The General Hospital Corporation

Siemens Healthcare Gmbh

Siemens Healthineers Ag

The United States of America, As Represented by the Secretary, Department of Health and Human Services

Case Western Reserve University

Stephen Farman Cauley

Lawrence Leroy Wald

Kawin Setsompop

Daniel Polak

Daniel Nicolas Splitthoff

Melissa Haskell

Jonathan R Polimeni

Berkin Bilgic

Bryan Clifford

Julian Hossbach

Mark A Griswold

Thorsten Feiweier

Josef Pfeuffer

Himanshu Bhat

Vibhas S Deshpande

Keith Aaron Heberlein

Thomas Witzel

Fei Han

Clarissa Zimmerman-Cooley

Cornelius Eichner

Sai Abitha Srinivas

Wei-Ching Lo

Steffen Bollman

Sohyun Han

A method is provided for acquiring a magnetic resonance image dataset of an object, using an imaging protocol in which several k-space lines are acquired in one echo train, wherein the echo train may include one or several sub echo trains, and wherein the imaging protocol includes a plurality of echo trains. Within the method, a set of additional k-space lines within a central region of k-space is acquired a number of times per echo train, wherein the number is greater than 1, in order to detect movement of the object during the echo train. The sets of additional k-space lines are acquired at pre-determined positions within the echo trains. The disclosure is also directed to a method for generating a motion-corrected magnetic resonance image dataset from the dataset thus acquired, a magnetic resonance imaging apparatus, and a computer program.

Method for acquiring a magnetic resonance image dataset and generating a motion-corrected image dataset

Systems and methods include segmentation of a first image of a subject to identify locations of anatomical structures of the subject, determination of a region of interest of the subject based on the locations of anatomical structures, determination of a coil-mixing matrix based on the region of interest, control of an MR scanner to acquire radial trajectory k-space data of the subject from each of a plurality of RF coils of the MR scanner, application of the coil-mixing matrix to the radial trajectory k-space data of the subject acquired from each of the plurality of RF coils to generate first radial trajectory k-space data, reconstruction of a second image of the subject based on the first radial trajectory k-space data, and display of the second image.

De-streaking algorithm for radial k-space data

Electromagnetic interference ('EMI') is mitigated for portable magnetic resonance imaging (“MRI”) systems using postprocessing interference suppression techniques that make use of EMI detectors external to the MRI system imaging volume to detect EMI signals and remove them from acquired magnetic resonance data. EMI correction models, including static transfer function-based models, dynamic transfer function-based models, correction weight-based models, or parallel imaging kernel-based models can be used to remove the EMI-related artifacts from the magnetic resonance data.

System and method for electromagnetic interference mitigation for portable MRI systems

A method for reconstructing a motion-corrected magnetic resonance image of a subject includes providing k-space magnetic resonance data including a plurality of shots, wherein each shot corresponds to an individual motion state of the subject. The method further includes providing motion parameters related to each motion state, determining redundancies across the motion states of the plurality of shots based on the motion parameters, compressing the plurality of motion states based on the determined redundancies across the motion states, and reconstructing the magnetic resonance image from the k-space magnetic resonance data based on the compressed plurality of motion states.

Method for reconstructing a motion-corrected magnetic resonance image of a subject

In a method and system for reducing motion artifacts in magnetic resonance image data, a scout scan of the region of the patient is performed, a magnetic resonance (MR) measurement of the region of the patient is performed to acquire MR image data of the region of the patient, and motion correction is performed on the acquired MR image data based on the scout scan to generate corrected MR image data. The motion correction technique advantageously reduces an influence of a patient motion on the magnetic resonance image data.

Scout acquisition enables rapid motion estimation and reduction (SAMER) systems and methods for retrospective motion mitigation

A combined physics-based and machine learning framework is used for reconstructing images from k-space data, in which motion artifacts are significantly reduced in the reconstructed images. In general, model-based retrospective motion correction techniques are accelerated using fast machine learning ('ML') steps, which may be implemented using a trained neural network such as a convolutional neural network. In this way, the confidence of a classical physics-based reconstruction is obtained with the computational benefits of an ML-based network.

Retrospective motion correction using a combined neural network and model-based image reconstruction of magnetic resonance data

A three-dimensional magnetic resonance image dataset of an object is acquired using a multi-shot imaging protocol in which several k-space lines are acquired in one shot. The three-dimensional k-space includes a central region and a periphery, wherein the sampling order of k-space lines differs between the central region and the periphery. At least one k-space line from each shot passes through the central region, whereas the periphery includes regions, which are sampled by k-space lines from a subset of the plurality of shots.

