公开(公告)号：US5666434A

公开(公告)日：1997-09-09

申请号：US536794

申请日：1995-09-29

申请人： Robert M. Nishikawa ,  Takehiro Ema ,  Hiroyuki Yoshida ,  Kunio Doi

发明人： Robert M. Nishikawa ,  Takehiro Ema ,  Hiroyuki Yoshida ,  Kunio Doi

IPC分类号： A61B6/00 ,  G06F19/00 ,  G06K9/52 ,  G06T7/00 ,  G06K9/00

CPC分类号： G06T7/0012 ,  A61B6/5258 ,  G06K9/527 ,  G06T2207/10116 ,  G06T2207/20064 ,  G06T2207/30068 ,  Y10S128/925

摘要： A method for automated detection of abnormal anatomic regions, wherein a mammogram is digitized to produce a digital image and the digital image is processed using local edge gradient analysis and linear pattern analysis in addition to feature extraction routines to identify abnormal anatomic regions. Noise reduction filtering and pit-filling/spike-removal filtering techniques are also provided. Multiple difference imaging techniques are also used in which difference images employing different filter characteristics are obtained and processing results logically OR'ed to identify abnormal anatomic regions. In another embodiment the processing results with and without noise reduction filtering are logically AND'ed to improve detection sensitivity. Also, in another embodiment the wavelet transform is utilized in the identification and detection of abnormal regions. The wavelet transform is preferably used in conjunction with the difference imaging technique with the results of the two techniques being logically OR'ed.

公开(公告)号：US5491627A

公开(公告)日：1996-02-13

申请号：US060531

申请日：1993-05-13

申请人： Wei Zhang ,  Kunio Doi

发明人： Wei Zhang ,  Kunio Doi

IPC分类号： A61B6/00 ,  G06F19/00 ,  G06T7/00

CPC分类号： G06T7/0012 ,  G06F19/345 ,  G06F19/321 ,  G06T2207/10116 ,  G06T2207/20084 ,  G06T2207/30068 ,  Y10S128/924 ,  Y10S128/925

摘要： A method and system for the detection of microcalcifications in digital mammograms. Digital mammograms are obtained and regions-of-interest (ROIs) are selected therefrom which contain suspected microcalcifications, either individual or clustered microcalcifications. The suspect ROIs are background-trend corrected, followed by Fourier transformation and power spectrum calculation to perform detection in the frequency domain. Detection can also be carried out in the spatial domain by omitting the Fourier transformation and power spectrum calculation. The ROI is then scaled for input into a neural network trained to detect microcalcifications. The neural network outputs ROIs with detected microcalcifications. The method and system can also include normalizing the background-trend corrected ROIs and imputing the normalized ROI to a shift-invariant neural network trained to detect microcalcifications. The output ROI of the shift-invariant neural network is thresholded to remove additional false positive detections, and then the thresholded ROI undergoes a cluster detection to detect clustered microcalcifications. Feature extraction techniques can be applied to the remaining ROI to remove additional false positive detections.

摘要翻译： 用于检测数字乳腺X线照片中微钙化的方法和系统。 获得数字乳腺X线照片，并从中选择感兴趣区域（ROI），其中包含怀疑的微钙化，单个或聚集的微钙化。 可疑ROI是背景趋势校正，随后进行傅里叶变换和功率谱计算，以在频域中进行检测。 也可以通过省略傅里叶变换和功率谱计算在空间域中进行检测。 然后，将ROI投影到经过训练以检测微钙化的神经网络中。 神经网络通过检测到的微钙化输出ROI。 该方法和系统还可以包括对背景趋势校正的ROI进行归一化，并将归一化的ROI估算为经过训练以检测微钙化的移位不变神经网络。 移位不变神经网络的输出ROI被阈值化以去除额外的假阳性检测，然后阈值ROI进行聚类检测以检测聚类微钙化。 特征提取技术可以应用于剩余的ROI以消除额外的假阳性检测。

