公开(公告)号：US11042803B2

公开(公告)日：2021-06-22

申请号：US16276135

申请日：2019-02-14

Applicant: General Electric Company

Inventor： Itzik Malkiel ,  Christopher Judson Hardy

IPC: G06K9/00 ,  G06N3/08 ,  G06N3/04 ,  G06T3/40 ,  G06N20/20

Abstract: A method of reconstructing imaging data into a reconstructed image may include training a generative adversarial network (GAN) to reconstruct the imaging data. The GAN may include a generator and a discriminator. Training the GAN may include determining a combined loss by adaptively adjusting an adversarial loss based at least in part on a difference between the adversarial loss and a pixel-wise loss. Additionally, the combined loss may be a combination of the adversarial loss and the pixel-wise loss. Training the GAN may also include updating the generator based at least in part on the combined loss. The method may also include receiving, into the generator, the imaging data and reconstructing, via the generator, the imaging data into a reconstructed image.

公开(公告)号：US20200337591A1

公开(公告)日：2020-10-29

申请号：US16394774

申请日：2019-04-25

Applicant: General Electric Company

Inventor： Michael Rotman ,  Rafael Shmuel Brada ,  Sangtae Ahn ,  Christopher Judson Hardy ,  Itzik Malkiel ,  Ron Wein

IPC: A61B5/055 ,  G01R33/565 ,  G01R33/48

Abstract: K-space data obtained from a magnetic resonance imaging scan where motion was detected is split into two parts in accordance with the timing of the motion to produce first and second sets of k-space data corresponding to different poses. Sub-images are reconstructed from the k first and second sets of k-space data, which are used as inputs to a deep neural network which transforms them into a motion-corrected image.

公开(公告)号：US10551458B2

公开(公告)日：2020-02-04

申请号：US15637708

申请日：2017-06-29

Applicant: General Electric Company

Inventor： Ek Tsoon Tan ,  Giang-Chau Ngo ,  Christopher Judson Hardy ,  Thomas Kwok-Fah Foo

IPC: G01R33/48 ,  G01R33/56

Abstract: A magnetic resonance (MR) imaging method performed by an MR imaging system includes acquiring MR data in multiple shots and multiple acquisitions (NEX), separately reconstructing the component magnitude and phase of images corresponding to the multiple shots and multiple NEX, removing the respective phase from each of the images, and combining, after removal of the respective phase, the shot images and the NEX images to produce a combined image. The method further includes using the combined image to calculate the full k-space data for each shot and NEX and replacing unacquired k-space data points with calculated k-space data points. The operations are repeated until the combined image reaches a convergence.

公开(公告)号：US20190266761A1

公开(公告)日：2019-08-29

申请号：US15907797

申请日：2018-02-28

Applicant: GENERAL ELECTRIC COMPANY

Inventor： Itzik Malkiel ,  Sangtae Ahn ,  Christopher Judson Hardy

IPC: G06T11/00 ,  G01R33/561 ,  G01R33/48

Abstract: A method for sparse image reconstruction includes acquiring coil data from a magnetic resonance imaging device. The coil data includes undersampled k-space data corresponding to a subject. The method further includes processing the coil data using an image reconstruction technique to generate an initial undersampled image. The method also includes generating a reconstructed image based on the coil data, the initial undersampled image, and a plurality of iterative blocks of a flared network. A first iterative block of the flared network receives the initial undersampled image. Each of the plurality of iterative blocks includes a data consistency unit and a regularization unit and the iterative blocks are connected both by direct connections from one iterative block to the following iterative block and by a plurality of dense skip connections to non-adjacent iterative blocks. The flared network is based on a neural network trained using previously acquired coil data.

公开(公告)号：US09689950B2

公开(公告)日：2017-06-27

申请号：US14272756

申请日：2014-05-08

Applicant: General Electric Company

Inventor： Christopher Judson Hardy ,  Anja Christina Sophie Brau ,  Ek Tsoon Tan

IPC: G01R33/563 ,  G01R33/56

CPC classification number: G01R33/56316 ,  G01R33/5607

Abstract: A system and method of self-calibrated correction for residual phase in phase-contrast magnetic resonance (PCMR) imaging data. The method includes receiving PCMR image data from an MR scanner system, segmenting static tissue from non-static cardiovascular elements of the image data, calculating a non-linear fitted-phase basis function, the non-linear fitted-phase basis function based on system artifacts of the PCMR system, adding the non-linear fitted-phase basis function to linear fit terms, and subtracting the result of the adding step from the PCMR imaging data. The system includes a PCMR scanning apparatus configured to provide PCMR image data, a scanner control circuit configured to control the scanning apparatus during image acquisition, the scanner control circuitry in communication with a control processor, the control processor configured to execute computer-readable instructions that cause the control processor to perform the method. A non-transitory computer-readable medium is also disclosed.

