General Electric Company

Ge Precision Healthcare LLC

Ersin Bayram

Daniel Vance Litwiller

Robert Marc Lebel

Xinzeng Wang

Anthony Tienhuan Vu

Graeme Colin McKinnon

Charles Robert Michelich

Sagar Mandava

Arnaud Guidon

Reed Frederick Busse

Kang Wang

Jingfei Ma

Ajeetkumar Gaddipati

Ramesh Venkatesan

Yuji Iwadate

Shaorong Chang

Naoyuki Takei

Manojkumar Saranathan

Michael R Hartley

Wei Sun

Ali Ersoz

Zachary Slavens

Ken-Pin Hwang

Ty A Cashen

Tim Sprenger

Daniel V Litwiller

Lloyd Estkowski

Dan W Rettmann

Vijay Shivalingappa Nimbargi

Daniel Vanc Litwiller

A method for producing an image of a subject with a magnetic resonance imaging (MRI) comprises acquiring a first set of partial k-space data from the subject and generating a phase corrected image based on a phase correction factor and the first set of the partial k-space data. The method further includes transforming the phase corrected image into a second set of partial k-space data and reconstructing the image of the subject from the second set of the partial k-space data and a weighting function.

System and method for reconstruction using a high-resolution phase in magnetic resonance images

Methods and systems are provided for independently removing streak artifacts and noise from medical images, using trained deep neural networks. In one embodiment, streak artifacts and noise may be selectively and independently removed from a medical image by receiving the medical image comprising streak artifacts and noise, mapping the medical image to a streak residual and a noise residual using the trained deep neural network, subtracting the streak residual from the medical image to a first extent, and subtracting the noise residual from the medical image to a second extent, to produce a de-noised medical image, and displaying the de-noised medical image via a display device.

Methods and system for selective removal of streak artifacts and noise from images using deep neural networks

A computer-implemented method of removing truncation artifacts in magnetic resonance (MR) images is provided. The method includes receiving a crude image that is based on partial k-space data from a partial k-space that is asymmetrically truncated in at least one k-space dimension. The method also includes analyzing the crude image using a neural network model trained with a pair of pristine images and corrupted images. The corrupted images are based on partial k-space data from partial k-spaces truncated in one or more partial sampling patterns. The pristine images are based on full k-space data corresponding to the partial k-space data of the corrupted images, and target output images of the neural network model are the pristine images. The method further includes deriving an improved image of the crude image based on the analysis, wherein the derived improved image includes reduced truncation artifacts and increased high spatial frequency data.

Deep learning systems and methods of removal of truncation artifacts in magnetic resonance images

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.

Systems and methods for reducing colored noise in medical images using deep neural network

A computer-implemented method of reducing noise and artifacts in medical images is provided. The method includes receiving a series of medical images along a first dimension, wherein the signals in the medical images having a higher correlation in the first dimension than the noise and the artifacts in the medical images. The method further includes, for each of a plurality of pixels in the medical images, deriving a series of data points along the first dimension based on the series of medical images, inputting the series of data points into a neural network model, and outputting the component of signals in the series of data points. The neural network model is configured to separate a component of signals from a component of noise and artifacts in the series of data points. The method further includes generating a series of corrected medical images based on the outputted component of signals.

Systems and methods of reducing noise and artifacts in magnetic resonance imaging

A computer-implemented method of correcting phase and reducing noise in magnetic resonance (MR) phase images is provided. The method includes executing a neural network model for analyzing MR images, wherein the neural network model is trained with a pair of pristine images and corrupted images, wherein the corrupted images include corrupted phase information, the pristine images are the corrupted images with the corrupted phase information reduced, and target output images of the neural network model are the pristine images. The method further includes receiving MR images including corrupted phase information, and analyzing the received MR images using the neural network model. The method also includes deriving pristine phase images of the received MR images based on the analysis, wherein the derived pristine phase images include reduced corrupted phase information, compared to the received MR images, and outputting MR images based on the derived pristine phase images.