公开(公告)号：US5463548A

公开(公告)日：1995-10-31

申请号：US53345

申请日：1993-04-28

申请人： Naoki Asada ,  Kunio Doi

发明人： Naoki Asada ,  Kunio Doi

IPC分类号： A61B5/00 ,  A61B6/00 ,  A61B10/00 ,  G06F19/00 ,  G06N3/04 ,  G06F159/00

CPC分类号： G06F19/345 ,  G06N3/0454 ,  G06F19/321 ,  G06F19/3431 ,  Y10S128/922 ,  Y10S128/924 ,  Y10S128/925

摘要： A method and system for computer-aided differential diagnosis of diseases, and in particular, computer-aided differential diagnosis using neural networks. A first embodiment of the neural network distinguishes between a plurality of interstitial lung diseases on the basis of inputted clinical parameters and radiographic information. A second embodiment distinguishes between malignant and benign mammographic cases based upon similar inputted clinical and radiographic information. The neural networks were first trained using a hypothetical data base made up of hypothetical cases for each of the interstitial lung diseases and for malignant and benign cases. The performance of the neural network was evaluated using receiver operating characteristics (ROC) analysis. The decision performance of the neural network was compared to experienced radiologists and achieved a high performance comparable to that of the experienced radiologists. The neural network according to the invention can be made up of a single network or a plurality of successive or parallel networks. The neural network according to the invention can also be interfaced to a computer which provides computerized automated lung texture analysis to supply radiographic input data in an objective and automated manner.

摘要翻译： 一种用于计算机辅助诊断疾病的方法和系统，特别是使用神经网络的计算机辅助鉴别诊断。 神经网络的第一实施例基于输入的临床参数和放射照相信息来区分多个间质性肺疾病。 第二实施例基于相似的输入的临床和放射学信息来区分恶性和良性乳房造影病例。 神经网络首先使用由假性病例组成的假性数据库进行训练，每个间质性肺疾病以及恶性和良性病例。 使用接收器操作特性（ROC）分析评估神经网络的性能。 将神经网络的决策性能与经验丰富的放射科医生进行比较，并获得与经验丰富的放射科医师相当的高性能。 根据本发明的神经网络可以由单个网络或多个连续或并行网络组成。 根据本发明的神经网络还可以与计算机接口，该计算机提供计算机化的自动肺结构分析，以客观和自动的方式提供射线照相输入数据。

公开(公告)号：US5359513A

公开(公告)日：1994-10-25

申请号：US981471

申请日：1992-11-25

申请人： Akiko Kano ,  Kunio Doi

发明人： Akiko Kano ,  Kunio Doi

IPC分类号： A61B6/00 ,  G06Q50/00 ,  G06T1/00 ,  G06T3/00 ,  G06T5/50 ,  G06T7/00 ,  G06F15/00 ,  G06K9/00

CPC分类号： G06T3/0081 ,  G06T5/50 ,  G06T2207/20224 ,  Y10S128/922

摘要： A computerized scheme for the analysis of digitized medical images which provides enhancement of interval changes occurring in a pair of temporally sequential images. In the method and apparatus of the invention, a pair of images are digitized and then are subjected to image registration including a nonlinear warping of one of the images so that corresponding locations in the two images are aligned with each other. Subsequent to image registration, a subtraction process is performed in order to generate a difference between the warped and unwarped images. In this manner, slight opacities which are only present in the later image may be detected based on the subtraction of the registered images.

摘要翻译： 一种用于分析数字化医学图像的计算机化方案，其提供在一对时间顺序图像中发生的间隔变化的增强。 在本发明的方法和装置中，一对图像被数字化，然后进行图像配准，其中包括一个图像的非线性翘曲，使得两个图像中的相应位置彼此对准。 在图像配准之后，执行减法处理，以便产生翘曲和未被绘制的图像之间的差异。 以这种方式，可以基于减少注册的图像来检测仅存在于稍后图像中的轻微不透明度。

公开(公告)号：US4918534A

公开(公告)日：1990-04-17

申请号：US184791

申请日：1988-04-22

申请人： Kwok L. Lam ,  Heang-Ping Chan ,  Kunio Doi ,  Kenichi Komatsu ,  Michitaka Honda