Method for acquiring a three-dimensional magnetic resonance image dataset and for generating a motion-corrected image dataset

A system and method include generation of modified imaging data and modified acquisition parameters of a training example based on the target imaging data and target acquisition parameters of each of a plurality of target examples, and generation, for each training example, of an initial reconstructed image based on the modified imaging data and modified acquisition parameters of the training example. A first network stage of a multi-stage network is trained based on the modified imaging data, modified acquisition parameters and initial reconstructed image of each training example, a first output image is generated for each training example by inputting the modified imaging data, modified acquisition parameters and initial reconstructed image of the training example to the trained first network stage, and a second network stage of the multi-stage network is trained based on the modified imaging data, modified acquisition parameters and first output image of each training example.

Training a multi-stage network to reconstruct MR images

Systems and Methods that identify the effect of motion during a medical imaging procedure. A neural network is trained to translate motion induced deviations of a coil-mixing matrix relative to a reference acquisition into a motion score. This score can be used for the prospective detection of the most corrupted echo trains for removal or triggering a replacement by reacquisition.

Coil mixing error matrix and deep learning for prospective motion assessment

Techniques are disclosed related to the compensation of phase variations introduced into k-space lines, which cause imaging artifacts. The techniques utilize the detection of motion via an encoding plus motion model, which does not require the use of additional prospective or retrospective motion detection techniques. The techniques described herein use the encoding plus motion model to reconstruct an initial image from a set of motion states, and then calculate phase information from images that are projected form the initial reconstructed image using a projection onto convex sets (POCS). The phase information is incorporated into the encoding plus motion model over several iterations to minimize data consistency error, thereby generating a refined image that compensates for patient motion over the set of motion states.

Phase estimation for retrospective motion correction

In a medical imaging auto-calibrated reconstruction method, an imaging scan is performed using a data acquisition scanner to generate image data, calibration data having a uniform sampling is determined, a point-spread function is determined based on the calibration data, and an image is reconstructed from the image data based on the point-spread function. A central region of k-space may have uniform sampling. The calibration data may be determined by extracting a uniformly-sampled central region of k-space from the image data. An outer region of k-space may have non-uniform sampling. A calibration scan may be performed to generate the calibration data.

Systems and methods for extending reconstructions to non-uniform k-space sampling

In a method and system for reducing motion artifacts in magnetic resonance image data, a scout scan (e.g. a three-dimensional (3D) scout scan) of the region of the patient is performed, a magnetic resonance (MR) measurement of the region of the patient is performed to acquire two-dimensional (2D) MR image data of the region of the patient, and motion correction is performed on the acquired 2D MR image data based on the scout scan to generate corrected MR image data. The motion correction technique advantageously reduces an influence of a patient motion on the magnetic resonance image data.

Systems and methods for separable motion estimation and correction using rapid three-dimensional (3D) volumetric scout acquisition

Systems and methods are provided for improving MRI data acquisition efficiency while providing more detailed information with high resolution and isotropic resolution without gaps. Improved data acquisition efficiency may be achieved by implementing a machine learning algorithm with a hardware processor and a memory to estimate imperfections in fast imaging sequences, such as a multi-shot echo planar imaging (MS-EPI) sequence. These imperfections, such as patient motion, physiological noise, and phase variations, may be difficult to model or otherwise estimate using standard physics-based reconstructions.

Multi-shot echo planar imaging through machine learning

The disclosure relates to a computer implemented method for magnetic resonance imaging. The method includes: receiving at least a first and a second subset of k-space data as radio frequency signals emitted from excited hydrogen atoms of a subject; sampling the first and second subset of k-space data; choosing the first subset of k-space data as a base subset of k-space data; estimating motion parameters of the second subset of k-space data against the base subset of k-space data; and correcting the second subset of k-space data based on the estimated motion parameters of the second subset of k-space data. The motion parameters of the second subset of k-space data are parameters of a non-linear motion estimating function representing a motion of the subject between receiving the first subset of k-space data and receiving the second subset of k-space data.

Computer implemented method and system for magnetic resonance imaging

Described here are systems and methods for producing images of a subject using magnetic resonance imaging ('MRI') in which data are acquired using a sparse approximate encoding scheme for controlled aliasing techniques. As one example, the sparse approximate encoding can be used for a Wave-CAIPI encoding scheme, which can enable faster image reconstruction using fewer computational resources, in addition to reducing noise in the reconstructed images relative to those reconstructed from data acquired using a Wave-CAIPI encoding scheme without sparse approximate encoding.

Growing community of inventors

Cambridge, MA, United States of America

Stephen Farman Cauley