General Electric Company

Ge Medical Systems Global Technology Company, LLC

Ge Precision Healthcare LLC

Leland Stanford Junior University

Ge Medical Systems Global Technology, LLC

Ge Medical Technology Services, Inc.

Graeme Colin McKinnon

Paul Edgar Licato

Robert Marc Lebel

Kevin Franklin King

Xiaohong Zhou

Ersin Bayram

Daniel Vance Litwiller

Xinzeng Wang

Eddy Benjamin Boskamp

Jingfei Ma

Shaorong Chang

Lawrence E Ploetz

Sagar Mandava

Qing Tan

Christopher Judson Hardy

Richard Louis Zinser

Jeffrey Michael Ashe

Glen Peter Koste

Brian A Hargreaves

Richard Scott Hinks

Jason Andrew Polzin

Joseph K Maier

Kevin Matthew Koch

Jean Helen Brittain

Dan Xu

Yongchuan Lai

Ileana Hancu

Dawei Gui

Robert Michael Vavrek

Josef P Debbins

Sarah Kathryn Patch

Sathish Ramani

Anton Max Linz

William James Balloni

Hongbin Wang

Daniel Joseph Weyers

Yoshihiro Tomoda

Daniel Joseph Schaefer

Weiwei Zhang

Ali Ersoz

Robert H Haworth

Mark T Radick

Tongzhou Wang

Mitsuhiro Bekku

Qin Liu

Bo Pettersson

Xucheng Zhu

Martin Andreas Janich

Eddy Benjemin Boskamp

Tejaswini Shubhachint

Rolf Feodor Schulte

Shumin Wang

Lawrence Edward Plo&eacute;tz

Sandhya Parameswaran

Andrew James Coristine

Arslan Amjad

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

Image reconstruction systems and methods include providing sensitivity maps for coils of a magnetic resonance imaging (MRI) system to a neural network. The systems and methods also include providing interleaved k-space data to the neural network, wherein the interleaved k-space data includes partial k-space data interleaved with zeros, or synthesized k-space data, to provide an extended field of view (FOV) different from a FOV utilized during acquisition of the partial k-space data, wherein the partial k-space data were obtained during a scan of a region of interest with the MRI system. The systems and methods further include outputting, from the neural network, a final reconstructed MR image based at least on the sensitivity maps and the interleaved k-space data, wherein the final reconstructed MR image includes the FOV utilized during the acquisition of the partial k-space data.

System and method for deep learning-based accelerated magnetic resonance imaging with extended field of view coil sensitivity calibration

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

Methods and systems are provided for reducing noise in medical images with deep neural networks. In one embodiment, a method for training a neural network comprises transforming each of a plurality of initial image data sets not acquired by a medical imaging modality into a target image data set, wherein each target image data set is in a format specific to the medical imaging modality, corrupting each target image data set to generate a corrupted image data set, and training the neural network to map each corrupted image data set to the corresponding target image data set. In this way, the high-resolution of digital non-medical photographs or images can be leveraged for the enhancement or correction of medical images, and the trained neural network can be used to reduce noise and image artifacts in medical images acquired by the medical imaging modality.

Systems and methods for noise reduction in medical images with deep neural networks

The present invention provides a magnetic resonance imaging method and system, the method comprising performing the following steps at least once: a composition step: performing image composition processing on raw images received by a receiving coil that is pre-determined as an artifact coil and a receiving coil that is pre-determined as a non-artifact coil to obtain a composite image; and a correction step: obtaining a product of the above composite image and space sensitivity of the above artifact coil to replace the raw image received by the above artifact coil, and performing the above composition step again.

Magnetic resonance imaging method and system

A system and method for localizing a deep brain stimulation electrode in vivo in a subject or object is provided. A magnetic resonance imaging system obtains MR image data from a volume-of-interest by way of a zero echo time (ZTE) or ultrashort echo time (UTE) pulse sequence, with one or more of a phase domain image and a magnitude domain image being analyzed from the MR image data acquired by the ZTE or UTE pulse sequence. One or more electrodes are localized within the volume-of-interest based on an analysis of the phase domain image and/or magnitude domain image. In localizing the electrodes, a multi-scale correlation-based analysis of the volume-of-interest is performed to estimate at least one of an electrode center and electrode contact locations of the electrode, with the localization being achieved with a fast scan-time and with a high level of accuracy and precision.