发明人： Kwok L. Lam ,  Heang-Ping Chan ,  Kunio Doi ,  Kenichi Komatsu ,  Michitaka Honda

IPC分类号： A61B6/00 ,  H04N5/235 ,  H04N5/238 ,  H04N5/32

CPC分类号： H04N5/32 ,  H04N5/2351 ,  H04N5/238

摘要： An optical image processing method and system in a radiographic system, wherein an image of an object is produced by an image intensifier and converted to video signal by a television (TV) camera system. A liquid crystal display (LCD) is placed bewteen the image intensifier and the T.V. camera system at a location shifted from a focal plane of the T.V. camera system. The liquid crystal display defines plural pixels which are controllable so that light produced by the image intensifier and impinging on the plural LCD pixels can be selectively attenuated by application of selected control signals to the LCD. Control signals are generated selectively based on predetermined parameters of the radiographic system and the image. These control signals are applied to the LCD to perform selective attenuation and corresponding dynamic range compression of light passing through respective LCD pixels, thereby to perform unsharp masking of the image produced by the image intensifier using the LCD.

摘要翻译： 一种放射线照相系统中的光学图像处理方法和系统，其中物体的图像由图像增强器产生并通过电视（TV）照相机系统转换为视频信号。 液晶显示器（LCD）将图像增强器和T.V.摄像机系统放置在从T.V.摄像机系统的焦平面偏移的位置处。 液晶显示器限定可控制的多个像素，使得可以通过将所选择的控制信号施加到LCD来选择性地衰减由图像增强器产生的光并撞击在多个LCD像素上的光。 基于射线照相系统和图像的预定参数选择性地产生控制信号。 这些控制信号被施加到LCD以对通过各个LCD像素的光进行选择性衰减和相应的动态范围压缩，从而对使用LCD的图像增强器产生的图像执行不清晰的掩蔽。

公开(公告)号：US4841555A

公开(公告)日：1989-06-20

申请号：US81001

申请日：1987-08-03

申请人： Kunio Doi ,  Yuichiro Kume

发明人： Kunio Doi ,  Yuichiro Kume

IPC分类号： A61B6/00 ,  G06T1/00 ,  G06T11/00

CPC分类号： G06T11/005

摘要： A method and system for reconstructing image data obtained from plural sequential exposures of an object by a radiation source through a multiple slit beam scanning assembly, wherein frames of original image data in the form of plural pixels are obtained for each discrete scanning step/exposure and a minimum pixel value of at least one frame is determined. Based on this determined minimum value, a cutoff value is determined. The pixel values of each frame are compared with the cutoff value, and the image is reconstructed by summation of all the pixel image data above the cutoff value for each respective pixel. In order to improve spatial resolution, the slit width of the scanning assembly is selected be the narrower than the pixel size. Pixel image data produced by exposure from radiation passing through the same slit of the scanning assembly in consecutive scanning steps are then assigned to adjacent pixels in a direction perpendicular to the scanning beam of an enlarged pixel matrix. To achieve a balanced aspect ratio, pixel values for pixels in a direction parallel to the scanning beam are assigned by interpolation between pixel values assigned in the perpendicular direction.

摘要翻译： 一种用于通过多缝光束扫描组件通过辐射源重建从物体的多次连续曝光获得的图像数据的方法和系统，其中为每个离散扫描步骤/曝光获得多个像素形式的原始图像数据的帧， 确定至少一帧的最小像素值。 基于该确定的最小值，确定截止值。 将每个帧的像素值与截止值进行比较，并通过对每个相应像素的截止值之上的所有像素图像数据求和来重建图像。 为了提高空间分辨率，选择扫描组件的狭缝宽度窄于像素尺寸。 然后在连续扫描步骤中通过扫描组件的相同狭缝的辐射曝光产生的像素图像数据在与放大的像素矩阵的扫描光束垂直的方向上分配给相邻像素。 为了实现平衡的长宽比，通过在垂直方向上分配的像素值之间的内插来分配与扫描光束平行的方向上的像素的像素值。