公开(公告)号：US09594144B2

公开(公告)日：2017-03-14

申请号：US14260069

申请日：2014-04-23

Applicant: General Electric Company

Inventor： Christopher Judson Hardy ,  Graeme Colin McKinnon

IPC: G01R33/565 ,  G01R33/34 ,  G01R33/36 ,  G01R33/38 ,  G01R33/48 ,  G01R33/385 ,  G01R33/561

CPC classification number: G01R33/56572 ,  G01R33/34092 ,  G01R33/36 ,  G01R33/3808 ,  G01R33/3854 ,  G01R33/4826 ,  G01R33/5616 ,  G01R33/5617

Abstract: 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.

Abstract translation: 公开了用于磁共振成像的系统和方法，其利用正弦梯度波形来驱动MRI系统中的梯度线圈。 正弦梯度波形可以施加在所有两个或更多（例如三个）梯度轴上，以产生相对纯的声音。 在某些实施例中，梯度方向可以三维地螺旋化以产生径向针垫k空间轨迹。

公开(公告)号：US20150309144A1

公开(公告)日：2015-10-29

申请号：US14272756

申请日：2014-05-08

Applicant: General Electric Company

Inventor： Christopher Judson Hardy ,  Anja Christina Sophie Brau ,  Ek Tsoon Tan

IPC: G01R33/563 ,  G01R33/56 ,  G01R33/341 ,  G01R33/385 ,  G01R33/58 ,  G01R33/30

CPC classification number: G01R33/56316 ,  G01R33/5607

Abstract: A system and method of self-calibrated correction for residual phase in phase-contrast magnetic resonance (PCMR) imaging data. The method includes receiving PCMR image data from an MR scanner system, segmenting static tissue from non-static cardiovascular elements of the image data, calculating a non-linear fitted-phase basis function, the non-linear fitted-phase basis function based on system artifacts of the PCMR system, adding the non-linear fitted-phase basis function to linear fit terms, and subtracting the result of the adding step from the PCMR imaging data. The system includes a PCMR scanning apparatus configured to provide PCMR image data, a scanner control circuit configured to control the scanning apparatus during image acquisition, the scanner control circuitry in communication with a control processor, the control processor configured to execute computer-readable instructions that cause the control processor to perform the method. A non-transitory computer-readable medium is also disclosed.

Abstract translation: 相位对比磁共振（PCMR）成像数据中残留相自校正校正的系统和方法。 该方法包括从MR扫描器系统接收PCMR图像数据，从图像数据的非静态心血管元件分割静态组织，计算非线性拟合相位基函数，基于系统的非线性拟合相位函数 PCMR系统的伪像，将非线性拟合相位基函数加到线性拟合项，并从PCMR成像数据中减去加法步骤的结果。 该系统包括配置成提供PCMR图像数据的PCMR扫描装置，被配置为在图像采集期间控制扫描装置的扫描仪控制电路，与控制处理器通信的扫描仪控制电路，该控制处理器被配置为执行计算机可读指令， 使控制处理器执行该方法。 还公开了一种非暂时的计算机可读介质。

公开(公告)号：US20150301131A1

公开(公告)日：2015-10-22

申请号：US14257713

申请日：2014-04-21

Applicant: General Electric Company

Inventor： Christopher Judson Hardy

IPC: G01R33/28 ,  G01R33/56 ,  G01R33/561 ,  G01R33/36

CPC classification number: G01R33/4833

Abstract: A method of designing quiet variable-rate MRI slice-select pulses includes creating discretized first slice-select constant-amplitude gradient and RF waveforms, associating discretized time points having a first constant time increment with the first waveforms, selecting a scaling function that smooths the gradient waveform when multiplied together, multiplying the gradient and RF waveforms by the corresponding value of the scaling function to create second gradient and RF waveforms, dividing the time increment between the discretized time points by the corresponding value of the scaling function to create a remapped time increment, cumulatively summing the remapped time increments to create a remapped time scale, interpolating the second gradient and RF waveforms along the remapped time scale to form final gradient and RF waveforms, and providing the final gradient and RF waveforms for incorporation into an MRI pulse sequence. A system implementing the method and a non-transitory computer-readable medium are disclosed.

