知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

73 records (took 0.024 seconds)

    HARDWARE SYSTEM DESIGN IMPROVEMENT USING DEEP LEARNING ALGORITHMS
    22.
    发明申请
    HARDWARE SYSTEM DESIGN IMPROVEMENT USING DEEP LEARNING ALGORITHMS 审中-公开

    公开(公告)号：US20180144243A1

    公开(公告)日：2018-05-24

    申请号：US15360042

    申请日：2016-11-23

    Applicant: General Electric Company

    Inventor： Jiang Hsieh ,  Gopal Avinash ,  Saad Sirohey

    IPC: G06N3/08 ,  G06F17/50

    CPC classification number: G06N3/08 ,  G06F11/30 ,  G06F17/5009 ,  G06N3/0454

    Abstract: Methods and apparatus for deep learning-based system design improvement are provided. An example system design engine apparatus includes a deep learning network (DLN) model associated with each component of a target system to be emulated, each DLN model to be trained using known input and known output, wherein the known input and known output simulate input and output of the associated component of the target system, and wherein each DLN model is connected as each associated component to be emulated is connected in the target system to form a digital model of the target system. The example apparatus also includes a model processor to simulate behavior of the target system and/or each component of the target system to be emulated using the digital model to generate a recommendation regarding a configuration of a component of the target system and/or a structure of the component of the target system.

    System and method for generating a CT slice image
    23.
    发明授权
    System and method for generating a CT slice image 有权

    公开(公告)号：US09940735B2

    公开(公告)日：2018-04-10

    申请号：US15185338

    申请日：2016-06-17

    Applicant: General Electric Company

    Inventor： Ping Liu ,  Jie Wu ,  Jiang Hsieh

    IPC: G06T11/00 ,  G06T7/00 ,  A61B6/02 ,  A61B6/03 ,  G06T7/11

    CPC classification number: G06T11/003 ,  A61B6/025 ,  A61B6/03 ,  A61B6/032 ,  A61B6/505 ,  A61B6/5223 ,  G06T7/0012 ,  G06T7/11 ,  G06T11/008 ,  G06T15/08 ,  G06T2207/10012 ,  G06T2207/10081 ,  G06T2207/30196

    Abstract: The present invention provides a system and method for generating a CT slice image. The system comprises an MIP image generation module, a region of interest determination module, an angle setting module, a curve determination module, a match module and a slice generation module. The MIP image generation module generates MIP images of a reconstructed image; the region of interest determination module determines an image range in an original slice, and determines the parts of the MIP images within the image range as regions of interest; the angle setting module rotates the regions of interest to a plurality of specific angles for a plurality of times; the curve determination module generates a plurality of two-dimensional projected curves of the regions of interest for the plurality of specific angles; the match module selects a two-dimensional projected curve matching with a part to be diagnosed based on features of the plurality of two-dimensional projected curves; the slice generation module determines a slice position range and a slice angle based on the features of the matched curve and the corresponding specific angle.

    APPARATUS AND METHOD FOR BEAM HARDENING ARTIFACT CORRECTION OF CT IMAGE
    25.
    发明申请
    APPARATUS AND METHOD FOR BEAM HARDENING ARTIFACT CORRECTION OF CT IMAGE 审中-公开

    公开(公告)号：US20170091932A1

    公开(公告)日：2017-03-30

    申请号：US15281396

    申请日：2016-09-30

    Applicant: General Electric Company

    Inventor： Ximiao Cao ,  Jiang Hsieh ,  Xueli Wang

    IPC: G06T7/00 ,  G06T5/00

    CPC classification number: G06T7/0012 ,  G06T5/005 ,  G06T11/003 ,  G06T2207/10081 ,  G06T2207/30008

    Abstract: The present invention provides an apparatus and method for beam hardening artifact correction of CT image, comprising a bone tissue image obtain module, a first correction module, an orthographic projection module, and a correction image obtaining module. The bone tissue image obtain module is used to extract a bone tissue image from a reconstructed original image; the first correction module is used to increase a current CT value of the bone tissue image; the orthographic projection module is used to perform an orthographic projection on the bone tissue image with the CT value being increased to obtain an orthographic projection data of the bone tissue image; the correction image obtaining module is used to perform image reconstruction according to the orthographic projection data of the bone tissue image described above and obtain a correction image.

