendicular to a common plane such that per k-space line, one intersection point within a plane results. The MR image is then reconstructed using the acquired MR data.

A method of medical imaging including receiving k-space data that is divided into multiple k-space data groups, selecting one of the multiple k-space data groups as a reference k-space data group, and calculating spatial transform data for each of the multiple k-space data groups by inputting the multiple k-space data groups and the reference k-space data group into a transformation estimation module. The spatial transformation estimation module is configured for outputting spatial transform data descriptive of a spatial transform between a reference k-space data group and multiple k-space data groups in response to receiving the reference k-space data group and the multiple k-space data groups as input. The method further comprises reconstructing a corrected magnetic resonance image according to the magnetic resonance imaging protocol using the multiple k-space data groups and the spatial transform data for each of the multiple k-space data groups.

Accelerated data acquisition using two-dimensional (“2D”) radio frequency (“RF”) pulse segments as virtual receivers for a parallel image reconstruction technique, such as GRAPPA, is provided. Data acquisition is accelerated using segmented RF pulses for excitation, refocusing, or both, and undersampling k-space along a dimension of the RF pulse segments. In this way, parallel image reconstruction techniques, such as GRAPPA, can be adapted to work with a single RF receive coil. By undersampling the data acquisition and finding correlations between the data from different segments, unsampled data can be recovered. This shortens scan times, yielding the advantages of segmented pulses without the formerly required long scans.

A method for magnetic resonance imaging (MRI) may include cause, based on a pulse sequence, a magnetic resonance (MR) scanner to perform a scan on an object. The pulse sequence may include a steady-state sequence and an acquisition sequence that is different from the steady-state sequence. The steady-state sequence may correspond to a steady-state phase of the scan in which no MR data is acquired. The acquisition sequence may correspond to an acquisition phase of the scan in which MR data of the object is acquired. The method may also include generating one or more images of the object based on the MR data.

Systems and methods for improving calibration of MRI imaging using echo-planar imaging (EPI) include a multi-shot radio frequency (RF) excitation during a calibration phase and a processor that calibrates the k-space for a slice by acquiring k-space data through multi-shot EPI data acquisition for a plurality of interleaved segments in the slice, each divided into a predetermined number of readout lines. Each EPI data acquisition includes providing a series of frequency encoding pulses throughout a readout period equal to the predetermined number of readout lines, providing a series of phase encoding pulses during a middle portion of the readout period corresponding to a middle section of the k-space, capturing magnetic resonance signals during the middle portion. The frequency and phase encoding pulse each include a rewinder pulse before a spoiler pulse after the magnetic resonance signals are captured. The processor creates a calibration model from the acquired k-space data based on the magnetic resonance signals during the middle portion, wherein k-space data corresponding to each segment in the slice is acquired before acquiring data for subsequent slices.

According to one embodiment, a medical information processing apparatus has processing circuitry. The processing circuitry acquires medical data on a subject, acquires numerical data obtained by digitizing an acquisition condition of the medical data, and applies a machine learning model to input data including the numerical data and the medical data, thereby generating output data based on the medical data.

In a method for avoiding artifacts during acquisition of MR data, a first measurement data set (MDS) of a target region of the examination object and at least one second MDS of the target region are acquired, and a combined MDS is created based on the acquired data sets. The first MDS does not sample a first region of k-space to be sampled according to Nyquist and corresponding to a first partial factor, and a second MDS does not sample a second region of k-space to be sampled according to Nyquist and corresponding to a second partial factor. The first and second regions of the k-space are different from each other. Advantageously, a k-space region acquired in none of the acquisitions made can be minimized by the inventive variation in the respective sampling pattern of the acquired MDS, so artifacts are reduced/avoided in MR images reconstructed from the MDS.

Various examples relate to SENSitivity Encoding (SENSE) reconstruction of Magnetic Resonance Imaging (MRI) images. Multiple coil sensitivity maps per coil of a receiver coil array are used, e.g., obtained from an Eigenvalue-based Spatially Constrained Iterative Reconstruction Technique (ESPIRiT) autocalibration protocol.

The invention provides a method for performing magnetic resonance imaging, MRI, which exploits prior knowledge of the interactions between electromagnetic fields and spins in the sampled object. This technique is able to provide shorter acquisition times with respect to traditional (Nyquist-Shannon limited) MRI. The method is based on an encoding matrix formalism constructed from the specific knowledge of how every spin would evolve in time depending on their position for a given pulse sequence. This particular previous knowledge has not been fully exploited previously by traditional MRI techniques. Moreover, the method of the invention can be used in combination with other schemes, such as compressed sensing, parallel imaging, or deep learning, for further shortening the MRI scan time.