Method Of And A System For Adapting A Geometric Model Using Multiple Partial Transformations
    11.
    发明申请
    Method Of And A System For Adapting A Geometric Model Using Multiple Partial Transformations 有权
    用于使用多个部分变换来适应几何模型的系统的方法和系统

    公开(公告)号:US20090037154A1

    公开(公告)日:2009-02-05

    申请号:US12067841

    申请日:2006-09-20

    IPC分类号: G06F17/10

    摘要: The invention relates to a method (100) of adapting a geometric model to an image data comprising determining a first partial transformation for mapping a first part of the geometric model into the image data and a second partial transformation for mapping a second part of the geometric model into the image data. By determining the first partial transformation of the first part of the geometric model and the second partial transformation of the second part of the geometric model, the geometric model can assume more shapes and therefore can be more accurately adapted to an object comprised in the image data.

    摘要翻译: 本发明涉及一种将几何模型适应于图像数据的方法(100),包括确定用于将几何模型的第一部分映射到图像数据中的第一部分变换,以及用于映射几何的第二部分的第二部分变换 模型进入图像数据。 通过确定几何模型的第一部分的第一部分变换和几何模型的第二部分的第二部分变换,几​​何模型可以呈现更多的形状,因此可以更准确地适应包含在图像数据中的对象 。

    Model-based coronary centerline localization
    12.
    发明授权
    Model-based coronary centerline localization 有权
    基于模型的冠状动脉中心线定位

    公开(公告)号:US08160332B2

    公开(公告)日:2012-04-17

    申请号:US12443490

    申请日:2007-09-28

    IPC分类号: G06K9/00

    摘要: The invention relates to a system (100) for registering a vessel model with an image data set based on a joined model comprising a reference object model and the vessel model, the system comprising: a placement unit (110) for placing the joined model in a space of the image data set, thereby creating a placed joined model comprising a placed reference object model and a placed vessel model; a computation unit (120) for computing a deformation field based on a landmark displacement field comprising displacements of landmarks of the placed reference object model relative to corresponding landmarks in the image data set; a transformation unit (130) for transforming the placed joined model using the deformation field, thereby creating a transformed joined model comprising a transformed reference object model and a transformed vessel model; and a registration unit (140) for registering the transformed vessel model with the image data set based on modifying the transformed vessel model and optimizing an objective function of the modified transformed vessel model, wherein the objective function comprises a location-prior term based on a localization of the modified transformed vessel model relative to the transformed joined model. Hence, the system is arranged to model a vessel taking into account the localization of a vessel model relative to a reference anatomical structure described by a reference model.

    摘要翻译: 本发明涉及一种用于根据包括参考对象模型和血管模型的联合模型将具有图像数据集的血管模型登记的系统(100),所述系统包括:放置单元(110),用于将所连接的模型放置在 图像数据集合的空间,从而创建包括放置的参考对象模型和放置的容器模型的放置的连接模型; 计算单元(120),用于基于地标位移场计算变形场,所述地标位移场包括相对于所述图像数据组中的对应地标放置的所述参考对象模型的界标的位移; 变换单元(130),用于使用变形场来变换放置的接合模型,从而创建包括变换的参考对象模型和变换的容器模型的变换的连接模型; 以及注册单元(140),用于基于修改所述经变换的血管模型并优化所述经修改的转化的血管模型的目标函数,将经变换的血管模型与所述图像数据集进行登记,其中所述目标函数包括基于 修改的转化血管模型相对于转化的连接模型的定位。 因此,该系统被设置成考虑到血管模型相对于由参考模型描述的参考解剖结构的定位来建模血管。

    Method for facilitating post-processing of images using deformable meshes
    13.
    发明授权
    Method for facilitating post-processing of images using deformable meshes 有权
    使用可变形网格促进图像后处理的方法

    公开(公告)号:US08144950B2

    公开(公告)日:2012-03-27

    申请号:US12097855

    申请日:2006-12-14

    IPC分类号: G06K9/00

    摘要: Method and system for facilitating post-processing of images using deformable meshes in which a deformable mesh model of an object such as an organ is extended by attaching information thereon in order to simplify and/or facilitate a desired post-processing task so that the post-processing task effected when the mesh is applied to the same object in an additional image can expeditiously use this information. The information may be attached to the mesh after its creation, for example, upon segmentation of the same object in some training image. The post-processing task can therefore be performed automatically without user interaction upon segmentation of the object in the additional image. Information is encoded on the mesh by enumerating a list of triangles or vertices of the mesh which have to be considered in the subsequent, post-processing task and by optionally providing additional parameters defining the precise post-processing algorithm(s) to be used.

