公开(公告)号：US06836558B2

公开(公告)日：2004-12-28

申请号：US09816217

申请日：2001-03-26

申请人： Kunio Doi ,  Hidetaka Arimura ,  Shigehiko Katsuragawa ,  Junji Morishita

发明人： Kunio Doi ,  Hidetaka Arimura ,  Shigehiko Katsuragawa ,  Junji Morishita

IPC分类号： G06K900

CPC分类号： G06K9/6203 ,  G06T7/0002 ,  G06T7/001 ,  G06T7/32 ,  G06T2207/30061

摘要： A method, system and computer readable medium for a computer-automated method for identifying given image data, including obtaining template image data corresponding to said given image data; calculating correlation values between the given image data and said template image data; and identifying said image data based on the correlation values calculated in the calculating step.

摘要翻译： 一种用于识别给定图像数据的计算机自动化方法的方法，系统和计算机可读介质，包括获得对应于所述给定图像数据的模板图像数据; 计算给定图像数据和所述模板图像数据之间的相关值; 以及基于在所述计算步骤中计算的相关值来识别所述图像数据。

公开(公告)号：US6011862A

公开(公告)日：2000-01-04

申请号：US098504

申请日：1998-06-17

申请人： Kunio Doi ,  Xin-Wei Xu ,  Shigehiko Katsuragawa ,  Junji Morishita

发明人： Kunio Doi ,  Xin-Wei Xu ,  Shigehiko Katsuragawa ,  Junji Morishita

IPC分类号： A61B6/00 ,  G06T1/00 ,  G06T7/00 ,  G06T7/60 ,  G06K9/62

CPC分类号： G06T7/0012

摘要： A computerized method for the detection and characterization of disease in an image derived from a chest radiograph, wherein an image in the chest radiograph is processed to determine the ribcage boundary, including lung top edges, right and left ribcage edges, and right and left hemidiaphragm edges. Texture measures including RMS variations of pixel values within regions of interest are converted to relative exposures and corrected for system noise existing in the system used to produce the image. Texture and/or geometric pattern indices are produced. A histogram(s) of the produced index (indices) is produced and values of the histograms) are applied as inputs to a trained artificial neural network, which classifies the image as normal or abnormal. In one embodiment, obviously normal and obviously abnormal images are determined based on the ratio of abnormal regions of interest to the total number of regions of interest in a rule-based method, so that only difficult cases to diagnose are applied to the artificial neural network.

摘要翻译： 一种用于检测和表征来自胸部X光照片的图像中的疾病的计算机化方法，其中处理胸部X光片中的图像以确定胸腔边界，包括肺顶缘，右和左胸廓边缘，以及右侧和左侧膈肌 边缘。 包括感兴趣区域内的像素值的RMS变化的纹理度量被转换为相对曝光并且对用于产生图像的系统中存在的系统噪声进行校正。 产生纹理和/或几何图案索引。 生成的索引（索引）的直方图（直方图的值）被应用为训练的人造神经网络的输入，其将图像分类为正常或异常。 在一个实施例中，基于基于规则的方法，基于感兴趣的异常区域与感兴趣区域总数的比率来确定明显的正常和明显异常的图像，使得仅将困难的诊断情况应用于人造神经网络 。

公开(公告)号：US5790690A

公开(公告)日：1998-08-04

申请号：US428867

申请日：1995-04-25

申请人： Kunio Doi ,  Xin-Wei Xu ,  Shigehiko Katsuragawa ,  Junji Morishita

发明人： Kunio Doi ,  Xin-Wei Xu ,  Shigehiko Katsuragawa ,  Junji Morishita

IPC分类号： A61B6/00 ,  G06T1/00 ,  G06T7/00 ,  G06T7/60 ,  G06K9/00 ,  G06F15/00 ,  G06K9/46 ,  H04N1/40

CPC分类号： G06T7/0012

摘要： A computerized method for the detection and characterization of disease in an image derived from a chest radiograph, wherein an image in the chest radiograph is processed to determine the ribcage boundary, including lung top edges, right and left ribcage edges, and right and left hemidiaphragm edges. Texture measures including RMS variations of pixel values within regions of interest are converted to relative exposures and corrected for system noise existing in the system used to produce the image. Texture and/or geometric pattern indices are produced. A histogram(s) of the produced index (indices) is produced and values of the histogram(s) are applied as inputs to a trained artificial neural network, which classifies the image as normal or abnormal. In one embodiment, obviously normal and obviously abnormal images are determined based on the ratio of abnormal regions of interest to the total number of regions of interest in a rule-based method, so that only difficult cases to diagnose are applied to the artificial neural network.

摘要翻译： 一种用于检测和表征来自胸部X光照片的图像中的疾病的计算机化方法，其中处理胸部X光片中的图像以确定胸腔边界，包括肺顶缘，右和左胸廓边缘，以及右侧和左侧膈肌 边缘。 包括感兴趣区域内的像素值的RMS变化的纹理度量被转换为相对曝光并且对用于产生图像的系统中存在的系统噪声进行校正。 产生纹理和/或几何图案索引。 产生生成的索引（索引）的直方图，并将直方图的值作为输入被应用于训练的人造神经网络，其将图像分类为正常或异常。 在一个实施例中，基于基于规则的方法，基于感兴趣的异常区域与感兴趣区域总数的比率来确定明显的正常和明显异常的图像，使得仅将困难的诊断情况应用于人造神经网络 。

公开(公告)号：US07251353B2

公开(公告)日：2007-07-31

申请号：US10358337

申请日：2003-02-05

申请人： Kunio Doi ,  Junji Morishita ,  Shigehiko Katsuagawa

发明人： Kunio Doi ,  Junji Morishita ,  Shigehiko Katsuagawa

IPC分类号： G06K9/00

CPC分类号： G06T7/0012 ,  G06F19/321 ,  G06K9/00362 ,  G06T2207/30061

摘要： A method for determining whether a first medical image and a second medical image are medical images of the same patient, comprising selecting a first region in the first medical image; selecting a second region in the second medical image; determining a common region based on a boundary of the first region and a boundary of the second region; calculating a correlation coefficient based on image data from the first medical image in the common region and image data from the second medical image in the common region; and determining whether the first medical image and the second medical image are medical images of the same patient based on the correlation coefficient. Biological fingerprints from parts of chest radiographs such as thoracic fields, cardiac shadows, lung apices, superior mediastinum, and the right lower lung that includes the costophrenic angle, are used for the purpose of patient recognition and identification.

摘要翻译： 一种用于确定第一医学图像和第二医学图像是否是同一患者的医学图像的方法，包括选择第一医学图像中的第一区域; 选择所述第二医疗图像中的第二区域; 基于所述第一区域的边界和所述第二区域的边界来确定公共区域; 基于来自公共区域中的第一医用图像的图像数据和来自公共区域中的第二医用图像的图像数据计算相关系数; 以及基于所述相关系数来确定所述第一医用图像和所述第二医用图像是否是相同患者的医学图像。 用于患者识别和识别的目的，使用胸部X线照片如胸部，心脏阴影，肺顶，上纵隔和右下肺的部分生物指纹。