知搜-全球专利检索

  1. 登录
  2. 注册

搜索结果

19 条记录 (耗时 0.016 秒)

    Automatic liver segmentation using adversarial image-to-image network
    4.
    发明授权
    Automatic liver segmentation using adversarial image-to-image network 有权

    公开(公告)号：US10600185B2

    公开(公告)日：2020-03-24

    申请号：US15877805

    申请日：2018-01-23

    申请人： Siemens Healthcare GmbH

    发明人： Dong Yang ,  Daguang Xu ,  Shaohua Kevin Zhou ,  Bogdan Georgescu ,  Mingqing Chen ,  Dorin Comaniciu

    IPC分类号： G06T7/11 ,  G06T7/187 ,  G06T7/143 ,  G06T7/00

    摘要： A method and apparatus for automated liver segmentation in a 3D medical image of a patient is disclosed. A 3D medical image, such as a 3D computed tomography (CT) volume, of a patient is received. The 3D medical image of the patient is input to a trained deep image-to-image network. The trained deep image-to-image network is trained in an adversarial network together with a discriminative network that distinguishes between predicted liver segmentation masks generated by the deep image-to-image network from input training volumes and ground truth liver segmentation masks. A liver segmentation mask defining a segmented liver region in the 3D medical image of the patient is generated using the trained deep image-to-image network.

    Multiple landmark detection in medical images based on hierarchical feature learning and end-to-end training
    8.
    发明授权
    Multiple landmark detection in medical images based on hierarchical feature learning and end-to-end training 有权

    公开(公告)号：US10210613B2

    公开(公告)日：2019-02-19

    申请号：US15591157

    申请日：2017-05-10

    申请人： Siemens Healthcare GmbH

    发明人： Daguang Xu ,  Tao Xiong ,  David Liu ,  Shaohua Kevin Zhou ,  Mingqing Chen ,  Dorin Comaniciu

    IPC分类号： G06T7/00 ,  G06T7/70 ,  G06K9/00 ,  G06N3/08 ,  G06T7/73 ,  A61B5/00

    摘要： The present embodiments relate to detecting multiple landmarks in medical images. By way of introduction, the present embodiments described below include apparatuses and methods for detecting landmarks using hierarchical feature learning with end-to-end training. Multiple neural networks are provided with convolutional layers for extracting features from medical images and with a convolutional layer for learning spatial relationships between the extracted features. Each neural network is trained to detect different landmarks using a different resolution of the medical images, and the convolutional layers of each neural network are trained together with end-to-end training to learn appearance and spatial configuration simultaneously. The trained neural networks detect multiple landmarks in a test image iteratively by detecting landmarks at different resolutions, using landmarks detected a lesser resolutions to detect additional landmarks at higher resolutions.

    Method and system for machine learning based classification of vascular branches
    9.
    发明授权
    Method and system for machine learning based classification of vascular branches 有权

    公开(公告)号：US10115039B2

    公开(公告)日：2018-10-30

    申请号：US15446252

    申请日：2017-03-01

    申请人： Siemens Healthcare GmbH

    发明人： Mehmet Akif Gulsun ,  Yefeng Zheng ,  Gareth Funka-Lea ,  Mingqing Chen

    IPC分类号： G06K9/62 ,  G06K9/46

    摘要： A method and apparatus for learning based classification of vascular branches to distinguish falsely detected branches from true branches is disclosed. A plurality of overlapping fixed size branch segments are sampled from branches of a detected centerline tree of a target vessel extracted from a medical image of a patient. A plurality of 1D profiles are extracted along each of the overlapping fixed size branch segments. A probability score for each of the overlapping fixed size branch segments is calculated based on the plurality of 1D profiles extracted for each branch segment using a trained deep neural network classifier. The trained deep neural network classifier may be a convolutional neural network (CNN) trained to predict a probability of a branch segment being fully part of a target vessel based on multi-channel 1D input. A final probability score is assigned to each centerline point in the branches of the detected centerline tree based on the probability scores of the overlapping branch segments containing that centerline point. The branches of the detected centerline tree of the target vessel are pruned based on the final probability scores of the centerline points.