Abstract translation: 设计安静的可变速率MRI切片选择脉冲的方法包括：创建离散化的第一切片选择恒幅度梯度和RF波形，将具有第一恒定时间增量的离散时间点与第一波形相关联;选择缩放函数， 当梯度波形相乘时，将梯度和RF波形乘以比例函数的相应值，以创建第二梯度和RF波形，将离散时间点之间的时间增量除以缩放函数的相应值，以创建重新映射的时间 增加，累积求和重新映射的时间增量以创建重新映射的时标，沿着重新映射的时间尺度内插第二梯度和RF波形以形成最终的梯度和RF波形，并提供最终的梯度和RF波形以便并入MRI脉冲序列 。 公开了实现该方法和非暂时性计算机可读介质的系统。

公开(公告)号：US20150115955A1

公开(公告)日：2015-04-30

申请号：US14067701

申请日：2013-10-30

Applicant: General Electric Company

Inventor： Seung-Kyun Lee ,  Christopher Judson Hardy ,  Ek Tsoon Tan

IPC: G01R33/561

CPC classification number: G01R33/561 ,  G01R33/4835 ,  G01R33/5611

Abstract: Magnetic resonance imaging systems and methods are provided. A method includes applying a slice selection gradient perpendicular to a desired slice plane and applying, substantially simultaneously with the slice selection gradient, a radiofrequency nuclear magnetic resonance excitation pulse having a bandwidth corresponding to the desired slice plane and a frequency corresponding to the frequency of protons present in the desired slice plane. The method also includes applying, during an encoding period and in a first direction, a phase encoding gradient having a phase encoding portion and a shearing portion and applying, during the readout period and in a second direction perpendicular to the first direction, a frequency encoding gradient having a portion having substantially the same shape as the shearing portion of the phase encoding gradient.

Abstract translation: 提供磁共振成像系统和方法。 一种方法包括应用垂直于期望切片平面的切片选择梯度并且基本上与切片选择梯度一起施加具有对应于期望切片平面的带宽和对应于质子频率的频率的射频核磁共振激励脉冲 存在于期望的切片平面中。 该方法还包括在编码周期期间和在第一方向上应用具有相位编码部分和剪切部分的相位编码梯度，并且在读出期间和垂直于第一方向的第二方向上应用频率编码 梯度具有与相位编码梯度的剪切部分基本上相同形状的部分。

公开(公告)号：US20140148684A1

公开(公告)日：2014-05-29

申请号：US13686226

申请日：2012-11-27

Applicant: GENERAL ELECTRIC COMPANY

Inventor： Thomas Kwok-Fah Foo ,  Christopher Judson Hardy ,  Charles William Stearns ,  Ravindra Mohan Manjeshwar ,  Florian Wiesinger ,  Dattesh Dayanand Shanbhag

IPC: A61B6/00 ,  A61B6/04 ,  A61B6/02 ,  A61B5/00 ,  A61B5/055

CPC classification number: A61B6/5247 ,  A61B5/0035 ,  A61B5/0037 ,  A61B5/055 ,  A61B5/0555 ,  A61B5/7289 ,  A61B5/7292 ,  A61B6/02 ,  A61B6/037 ,  A61B6/04 ,  A61B6/0407 ,  A61B6/4417 ,  A61B6/461 ,  A61B6/469 ,  A61B6/488 ,  A61B6/541 ,  A61B6/542 ,  A61B6/545 ,  A61B6/548

Abstract: Exemplary embodiments of the present disclosure are directed to scheduling positron emission tomography (PET) scans for a combined PET-MRI scanner based on an acquisition of MR scout images of a subject. An anatomy and orientation of the subject can be determined based on the MR scout images and the schedule for acquiring PET scans of the subject can be determined from the anatomy of the subject. The schedule generated using exemplary embodiments of the present disclosure can specify a sequence of bed positions, scan durations at each bed position, and whether respiratory gating will be used at one or more of the bed positions.

Abstract translation: 本公开的示例性实施例涉及基于对受试者的MR侦察图像的获取来调度用于组合的PET-MRI扫描仪的正电子发射断层摄影（PET）扫描。 可以基于MR侦察图像来确定被摄体的解剖结构和取向，并且可以从对象的解剖结构确定用于获取对象的PET扫描的时间表。 使用本公开的示例性实施例产生的日程表可以指定床位置的顺序，每个床位置的扫描持续时间，以及是否在一个或多个床位置使用呼吸门控。