知搜-全球专利检索

  1. 登录
  2. 注册

搜索结果

41 条记录 (耗时 0.03 秒)

    Detection of 3D pose of a TEE probe in x-ray medical imaging
    3.
    发明授权
    Detection of 3D pose of a TEE probe in x-ray medical imaging 有权

    公开(公告)号：US10515449B2

    公开(公告)日：2019-12-24

    申请号：US15344210

    申请日：2016-11-04

    申请人： Siemens Medical Solutions USA, Inc.

    发明人： Shun Miao ,  Rui Liao ,  Ryan Spilker

    IPC分类号： A61B5/05 ,  G06T7/00 ,  A61B8/00 ,  G06T7/73 ,  A61B6/12 ,  A61B6/00 ,  A61B8/12

    摘要： Pose of a probe is detected in x-ray medical imaging. Since the TEE probe is inserted through the esophagus of a patient, the pose is limited to being within the esophagus. The path of the esophagus is determined from medical imaging prior to the intervention. During the intervention, the location in 2D is found from one x-ray image at a given time. The 3D probe location is provided by assigning the depth of the esophagus at that 2D location to be the depth of the probe. A single x-ray image may be used to determine the probe location in 3D, allowing for real-time pose determination without requiring space to rotate a C-arm during the intervention.

    Method for 2-D/3-D registration based on hierarchical pose regression
    5.
    发明申请
    Method for 2-D/3-D registration based on hierarchical pose regression 审中-公开
    基于层次姿态回归的2-D / 3-D注册方法

    公开(公告)号：US20170024634A1

    公开(公告)日：2017-01-26

    申请号：US15207033

    申请日：2016-07-11

    申请人： Siemens Medical Solutions USA, Inc. ,  The University of British Columbia

    发明人： Shun Miao ,  Rui Liao ,  Zhen Wang

    IPC分类号： G06K9/62 ,  G06N3/08 ,  G06N3/04 ,  G06K9/46 ,  G06T7/00

    CPC分类号： G06K9/6277 ,  G06K9/4671 ,  G06N3/04 ,  G06N3/08 ,  G06T7/33 ,  G06T2200/04 ,  G06T2207/10081 ,  G06T2207/10088 ,  G06T2207/10116 ,  G06T2207/10124 ,  G06T2207/20021 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30052

    摘要： A method and apparatus for convolutional neural network (CNN) regression based 2D/3D registration of medical images is disclosed. A parameter space zone is determined based on transformation parameters corresponding to a digitally reconstructed radiograph (DRR) generated from the 3D medical image. Local image residual (LIR) features are calculated from local patches of the DRR and the X-ray image based on a set of 3D points in the 3D medical image extracted for the determined parameter space zone. Updated transformation parameters are calculated based on the LIR features using a hierarchical series of regressors trained for the determined parameter space zone. The hierarchical series of regressors includes a plurality of regressors each of which calculates updates for a respective subset of the transformation parameters.

    摘要翻译： 公开了一种用于基于卷积神经网络（CNN）回归的2D / 3D医学图像配准的方法和装置。 基于与从3D医学图像生成的数字重构X射线照片（DRR）对应的变换参数来确定参数空间区域。 基于针对确定的参数空间区域提取的3D医学图像中的一组3D点，从DRR和X射线图像的局部斑块计算局部图像残差（LIR）特征。 基于LIR特征使用针对确定的参数空间区训练的分级序列回归计算更新的变换参数。 回归分级系列包括多个回归器，每个回归计算器计算变换参数的相应子集的更新。

    Unsupervised deformable registration for multi-modal images
    7.
    发明授权
    Unsupervised deformable registration for multi-modal images 有权

    公开(公告)号：US11158069B2

    公开(公告)日：2021-10-26

    申请号：US16428092

    申请日：2019-05-31

    申请人： Siemens Healthcare GmbH

    发明人： Bibo Shi ,  Chen Qin ,  Rui Liao ,  Tommaso Mansi ,  Ali Kamen

    IPC分类号： G06T7/30

    摘要： In order to reduce computation time and provide more accurate solutions for bi-directional, multi-modal image registration, a learning-based unsupervised multi-modal deformable image registration method that does not require any aligned image pairs or anatomical landmarks is provided. A bi-directional registration function is learned based on disentangled shape representation by optimizing a similarity criterion defined on both latent space and image space.