Systems and methods of generating robust phase images in magnetic resonance images

An MRI system for performing time resolved MR imaging of an object with grouped data acquisition is provided. The MRI system includes an MRI controller in electronic communication with a magnet assembly and operative to sample a group of data points within a first region of a k-space. The first region includes a central sub-region and a first peripheral sub-region. The MRI controller is further operative to sample a group of data points within a second region of the k-space. The second region includes the central sub-region and a second peripheral sub-region different from the first peripheral sub-region.

MRI system and method for performing time resolved MR imaging of an object with grouped data acquisition

A method that includes obtaining an MRI gradient echo train of at least three echo data sets at differing phase angles; producing a plurality of phase error maps among the at least three echo data sets; and imaging at least three distinct chemical species based on the plurality of phase error maps.

Method and apparatus for separating chemical species in magnetic resonance imaging

A method of magnetic resonance (MR) imaging includes segmenting a k-kplane comprising a plurality of encoding points into a plurality of annular segments. For each annular segment, a view ordering is defined based on a polar angle associated with each encoding point contained within the annular segment. MR data is acquired for the plurality of encoding points based on the view ordering for each annular segment and at least one MR image is generated using the acquired MR data.

System and method of angular elliptic centric view ordering for 3D MR acquisitions

Methods and systems are provided for modifying a pulse sequence. In one embodiment, a determination is made whether an estimated peripheral nerve stimulation (PNS) associated with a pulse sequence exceeds a PNS limit. If the estimated PNS exceeds the PNS limit, a slew rate associated with one or more axes of the pulse sequence may be reduced and the maximum gradient amplitudes for each axis of the pulse sequence may be adjusted. In one embodiment, adjustment of the maximum gradient amplitudes or local slew rate may be based upon a cost analysis performed on the pulse sequence.

Method and system for modifying pulse sequences

A method for acquiring magnetic resonance (MR) data for a three-dimensional (3D) dynamic study includes partitioning a k-kplane with a plurality of views into an inner region and a plurality of outer regions. The inner region includes a set of views in a central region of the k-kplane and each outer region includes a plurality of views outside of the central region of the k-kplane. The method also includes partitioning each outer region into a plurality of radial fan beam segments, defining a first view ordering for the inner region and defining a second view ordering for each outer region. Once the k-kplane is partitioned and the view orderings are defined, MR data is acquired for the set of views in the inner region and for all of the views in each of the outer regions in an alternating acquisition order where the set of views in the inner region are acquired more frequently than the views in each of the outer regions. At least one MR image is generated based on the acquired MR data.

Method and apparatus for view ordering of magnetic resonance imaging data for dynamic studies

A multi-shot three-dimensional (3D) magnetic resonance imaging (MRI) data view-ordering strategy for uniform or variable density k-space sampling schemes is presented. The k-kplane is partitioned into multiple wedge-shaped 'blades' (or radial fan beams). The angular size of individual blades may be adjusted according to the desired number of views for each blade. Each blade includes views near the origin that contain low-frequency information about the imaged object such that views having the most desirable characteristics may be used to fill the central portions of k-space.

Method and apparatus for acquiring magnetic resonance imaging data

A magnetic resonance imaging (MRI) apparatus includes an MRI system having a plurality of gradient coils positioned about a bore of a magnet. An RF transceiver system and an RF switch are controlled by a pulse module to transmit and receive RF signals to and from an RF coil assembly to acquire MR images. The MRI apparatus also includes a computer programmed to calculate a cost for each point to be acquired in k-space, acquire MR data based on the calculated cost for each point, and generate an MR image using the acquired MR data.

A system and method of MR imaging enables PROPELLER imaging to be feasibly carried out independently of slice orientation or anatomy of interest. The invention is directed to accelerated acquisition of blades of MR data that are rotated about a central region of k-space and reconstructing an image of arbitrary slice orientation from the blades of MR data that preserves contrast and reduces acceleration artifacts caused by signal amplitude variances.

Growing community of inventors

Delafield, WI, United States of America

Ersin Bayram