公开(公告)号：US07545965B2

公开(公告)日：2009-06-09

申请号：US10703617

申请日：2003-11-10

申请人： Kenji Suzuki ,  Kunio Doi

发明人： Kenji Suzuki ,  Kunio Doi

IPC分类号： G06K9/00

CPC分类号： G06T7/0012 ,  G06T5/007 ,  G06T5/30 ,  G06T2207/10081 ,  G06T2207/20016 ,  G06T2207/20084 ,  G06T2207/20216 ,  G06T2207/30061

摘要： A method, system, and computer program product for modifying an appearance of an anatomical structure in a medical image, e.g., rib suppression in a chest radiograph. The method includes: acquiring, using a first imaging modality, a first medical image that includes the anatomical structure; applying the first medical image to a trained image processing device to obtain a second medical image, corresponding to the first medical image, in which the appearance of the anatomical structure is modified; and outputting the second medical image. Further, the image processing device is trained using plural teacher images obtained from a second imaging modality that is different from the first imaging modality. In one embodiment, the method also includes processing the first medical image to obtain plural processed images, wherein each of the plural processed images has a corresponding image resolution; applying the plural processed images to respective multi-training artificial neural networks (MTANNs) to obtain plural output images, wherein each MTANN is trained to detect the anatomical structure at one of the corresponding image resolutions; and combining the plural output images to obtain a second medical image in which the appearance of the anatomical structure is enhanced.

摘要翻译： 一种用于修改医学图像中的解剖结构外观的方法，系统和计算机程序产品，例如胸部X光照片中的肋骨抑制。 该方法包括：使用第一成像模式获取包括解剖结构的第一医学图像; 将第一医学图像应用于训练有素的图像处理装置，以获得对应于第一医学图像的第二医学图像，其中解剖结构的外观被修改; 并输出第二医用图像。 此外，使用从与第一成像模式不同的第二成像模式获得的多个教师图像来训练图像处理装置。 在一个实施例中，该方法还包括处理第一医用图像以获得多个经处理的图像，其中多个处理图像中的每一个具有对应的图像分辨率; 将多个经处理的图像应用于相应的多训练人造神经网络（MTANN）以获得多个输出图像，其中训练每个MTANN以检测其中一个相应图像分辨率的解剖结构; 以及组合所述多个输出图像以获得其中所述解剖结构的外观增强的第二医用图像。

公开(公告)号：US20080298657A1

公开(公告)日：2008-12-04

申请号：US12094161

申请日：2006-11-22

申请人： Junji Shiraishi ,  Kunio Doi ,  Daniel Appelbaum ,  Yonglin Pu ,  Qiang Li

发明人： Junji Shiraishi ,  Kunio Doi ,  Daniel Appelbaum ,  Yonglin Pu ,  Qiang Li

IPC分类号： G06K9/00

CPC分类号： G06T7/0012 ,  G06T5/50 ,  G06T7/32 ,  G06T2207/30008

摘要： A method of producing an image to aid detection of a change in progress of a disease in a patient is described. In the method, a first image of a distribution of a radioisotope in the patient is obtained. A second image of the distribution of the radioisotope in the patient is also obtained. At least one of the first and second images are then normalized (1:140). One of the images is warped to match the other image using a multiple-segment matching method (1:160). The first image is subtracted from the second image to form a subtraction image (1:220). Finally, the resulting subtraction image is displayed.

摘要翻译： 描述了一种产生图像以帮助检测患者疾病进展变化的方法。 在该方法中，获得患者中放射性同位素分布的第一图像。 还获得了患者中放射性同位素分布的第二图像。 然后对第一和第二图像中的至少一个进行归一化（1：140）。 使用多段匹配方法（1：160），其中一个图像被扭曲以匹配其他图像。 从第二图像中减去第一图像以形成减影图像（1:220）。 最后，显示所得的减法图像。