    摘要翻译: 使用可变形网格促进图像后处理的方法和系统,其中通过在其上附加信息来扩展诸如器官的对象的可变形网格模型,以简化和/或促进期望的后处理任务,使得该帖子 当将网格应用于附加图像中的同一对象时,处理任务即可快速使用此信息。 信息可以在其创建之后附加到网格上,例如,在一些训练图像中对相同对象进行分割时。 因此,在附加图像中的对象的分割之后,可以自动执行后处理任务,而无需用户交互。 通过枚举在后续的后处理任务中必须考虑的网格的三角形或顶点列表以及可选地提供定义要使用的精确后处理算法的附加参数,将信息编码在网格上。

    MODEL-BASED CORONARY CENTERLINE LOCALIZATION
    14.
    发明申请
    MODEL-BASED CORONARY CENTERLINE LOCALIZATION 有权
    基于模型的冠状动脉中心定位

    公开(公告)号:US20100046815A1

    公开(公告)日:2010-02-25

    申请号:US12443490

    申请日:2007-09-28

    IPC分类号: G06K9/00

    摘要: The invention relates to a system (100) for registering a vessel model with an image data set based on a joined model comprising a reference object model and the vessel model, the system comprising: a placement unit (110) for placing the joined model in a space of the image data set, thereby creating a placed joined model comprising a placed reference object model and a placed vessel model; a computation unit (120) for computing a deformation field based on a landmark displacement field comprising displacements of landmarks of the placed reference object model relative to corresponding landmarks in the image data set; a transformation unit (130) for transforming the placed joined model using the deformation field, thereby creating a transformed joined model comprising a transformed reference object model and a transformed vessel model; and a registration unit (140) for registering the transformed vessel model with the image data set based on modifying the transformed vessel model and optimizing an objective function of the modified transformed vessel model, wherein the objective function comprises a location-prior term based on a localization of the modified transformed vessel model relative to the transformed joined model. Hence, the system is arranged to model a vessel taking into account the localization of a vessel model relative to a reference anatomical structure described by a reference model.

    摘要翻译: 本发明涉及一种用于根据包括参考对象模型和血管模型的联合模型将具有图像数据集的血管模型登记的系统(100),该系统包括:一个放置单元(110),用于将连接的模型放置在 图像数据集合的空间,从而创建包括放置的参考对象模型和放置的容器模型的放置的连接模型; 计算单元(120),用于基于地标位移场计算变形场,所述地标位移场包括相对于所述图像数据组中的对应地标放置的所述参考对象模型的界标的位移; 变换单元(130),用于使用变形场来变换放置的接合模型,从而创建包括变换的参考对象模型和变换的容器模型的变换的连接模型; 以及注册单元(140),用于基于修改所述经变换的血管模型并优化所述经修改的转化的血管模型的目标函数,将经变换的血管模型与所述图像数据集进行登记,其中所述目标函数包括基于 修改的转化血管模型相对于转化的连接模型的定位。 因此,该系统被设置成考虑到血管模型相对于由参考模型描述的参考解剖结构的定位来建模血管。

    APPARATUS AND METHOD FOR REMOVING A COVERING FILM FROM A CONTAINER
    17.
    发明申请
    APPARATUS AND METHOD FOR REMOVING A COVERING FILM FROM A CONTAINER 审中-公开
    从容器中移除覆盖膜的装置和方法

    公开(公告)号:US20090208316A1

    公开(公告)日:2009-08-20

    申请号:US12278897

    申请日:2007-01-03

    IPC分类号: B65B69/00

    摘要: An apparatus and a method are proposed for removing a covering film from a container. The apparatus has at least one gripping apparatus and a mechanism for producing a predetermined breaking point on a container. After production of the predetermined breaking point, the gripping apparatus grips a gripping face which is delimited by the predetermined breaking point, for removing the covering film from the container, and separates the gripping face during the removal of the covering film from the container.

    摘要翻译: 提出了从容器中除去覆盖膜的装置和方法。 该装置具有至少一个夹持装置和用于在容器上产生预定断裂点的机构。 在产生预定的断裂点之后,夹持装置夹持由预定断裂点限定的夹持面,用于从容器移除覆盖膜,并且在从容器移除覆盖膜期间分离夹持面。

    Intelligent speech recognition with user interfaces
    18.
    发明申请
    Intelligent speech recognition with user interfaces 有权
    智能语音识别与用户界面

    公开(公告)号:US20070083366A1

    公开(公告)日:2007-04-12

    申请号:US10576329

    申请日:2004-10-13

    申请人: Jochen Peters

    发明人: Jochen Peters

    IPC分类号: G10L15/26

    CPC分类号: G10L15/26 G10L2015/225

    摘要: The invention relates to a method, an apparatus, a Graphical User Interface and a computer program product for transforming speech to text. In a speech recognition step (100) a text is provided to which an automatic text modification based on text modification rules is applied (102). Text portions to which the modification applies, are highlighted in the text (106) and at least one suggestion for the modification is displayed to a user (110), being enabled to select one of the suggestions and therefore having control over the automatic text modification process (112). For example the rules identify text portions relating to ambiguous speech or missing text formatting characters that might have been misinterpreted in the speech recognition step.

