Abstract: Methods and systems are provided for de-noising medical images using deep neural networks. In one embodiment, a method comprises receiving a medical image acquired by an imaging system, wherein the medical image comprises colored noise; mapping the medical image to a de-noised medical image using a trained convolutional neural network (CNN); and displaying the de-noised medical image via a display device. The deep neural network may thereby reduce colored noise in the acquired noisy medical image, increasing a clarity and diagnostic quality of the image.

Abstract: A magnetic resonance (MR) imaging method of correcting motion in precorrection MR images of a subject is provided. The method includes applying, by an MR system, a pulse sequence having a k-space trajectory of a blade being rotated in k-space. The method also includes acquiring k-space data of a three-dimensional (3D) imaging volume of the subject, the k-space data of the 3D imaging volume corresponding to the precorrection MR images and acquired by the pulse sequence. The method further includes receiving a 3D MR calibration data of a 3D calibration volume, wherein the 3D calibration volume is greater than or equal to the 3D imaging volume, jointly estimating rotation and translation in the precorrection MR images based on the k-space data of the 3D imaging volume and the calibration data, correcting motion in the precorrection images based on the estimated rotation and the estimated translation, and outputting the motion-corrected images.

Abstract: Methods and systems are provided for de-noising medical images using deep neural networks. In one embodiment, a method comprises receiving a medical image acquired by an imaging system, wherein the medical image comprises colored noise; mapping the medical image to a de-noised medical image using a trained convolutional neural network (CNN); and displaying the de-noised medical image via a display device. The deep neural network may thereby reduce colored noise in the acquired noisy medical image, increasing a clarity and diagnostic quality of the image.

Abstract: Methods and systems are provided for optimizing gradient waveforms for oblique imaging. In one embodiment, a method comprises generating initial gradient waveforms in logical axes, evaluating area demand of each of the initial gradient waveforms, increasing a maximum amplitude of the initial gradient waveform in a first logical axis, reducing a maximum amplitude of the initial gradient waveform in a second logical axis, wherein the area demand in the first logical axis is greater than the area demand in the second logical axis, converting the gradient waveforms to physical gradient waveforms, and driving physical amplifiers of an imaging system with the physical gradient waveforms during a scan. In this way, oblique scans may be performed without a performance reduction caused by increases in echo time, repetition time, and echo spacing.

Abstract: Various methods and systems are provided for acquiring a plurality blades of k-space data for magnetic resonance (MR) data acquisition. The plurality blades are arranged in a rotational manner around a center of the k-space. Each of the blades includes a plurality of parallel phase encoding lines indexed sequentially along a phase encoding direction of the blade. The phase encoding lines of each blade are sampled according to an asymmetric phase encoding order. The blade phase encoding orders of at least two adjacent blades are opposite to each other. This results in reducing shading and blurring artifacts in MRI images.

Abstract: Systems and methods for accelerated multi-contrast PROPELLER are disclosed herein. K-space is sampled in a rotating fashion using a plurality of radially directed blades around a center of k-space. A first subset of blades is acquired for a first contrast and a second subset of blades is acquired for a second contrasts. The first subset of blades is combined with high frequency components of the second subset of blades to produce an image of the first contrast. And the second subset of blades are combined with high frequency components of the first subset of blades to produce an image of the second contrast.

Abstract: Various methods and systems are provided for acquiring a plurality blades of k-space data for magnetic resonance (MR) data acquisition. The plurality blades are arranged in a rotational manner around a center of the k-space. Each of the blades includes a plurality of parallel phase encoding lines indexed sequentially along a phase encoding direction of the blade. The phase encoding lines of each blade are sampled according to an asymmetric phase encoding order. The blade phase encoding orders of at least two adjacent blades are opposite to each other. This results in reducing shading and blurring artifacts in MM images.

Abstract: Methods and systems are provided for optimizing gradient waveforms for oblique imaging. In one embodiment, a method comprises generating initial gradient waveforms in logical axes, evaluating area demand of each of the initial gradient waveforms, increasing a maximum amplitude of the initial gradient waveform in a first logical axis, reducing a maximum amplitude of the initial gradient waveform in a second logical axis, wherein the area demand in the first logical axis is greater than the area demand in the second logical axis, converting the gradient waveforms to physical gradient waveforms, and driving physical amplifiers of an imaging system with the physical gradient waveforms during a scan. In this way, oblique scans may be performed without a performance reduction caused by increases in echo time, repetition time, and echo spacing.

Abstract: A method for highly accelerated projection imaging (“HAPI”) is provided. In this method, conventional linear gradients are used to obtain coil sensitivity-weighted projections of the object being imaged. Only a relatively small number of projections, such as sixteen or less, of the object are required to reconstruct a two-dimensional image of the object, unlike conventional projection imaging techniques. The relationship between the voxel values of the imaged object and the coil sensitivity-weighted projections is formulated as a linear system of equations and the reconstructed images are obtained by solving this matrix equation. This method advantageously allows higher acceleration rates compared to echo planar imaging (“EPI”) with SENSE or GRAPPA acceleration. Moreover, the method does not require any additional or specialized hardware because hardware in conventional MRI scanners is adequate to implement the method.

Abstract: A method for highly accelerated projection imaging (“HAPI”) is provided. In this method, conventional linear gradients are used to obtain coil sensitivity-weighted projections of the object being imaged. Only a relatively small number of projections, such as sixteen or less, of the object are required to reconstruct a two-dimensional image of the object, unlike conventional projection imaging techniques. The relationship between the voxel values of the imaged object and the coil sensitivity-weighted projections is formulated as a linear system of equations and the reconstructed images are obtained by solving this matrix equation. This method advantageously allows higher acceleration rates compared to echo planar imaging (“EPI”) with SENSE or GRAPPA acceleration. Moreover, the method does not require any additional or specialized hardware because hardware in conventional MRI scanners is adequate to implement the method.