公开(公告)号：US20050100208A1

公开(公告)日：2005-05-12

申请号：US10703617

申请日：2003-11-10

申请人： Kenji Suzuki ,  Kunio Doi

发明人： Kenji Suzuki ,  Kunio Doi

IPC分类号： G06K9/00 ,  G06K9/62 ,  G06T5/00 ,  G06T5/40 ,  G06T7/00

CPC分类号： G06T7/0012 ,  G06T5/007 ,  G06T5/30 ,  G06T2207/10081 ,  G06T2207/20016 ,  G06T2207/20084 ,  G06T2207/20216 ,  G06T2207/30061

摘要： A method, system, and computer program product for modifying an appearance of an anatomical structure in a medical image, e.g., rib suppression in a chest radiograph. The method includes: acquiring, using a first imaging modality, a first medical image that includes the anatomical structure; applying the first medical image to a trained image processing device to obtain a second medical image, corresponding to the first medical image, in which the appearance of the anatomical structure is modified; and outputting the second medical image. Further, the image processing device is trained using plural teacher images obtained from a second imaging modality that is different from the first imaging modality. In one embodiment, the method also includes processing the first medical image to obtain plural processed images, wherein each of the plural processed images has a corresponding image resolution; applying the plural processed images to respective multi-training artificial neural networks (MTANNs) to obtain plural output images, wherein each MTANN is trained to detect the anatomical structure at one of the corresponding image resolutions; and combining the plural output images to obtain a second medical image in which the appearance of the anatomical structure is enhanced.

摘要翻译： 一种用于修改医学图像中的解剖结构外观的方法，系统和计算机程序产品，例如胸部X光照片中的肋骨抑制。 该方法包括：使用第一成像模式获取包括解剖结构的第一医学图像; 将第一医学图像应用于训练有素的图像处理装置，以获得对应于第一医学图像的第二医学图像，其中解剖结构的外观被修改; 并输出第二医用图像。 此外，使用从与第一成像模式不同的第二成像模式获得的多个教师图像来训练图像处理装置。 在一个实施例中，该方法还包括处理第一医用图像以获得多个经处理的图像，其中多个处理图像中的每一个具有对应的图像分辨率; 将多个经处理的图像应用于相应的多训练人造神经网络（MTANN）以获得多个输出图像，其中训练每个MTANN以检测其中一个相应图像分辨率的解剖结构; 以及组合所述多个输出图像以获得其中所述解剖结构的外观增强的第二医用图像。

公开(公告)号：US06891964B2

公开(公告)日：2005-05-10

申请号：US09990310

申请日：2001-11-23

申请人： Kunio Doi ,  Masahito Aoyama ,  Qiang Li

发明人： Kunio Doi ,  Masahito Aoyama ,  Qiang Li

IPC分类号： G06T7/00 ,  G06K9/00 ,  G06K9/40 ,  G06K9/48

CPC分类号： G06T7/0012 ,  G06T2207/30061

摘要： An automated computerized scheme for determination of the likelihood of malignancy in pulmonary nodules. The present invention includes steps of obtaining at least one computed tomography medical image of a pulmonary nodule in determining if the pulmonary nodule is malignant based on the examination of seven patient or image features. The method can be implemented when instructions are loaded into a computer to program the computer. The significance of employing seven patient or image features is that statistically, seven features are the most practical based on the unique implementation of statistical analysis. Out of the seven features that are now analyzed to determine if a pulmonary nodule is malignant, these features are selected to optimize the accuracy of the diagnosis of a pulmonary nodule. Through a unique sampling scheme, different embodiments of the present invention utilize different combinations of features to optimize the accuracy of the method of the present invention.

摘要翻译： 一种用于确定肺结节恶性肿瘤可能性的自动计算机化方案。 本发明包括以下步骤：基于七个患者或图像特征的检查，在确定肺结节是否恶性时获得肺结节的至少一个计算机断层摄影医学图像。 当将指令加载到计算机中以对计算机进行编程时，可以实现该方法。 采用七个患者或图像特征的意义在统计学上，七个特征是最实际的，基于统计分析的独特实现。 在现在分析以确定肺结节是否恶性的七个特征中，选择这些特征以优化肺结节诊断的准确性。 通过独特的采样方案，本发明的不同实施例利用不同的特征组合来优化本发明的方法的精度。