    摘要翻译: 本发明涉及一种将语音转换为文本的方法,装置,图形用户接口和计算机程序产品。 在语音识别步骤(100)中,提供了基于文本修改规则的自动文本修改的文本(102)。 修改应用的文本部分在文本(106)中突出显示,并且至少一个用于修改的建议被显示给用户(110),使得能够选择其中一个建议,因此具有对自动文本修改的控制 过程(112)。 例如,规则识别与语音识别步骤中可能被误解的模糊语音或缺失的文本格式化字符有关的文本部分。

    Text segmentation and label assignment with user interaction by means of topic specific language models and topic-specific label statistics
    19.
    发明授权
    Text segmentation and label assignment with user interaction by means of topic specific language models and topic-specific label statistics 有权
    通过主题特定语言模型和主题特定标签统计信息,通过用户交互进行文本分割和标签分配

    公开(公告)号:US08688448B2

    公开(公告)日:2014-04-01

    申请号:US13619972

    申请日:2012-09-14

    IPC分类号: G10L15/00

    摘要: The invention relates to a method, a computer program product, a segmentation system and a user interface for structuring an unstructured text by making use of statistical models trained on annotated training data. The method performs text segmentation into text sections and assigns labels to text sections as section headings. The performed segmentation and assignment is provided to a user for general review. Additionally, alternative segmentations and label assignments are provided to the user being capable to select alternative segmentations and alternative labels as well as to enter a user defined segmentation and user defined label. In response to the modifications introduced by the user, a plurality of different actions are initiated incorporating the re-segmentation and re-labeling of successive parts of the document or the entire document.

    摘要翻译: 本发明涉及通过利用在注释训练数据上训练的统计模型来构造非结构化文本的方法,计算机程序产品,分割系统和用户界面。 该方法执行文本分段到文本部分,并将标签分配给文本部分作为标题。 执行的分割和分配被提供给用户进行一般审查。 此外,替代分割和标签分配被提供给能够选择替代分割和替代标签以及输入用户定义的分割和用户定义标签的用户。 响应于用户引入的修改,启动了多个不同的动作,其中包括文档或整个文档的连续部分的重新分割和重新标记。

    TEXT SEGMENTATION AND LABEL ASSIGNMENT WITH USER INTERACTION BY MEANS OF TOPIC SPECIFIC LANGUAGE MODELS AND TOPIC-SPECIFIC LABEL STATISTICS
    20.
    发明申请
    TEXT SEGMENTATION AND LABEL ASSIGNMENT WITH USER INTERACTION BY MEANS OF TOPIC SPECIFIC LANGUAGE MODELS AND TOPIC-SPECIFIC LABEL STATISTICS 有权
    用主题特定语言模型和主题特定标签统计的用户交互的文本分段和标签分配

    公开(公告)号:US20130066625A1

    公开(公告)日:2013-03-14

    申请号:US13619972

    申请日:2012-09-14

    IPC分类号: G06F17/27

    摘要: The invention relates to a method, a computer program product, a segmentation system and a user interface for structuring an unstructured text by making use of statistical models trained on annotated training data. The method performs text segmentation into text sections and assigns labels to text sections as section headings. The performed segmentation and assignment is provided to a user for general review. Additionally, alternative segmentations and label assignments are provided to the user being capable to select alternative segmentations and alternative labels as well as to enter a user defined segmentation and user defined label. In response to the modifications introduced by the user, a plurality of different actions are initiated incorporating the re-segmentation and re-labeling of successive parts of the document or the entire document.

    摘要翻译: 本发明涉及通过利用在注释训练数据上训练的统计模型来构造非结构化文本的方法,计算机程序产品,分割系统和用户界面。 该方法执行文本分段到文本部分,并将标签分配给文本部分作为标题。 执行的分割和分配被提供给用户进行一般审查。 此外,替代分割和标签分配被提供给能够选择替代分割和替代标签以及输入用户定义的分割和用户定义标签的用户。 响应于用户引入的修改,启动了多个不同的动作,其中包括文档或整个文档的连续部分的重新分割和重新标记。