知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

17 records (took 0.016 seconds)

    Systems and methods for deep learning-based image reconstruction
    1.
    发明授权
    Systems and methods for deep learning-based image reconstruction 有权

    公开(公告)号：US11580677B2

    公开(公告)日：2023-02-14

    申请号：US16806727

    申请日：2020-03-02

    Applicant: General Electric Company

    Inventor： Giovanni John Jacques Palma ,  Razvan Iordache

    IPC: G06K9/00 ,  G06T11/00 ,  A61B6/00 ,  A61B6/02 ,  G06N3/04 ,  G06N3/08 ,  G06T7/33

    Abstract: Methods and systems for deep learning based image reconstruction are disclosed herein. An example method includes receiving a set of imaging projections data, identifying a voxel to reconstruct, receiving a trained regression model, and reconstructing the voxel. The voxel is reconstructed by: projecting the voxel on each imaging projection in the set of imaging projections according to an acquisition geometry, extracting adjacent pixels around each projected voxel, feeding the regression model with the extracted adjacent pixel data to produce a reconstructed value of the voxel, and repeating the reconstruction for each voxel to be reconstructed to produce a reconstructed image.

    Iterative X-ray imaging optimization method and system
    4.
    发明授权
    Iterative X-ray imaging optimization method and system 有权

    公开(公告)号：US10368822B2

    公开(公告)日：2019-08-06

    申请号：US14840675

    申请日：2015-08-31

    Applicant: General Electric Company

    Inventor： Giovanni John Jacques Palma ,  Razvan Gabriel Iordache ,  Serge Louis Wilfrid Muller ,  Laurence Vancamberg

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

    Abstract: A method of optimizing images of a patient utilizing a medical imaging device includes the steps of providing a medical imaging device having an x-ray source, an x-ray detector, a controller for adjusting the positions of the x-ray source and detector, an image reconstructor/generator connected to the x-ray detector to receive x-ray data and reconstruct an x-ray image, and a processor connected to the image reconstructor/generator and the controller to perform analyses on the x-ray image, acquiring a first data set S1 of images, processing the first data set S1 to reconstruct a first computerized data set D1, analyzing the first computerized data set D1, acquiring at least one additional data set Sn in response to the analysis of the first computerized data set D1 and processing the at least one additional data set Sn in combination with the first data set S1 to reconstruct an optimized computerized data set Dn.

    Method and system for performing a guided biopsy using digital tomosynthesis
    6.
    发明授权
    Method and system for performing a guided biopsy using digital tomosynthesis 有权

    公开(公告)号：US10568694B2

    公开(公告)日：2020-02-25

    申请号：US14693863

    申请日：2015-04-22

    Applicant: GENERAL ELECTRIC COMPANY

    Inventor： Giovanni John Jacques Palma ,  Remy Andre Klausz

    IPC: A61B6/12 ,  A61B34/10 ,  A61B6/00 ,  A61B90/00 ,  A61B10/02 ,  A61B6/04 ,  A61B6/02

    Abstract: A method and system for performing a biopsy guided by a 3D image of the object obtained from digital tomosynthesis is performed. The method includes applying a compression paddle to an object; performing a tomosynthesis scan; reconstructing at least a portion of the scan; locating a lesion using a displayed or reprojected marker to determine a location of the lesion; correlating the location of the lesion in the scan to a marker on the compression paddle or tool holder; and proposing a needle entry point of a biopsy tool based off of the correlation so that a needle or penetration device will effectively reach the lesion or target if it is inserted at the proposed entry point. A system for performing a biopsy is further disclosed. In one embodiment, the system comprises: a tomosynthesis imaging apparatus for performing a tomosynthesis scan comprising an x-ray source and an x-ray detector; at least one marker; a compression paddle; a controller; and a display screen for displaying at least a portion of a reproduction of a tomosynthesis scan generated by the imaging apparatus, wherein (i) the at least one marker is displayed on the reproduction, the marker being used to determine a lesion location or target, and (ii) the controller correlates the lesion location or target with the at least one marker and generates a proposed needle entry point based on the correlation so that a needle or penetration device will effectively reach the lesion or target if it is inserted at the proposed entry point.

    SYSTEMS AND METHODS FOR DEEP LEARNING-BASED IMAGE RECONSTRUCTION
    8.
    发明申请
    SYSTEMS AND METHODS FOR DEEP LEARNING-BASED IMAGE RECONSTRUCTION 审中-公开

    公开(公告)号：US20190102916A1

    公开(公告)日：2019-04-04

    申请号：US15720632

    申请日：2017-09-29

    Applicant: General Electric Company

    Inventor： Giovanni John Jacques Palma ,  Razvan Iordache

    IPC: G06T11/00 ,  A61B6/00 ,  A61B6/02 ,  G06N3/08 ,  G06T7/33 ,  G06N3/04

    Abstract: Methods and systems for deep learning based image reconstruction are disclosed herein. An example method includes receiving a set of imaging projections data, identifying a voxel to reconstruct, receiving a trained regression model, and reconstructing the voxel. The voxel is reconstructed by: projecting the voxel on each imaging projection in the set of imaging projections according to an acquisition geometry, extracting adjacent pixels around each projected voxel, feeding the regression model with the extracted adjacent pixel data to produce a reconstructed value of the voxel, and repeating the reconstruction for each voxel to be reconstructed to produce a reconstructed image.

    Interpolated tomosynthesis projection images
    10.
    发明授权
    Interpolated tomosynthesis projection images 有权

    公开(公告)号：US10157460B2

    公开(公告)日：2018-12-18

    申请号：US15334025

    申请日：2016-10-25

    Applicant: General Electric Company

    Inventor： Giovanni John Jacques Palma

    IPC: G06K9/00 ,  G06T7/00 ,  G06T11/00 ,  G06T15/08 ,  A61B6/00 ,  A61B6/02

    Abstract: Systems and methods of medical imaging includes acquiring a plurality of projection images. A first projection image and a second projection image from the plurality of projection images are selected that are adjacent to a received focal point. A first set of object locations in the first projection image and a second set of object locations in the second projection image are identified that contribute to a pixel of the synthetic projection image. A value for the pixel of the synthetic projection image is calculated from the pixels of the first set of object locations and the pixels of the second set of object locations. The synthetic projection image is created with the calculated value.

Patent Agency Ranking