公开(公告)号：WO2015126736A1

公开(公告)日：2015-08-27

申请号：PCT/US2015/015741

申请日：2015-02-13

Applicant: SIEMENS AKTIENGESELLSCHAFT ,  SIEMENS CORPORATION

Inventor： ZHENG, Yefeng ,  LU, Shiyang ,  HUANG, Xiaojie

IPC: G06T7/00

CPC classification number: G06K9/4604 ,  G06K9/46 ,  G06K9/52 ,  G06K9/6267 ,  G06K2009/4666 ,  G06T7/12 ,  G06T7/143 ,  G06T7/162 ,  G06T2207/10072 ,  G06T2207/10081 ,  G06T2207/10088 ,  G06T2207/20072 ,  G06T2207/20076 ,  G06T2207/20081 ,  G06T2207/30048 ,  G06T2207/30101 ,  G06T2207/30172

Abstract: The coronary sinus or other vessel is segmented (16) by finding (14) a centerline and then using (30) the centerline to locate the boundary of the vessel. For finding the centerline, a refinement process uses multi-scale sparse appearance learning (22). For locating the boundary, the lumen is segmented as a graph cut problem (36).

Abstract translation: 通过找到（14）中心线，然后使用（30）中心线来定位血管的边界，将冠状窦或其他血管分段（16）（16）。 为了找到中心线，细化过程使用多尺度稀疏外观学习（22）。 为了定位边界，管腔被分割为图形切割问题（36）。

公开(公告)号：WO2015058044A1

公开(公告)日：2015-04-23

申请号：PCT/US2014/061063

申请日：2014-10-17

Applicant: SIEMENS AKTIENGESELLSCHAFT ,  SIEMENS CORPORATION

Inventor： SHARMA, Puneet ,  KAMEN, Ali ,  GEORGESCU, Bogdan ,  SAUER, Frank ,  COMANICIU, Dorin ,  ZHENG, Yefeng ,  NGUYEN, Hien ,  SINGH, Vivek Kumar

IPC: A61B6/00 ,  G06F19/00 ,  G06T7/00 ,  G06T7/20

CPC classification number: A61B5/026 ,  A61B5/0261 ,  A61B5/0263 ,  A61B5/7264 ,  A61B5/7267 ,  A61B5/7282 ,  A61B6/032 ,  A61B6/504 ,  A61B6/507 ,  A61B6/5217 ,  A61B6/563 ,  A61B8/06 ,  A61B8/0891 ,  A61B8/12 ,  A61B8/5223 ,  A61B8/565 ,  G06K9/46 ,  G06K9/6256 ,  G06T7/0012 ,  G06T7/0016 ,  G06T7/20 ,  G06T2207/10081 ,  G06T2207/10088 ,  G06T2207/10101 ,  G06T2207/10132 ,  G06T2207/20081 ,  G06T2207/30048 ,  G06T2207/30104 ,  G06T2211/404 ,  G16H50/20 ,  G16H50/30 ,  G16H50/50

Abstract: A method and system for determining fractional flow reserve (FFR) for a coronary artery stenosis of a patient is disclosed. In one embodiment, medical image data of the patient including the stenosis is received, a set of features for the stenosis is extracted from the medical image data of the patient, and an FFR value for the stenosis is determined based on the extracted set of features using a trained machine-learning based mapping. In another embodiment, a medical image of the patient including the stenosis of interest is received, image patches corresponding to the stenosis of interest and a coronary tree of the patient are detected, an FFR value for the stenosis of interest is determined using a trained deep neural network regressor applied directly to the detected image patches

Abstract translation: 公开了一种用于确定患者冠状动脉狭窄的分数流量储备（FFR）的方法和系统。 在一个实施例中，接收包括狭窄的患者的医学图像数据，从患者的医学图像数据中提取用于狭窄的一组特征，并且基于所提取的特征集来确定狭窄的FFR值 使用训练有素的基于机器学习的映射。 在另一个实施例中，接收到包括感兴趣狭窄的患者的医学图像，检测到与感兴趣的狭窄相对应的图像斑块和患者的冠状动脉树，使用训练有素的深度来确定感兴趣狭窄的FFR值 神经网络回归器直接应用于检测到的图像补丁

公开(公告)号：WO2016025189A1

公开(公告)日：2016-02-18

申请号：PCT/US2015/043057

申请日：2015-07-31

Applicant: SIEMENS AKTIENGESELLSCHAFT ,  SIEMENS CORPORATION

Inventor： NGUYEN, Hien ,  SINGH, Vivek Kumar ,  ZHENG, Yefeng ,  GEORGESCU, Bogdan ,  COMANICIU, Dorin ,  ZHOU, Shaohua Kevin

IPC: G06K9/46 ,  G06K9/62 ,  G06K9/00

CPC classification number: G06K9/6256 ,  G06K9/00624 ,  G06K9/00771 ,  G06K9/4628 ,  G06K9/6255

Abstract: Object detection (58) uses a deep or multiple layer network (72-80) to learn features for detecting (58) the object in the image. Multiple features from different layers are aggregated (46) to train (48) a classifier for the object. In addition or as an alternative to feature aggregation from different layers, an initial layer (72) may have separate learnt nodes for different regions of the image (70) to reduce the number of free parameters. The object detection (58) is learned or a learned object detector is applied.

Abstract translation: 对象检测（58）使用深层或多层网络（72-80）来学习用于检测（58）图像中的对象的特征。 来自不同层的多个特征被聚合（46）以训练（48）对象的分类器。 另外或作为来自不同层的特征聚合的替代，初始层（72）可以具有用于图像（70）的不同区域的单独的学习节点，以减少自由参数的数量。 学习对象检测（58）或应用学习对象检测器。

公开(公告)号：WO2016182551A1

公开(公告)日：2016-11-17

申请号：PCT/US2015/030084

申请日：2015-05-11

Applicant: SIEMENS HEALTHCARE GMBH ,  SIEMENS CORPORATION

Inventor： LIU, David ,  GEORGESCU, Bogdan ,  ZHENG, Yefeng ,  NGUYEN, Hien ,  ZHOU, Shaohua Kevin ,  SINGH, Vivek Kumar ,  COMANICIU, Dorin

IPC: G06K9/46 ,  G06K9/62

CPC classification number: G06K9/4628 ,  A61B8/52 ,  G06K9/627 ,  G06K2209/05 ,  G06T7/0012

Abstract: A method and system for anatomical landmark detection in medical images using deep neural networks is disclosed. For each of a plurality of image patches centered at a respective one of a plurality of voxels in the medical image, a subset of voxels within the image patch is input to a trained deep neural network based on a predetermined sampling pattern. A location of a target landmark in the medical image is detected using the trained deep neural network based on the subset of voxels input to the trained deep neural network from each of the plurality of image patches.

Abstract translation: 公开了一种使用深层神经网络的医学图像中解剖学标记检测的方法和系统。 对于以医学图像中的多个体素的相应一个为中心的多个图像补丁中的每一个，基于预定的采样模式，将图像补片内的体素子集输入训练的深层神经网络。 使用经训练的深层神经网络，基于从多个图像块中的每一个输入到经过训练的深层神经网络的体素子集来检测医学图像中的目标地标的位置。