    Deep variational method for deformable image registration
    8.
    发明授权
    Deep variational method for deformable image registration 有权

    公开(公告)号：US10909416B2

    公开(公告)日：2021-02-02

    申请号：US16440215

    申请日：2019-06-13

    申请人： Siemens Healthcare GmbH

    发明人： Tommaso Mansi ,  Boris Mailhe ,  Rui Liao ,  Shun Miao

    IPC分类号： G06K9/00 ,  G06K9/62 ,  G06N20/20 ,  G06F17/18 ,  G06N3/08 ,  G06N5/04

    摘要： A correspondence between a source image and a reference image is determined. A generative model corresponds to a prior probability distribution of deformation fields, each deformation field corresponding to a respective coordinate transformation. A conditional model generates a style transfer probability distribution of reference images, given a source image and a deformation field. The first image data is the source image, and the second image data is the reference image. An initial first deformation field is determined. An update process is iteratively performed until convergence to update the first deformation field, to generate a converged deformation field representing the correspondence between the source image and the reference image. The update process includes: determining a change in one or more characteristics of the first deformation field to increase a posterior probability density associated with the first deformation field, given the source image and reference image; and changing the one or more characteristics in accordance with the determined change.

    Dilated fully convolutional network for multi-agent 2D/3D medical image registration
    9.
    发明授权
    Dilated fully convolutional network for multi-agent 2D/3D medical image registration 有权

    公开(公告)号：US10818019B2

    公开(公告)日：2020-10-27

    申请号：US16103196

    申请日：2018-08-14

    申请人： Siemens Healthcare GmbH

    发明人： Sebastien Piat ,  Shun Miao ,  Rui Liao ,  Tommaso Mansi ,  Jiannan Zheng

    IPC分类号： G06T7/33 ,  G06T15/08 ,  G06K9/62 ,  G06K9/32 ,  G06T19/20

    摘要： A method and system for 3D/3D medical image registration. A digitally reconstructed radiograph (DRR) is rendered from a 3D medical volume based on current transformation parameters. A trained multi-agent deep neural network (DNN) is applied to a plurality of regions of interest (ROIs) in the DRR and a 2D medical image. The trained multi-agent DNN applies a respective agent to each ROI to calculate a respective set of action-values from each ROI. A maximum action-value and a proposed action associated with the maximum action value are determined for each agent. A subset of agents is selected based on the maximum action-values determined for the agents. The proposed actions determined for the selected subset of agents are aggregated to determine an optimal adjustment to the transformation parameters and the transformation parameters are adjusted by the determined optimal adjustment. The 3D medical volume is registered to the 2D medical image using final transformation parameters resulting from a plurality of iterations.

    UNSUPERVISED DEFORMABLE REGISTRATION FOR MULTI-MODAL IMAGES
    10.
    发明申请
    UNSUPERVISED DEFORMABLE REGISTRATION FOR MULTI-MODAL IMAGES 审中-公开

    公开(公告)号：US20200184660A1

    公开(公告)日：2020-06-11

    申请号：US16428092

    申请日：2019-05-31

    申请人： Siemens Healthcare GmbH

    发明人： Bibo Shi ,  Chen Qin ,  Rui Liao ,  Tommaso Mansi ,  Ali Kamen

    IPC分类号： G06T7/30

    摘要： In order to reduce computation time and provide more accurate solutions for bi-directional, multi-modal image registration, a learning-based unsupervised multi-modal deformable image registration method that does not require any aligned image pairs or anatomical landmarks is provided. A bi-directional registration function is learned based on disentangled shape representation by optimizing a similarity criterion defined on both latent space and image space.