System and method for localization of deep brain stimulation electrode via magnetic resonance imaging

A method of magnetic resonance imaging includes executing an imaging sequence, in response to the imaging sequence, acquiring magnetic resonance data, entering the acquired magnetic resonance data in k-space in a memory along a predetermined k-space trajectory, and modifying the k-space trajectory during acquisition of the magnetic resonance data.

Magnetic resonance imaging system and method

Systems and method for magnetic resonance imaging are disclosed which utilize sinusoidal gradient waveforms to drive gradient coils in an MRI system. The sinusoidal gradient waveforms may be applied on all two or more (e.g. three) gradient axes to produce a relatively pure acoustic tone. In certain embodiments, gradient directions may be spiraled in three-dimensions to generate a radial pin-cushion k-space trajectory.

Low-noise magnetic resonance imaging using low harmonic pulse sequences

An imaging apparatus that incorporates an RF coil assembly and RF shield as part of a magnetic resonance imaging (MRI) system is disclosed. The imaging apparatus includes an MRI system having a plurality of gradient coils, an RF coil former having inner and outer surfaces, an RF shield positioned on the outer surface of the RF coil former so as to be formed about the RF coil former, and an RF coil positioned on the inner surface of the RF coil former, with the RF coil coupled to a pulse generator to emit an RF pulse sequence and receive resulting MR signals from a subject of interest. The RF coil former comprises a generally cylindrical member having an indented portion indented in a radial direction inwardly from the outer surface, with the RF shield conforming to the RF coil former so as to also have an indented portion.

Structured RF coil assembly for MRI scanner

A system and method for multi-spectral MR imaging near metal include a computer programmed to calculate an MR pulse sequence comprising a plurality of RF pulses configured to excite spins in an imaging object and comprising a plurality of volume selection gradients and determine a plurality of distinct offset frequency values. For each respective determined offset frequency value, the computer is programmed to execute the MR pulse sequence having a central transmit frequency and a central receive frequency of the MR pulse sequence set to the respective determined offset frequency value. The computer is also programmed to acquire a three-dimensional (3D) MR data set for each MR pulse sequence execution and generate a composite image based on data from each of the acquired 3D MR data sets.

System and method for multi-spectral MR imaging near metal

Systems and methods for reducing an amount of space occupied by a radio frequency coil assembly in a magnetic resonance imaging system are provided. In one embodiment, a radio frequency coil assembly for a magnetic resonance imaging system includes a radio frequency coil disposed cylindrically around a patient space and a radio frequency shield disposed cylindrically around the patient space and electrically coupled to the axial ends of the radio frequency coil. The radio frequency shield may be configured to extend behind the radio frequency coil, and the axial length of the radio frequency shield may be at least two times the axial length of the radio frequency coil.

Thin extended-cavity RF coil for MRI

Systems and methods for controlling a magnetic resonance imaging system are provided. In one embodiment, a magnetic resonance imaging system includes a radio frequency coil with a plurality of conductive coil elements, control circuitry that determines, based at least in part on a measurement of scattering parameters, a plurality of forward voltages that will cause power deposition into an object within a predetermined specific absorption rate, and an amplifier configured to apply the determined plurality of forward voltages respectively to the plurality of coil elements. The control circuitry may determine the plurality of forward voltages based at least in part on an unloaded measurement of scattering parameters and a loaded measurement of scattering parameters.

Using S-parameter measurements to manage SAR and transmit gain in MRI

Techniques for designing RF pulses may be configured to produce improved magnitude profiles of the resulting magnetization by relaxing the phase constraint and optimizing the phase profiles. In one embodiment, a spinor-based, optimal control, optimal phase technique may be used to design arbitrary-tip-angle (e.g., large and small tip angle) RF pulses (both parallel transmission and single channel). In another embodiment, small tip angle RF pulses (both parallel transmission and single channel) may be designed using a small-tip-angle (STA) pulse design without phase constraint that is formulated as a parameter optimization problem.

System and method for phase relaxed RF pulse design

A system and method for transmitting multiple radio frequency (RF) channels via an RF coil assembly are provided. An RF coil assembly having a number of coil elements may be configured to transmit a number of RF channels which is less than the number of coil elements thereof. Some implementations may use signal splitters for some or all of the RF channels to produce driving inputs for each coil element. By using more coil elements than RF channels, various embodiments may exhibit increased power efficiency and improved B1 uniformity.

System and method for multi-channel MR transmission

Growing community of inventors

Hartland, WI, United States of America

Graeme Colin McKinnon