    SYSTEM AND METHOD FOR REDUCING HIGH DENSITY ARTIFACTS IN COMPUTED TOMOGRAPHY IMAGING
    28.
    发明申请
    SYSTEM AND METHOD FOR REDUCING HIGH DENSITY ARTIFACTS IN COMPUTED TOMOGRAPHY IMAGING 有权
    用于减少计算机图像成像中的高密度伪像的系统和方法

    公开(公告)号：US20140328450A1

    公开(公告)日：2014-11-06

    申请号：US13875369

    申请日：2013-05-02

    Applicant: GENERAL ELECTRIC COMPANY

    Inventor： Debashish Pal ,  Jiang Hsieh ,  Paavana Sainath ,  Girijesh Yadava ,  Brian Edward Nett

    IPC: G01N23/04

    CPC classification number: G01N23/046 ,  A61B6/032 ,  A61B6/482 ,  A61B6/488 ,  A61B6/5258 ,  A61B6/545 ,  G01V5/005

    Abstract: An imaging system includes a rotatable gantry for receiving an object to be scanned, a generator configured to energize an x-ray source to generate x-rays, a detector positioned to receive the x-rays that pass through the object, and a computer. The computer is programmed to obtain knowledge of a metal within the object, scan the object using system scanning parameters, reconstruct an image of the object using a reconstruction algorithm, and automatically select at least one of the system scanning parameters and the reconstruction algorithm based on the obtained knowledge.

    Abstract translation: 成像系统包括用于接收待扫描物体的可旋转机架，被配置为激励X射线源以产生X射线的发生器，定位成接收穿过物体的x射线的检测器以及计算机。 计算机被编程以获得对象内的金属的知识，使用系统扫描参数扫描对象，使用重建算法重构对象的图像，并且基于以下方式自动选择系统扫描参数和重建算法中的至少一个： 获得知识。

    System and method for multi-material correction of image data
    29.
    发明授权
    System and method for multi-material correction of image data 有权
    图像数据多物质校正的系统和方法

    公开(公告)号：US08811709B2

    公开(公告)日：2014-08-19

    申请号：US13677010

    申请日：2012-11-14

    Applicant: General Electric Company

    Inventor： Xiaoye Wu ,  Jiang Hsieh ,  Paavana Sainath ,  Dan Xu ,  Yannan Jim ,  Girijesh Kumar Yadava ,  Adam Israel Cohen ,  Hewei Gao

    IPC: G06K9/00 ,  A61B6/00

    CPC classification number: A61B6/03 ,  G06T7/0012 ,  G06T11/008 ,  G06T2211/408

    Abstract: A method is provided. The method includes acquiring projection data of an object from a plurality of pixels, reconstructing the acquired projection data from the plurality of pixels into a reconstructed image, performing material characterization and decomposition of an image volume of the reconstructed image to reduce a number of materials analyzed in the image volume to two basis materials. The method also includes generating a re-mapped image volume for at least one basis material of the two basis materials, and performing forward projection on at least the re-mapped image volume for the at least one basis material to produce a material-based projection. The method further includes generating multi-material corrected projections based on the material-based projection and a total projection attenuated by the object, which represents both of the two basis materials, wherein the multi-material corrected projections include linearized projections.

    Abstract translation: 提供了一种方法。 该方法包括从多个像素获取对象的投影数据，将所获取的投影数据从多个像素重建成重建图像，对重建图像执行材料表征和分解，以减少分析的材料数量 在图像体积为两个基础材料。 所述方法还包括为所述两种基础材料的至少一种基础材料生成重新映射的图像体积，以及针对所述至少一种基础材料至少对所述重新映射的图像体积执行向前投影以产生基于材料的投影 。 该方法还包括基于基于材料的投影和由对象衰减的总投影来生成多材料校正投影，其表示两种基础材料，其中多材料校正投影包括线性化投影。

Patent Agency Ranking