Described here are systems and methods for designing and implementing spatially selective, multidimensional adiabatic radio frequency (“RF”) pulses for use in magnetic resonance imaging (“MRI”). Spatially selective inversion can be achieved adiabatically in both two-dimensional (“2D”) and three-dimensional (“3D”) regions-of-interest. The multidimensional adiabatic pulses are generally designed using sub-pulses that are adiabatically driven using a parent adiabatic pulse.

In a method to generate a spatially selective excitation in an imaging region of a magnetic resonance apparatus that precedes an acquisition of magnetic resonance data, in the course of the excitation an excitation trajectory in k-space is traversed, the excitation trajectory having a symmetry relative to the k-space center in at least one direction of k-space in the sense that a first traversed extreme value in this direction corresponds to the negative of the other extreme value traversed in this direction, so the excitation trajectory is shortened in the at least one directions on one side of the zero point between the extreme values, and the shortened excitation trajectory is used for excitation.

Representative methods and systems are disclosed for reducing image distortion or increasing spatial resolution in echo planar magnetic resonance imaging. In representative embodiments, a targeted field of view (FOV) image is divided into segments, with each segment having a predetermined overlap region with an adjacent segment, such as in a phase-encoding direction. Image data is acquired for each segment, sequentially or simultaneously, using a reduced phase-encoding FOV with a 2D radiofrequency (RF) excitation pulse, and rotated and scaled magnetic field gradients. The 2D RF excitation pulse may also be modulated, such as onto a plurality of different carrier frequencies, for simultaneous acquisition of multiple segments in the same imaging plane. Using the spatial response of the 2D RF excitation pulse, the acquired image data for each segment of the plurality of segments is combined to generate a combined magnetic resonance image having the targeted field of view.

A computer-implemented method of visualizing blood flow through a patient using magnetic resonance imaging (MRI) includes receiving an image of the portal venous system of the patient's liver at a full field of view. A reduced field of view is defined which encompasses the portal venous system of the patient's liver and excludes extraneous anatomy in the full field of view. A navigator area is defined in the full field of view and outside of the reduced field of view. Transmit channels are used to selectively excite the reduced field of view and the navigator area throughout a cardiac cycle of the patient. Measurement data is acquired in response to the selective excitation. The acquired data is used to generate time-resolved 3D datasets. Additionally, a 3D visualization of blood flow though the portal venous system is generated based on the time-resolved 3D datasets.

In magnetic resonance imaging using a measurement sequence of the “free precession of transverse magnetization in the steady state”-type i.e., an SSFP measurement sequence, during the SSFP measurement sequence, the implementation of a preparation sequence takes place to reduce a signal contribution of the transverse magnetization in an outer region surrounding a measurement region in the MR imaging. The implementation of the preparation sequence includes the radiation of a multidimensional, spatially selective RF pulse that acts in a spatially selective manner on the transverse magnetization in the outer region. Saturation of the transverse magnetization and/or dephasing of the transverse magnetization in the outer region can be achieved by the multidimensional, spatially selective RF pulse.

A system and method to detect abnormality of subjects directly from MRI k-space data are provided. The system includes: at least one computer hardware processor, at least one non-transitory computer-readable storage medium, and at least one computer program stored in the at least one non-transitory computer-readable storage medium and executable on the at least one computer hardware processor, wherein the at least one computer program includes: an acquisition module, configured to obtain target MRI k-space data by scanning a subject, wherein the target MRI k-space data are fully-sampled or undersampled or sparse MRI k-space data; a detection module, configured to obtain and output detection outcome from the target MRI k-space data using detection models; and a model training module, configured to train the detection models based on training data. Hence, the MRI scan time and related cost are reduced, and the accuracy of the detecting results is increased.

A method for operating a magnetic resonance imaging scanner, comprising: providing 3D data from a patient; providing target parameters, wherein the target parameters include an excitation of nuclear spins to be achieved; ascertaining a spectrally selective excitation pulse for emission by a transmitter based on the 3D data from the patient, wherein the spectrally selective excitation pulse is configured to generate the target parameters; and outputting the spectrally selective excitation pulse via the transmitter.

Described here are systems and methods for volumetric excitation in magnetic resonance imaging (“MRI”) using frequency modulated radio frequency (“RF”) pulses. In general, quadratic phase modulation along the slice encoding direction is implemented for additional spatiotemporal encoding, which better distributes signal content in the slice direction and enables higher acceleration rates that are robust to slice-undersampling.

According to one embodiment, a magnetic resonance imaging apparatus includes processing circuitry. The processing circuitry sets an excitation pulse sequence that applies an excitation pulse including an inversion pulse between at least one set of sub pulses of a local excitation radio frequency pulse formed of a plurality of sub pulses, and applies a spoiler gradient magnetic field that disperses transverse magnetization while applying the inversion pulse. The processing circuitry controls execution of the excitation pulse sequence by applying the excitation pulse and the spoiler gradient magnetic field according to the excitation pulse sequence, and collects a magnetic resonance signal based on a data collecting sequence after execution of the excitation pulse sequence.

Disclosed herein is a medical system (100, 300) comprising a memory (110) storing machine executable instructions (120) and a convolutional neural network (122). The convolutional neural network is configured to receive as input a complex array (128) encoding a selection of at least one excitation field of view (324, 900) and in response output a radio frequency wave form (130) and multiple spatially selective gradient pulse waveforms (132). The convolutional neural network is a multi-task convolutional neural network. The execution of the machine executable instructions causes a computational system (104) to: receive (200) a selection (124) of the at least one excitation field of view; receive (202) initial pulse sequence commands (126); encode (204) the complex array using the at least one excitation field of view; receive (206) the radio frequency wave form and the multiple spatially selective gradient pulse waveforms in response to inputting the complex array into the convolutional neural network; and construct (208) modified pulse sequence commands (134) by modifying the initial pulse sequence commands with the radio frequency wave form and the multiple spatially selective gradient pulse waveforms.