-
公开(公告)号：KR101312459B1
公开(公告)日：2013-09-27
申请号：KR1020120054557
申请日：2012-05-23
Applicant: 서울대학교산학협력단
Inventor： 김종효
IPC: G06T5/00
CPC classification number: G06T5/002 , A61B6/5258 , G06T5/20 , G06T2207/10072 , G06T2207/10116 , G06T2207/20024 , G06T2207/20182 , G06T2207/30004
Abstract: PURPOSE: A method for reducing noise of a medical image is provided to utilize the estimated amplitude of noise in a pixel of the image in noise cancellation, by estimating the amplitude of the noise per pixel in the image using physical principle of a medical image acquisition process. CONSTITUTION: The amount of noise in each pixel position of an image is estimated, by using a previously stored reference table (S200). A first signal coherence and structure direction of a first signal are extracted from each pixel position (S300). An anisotropic smoothing image of a middle step is acquired, by filtering using a first anisotropic smoothing kernel (S400,S500). Structure direction and a second signal coherence of a second signal are extracted from each pixel of the anisotropic smoothed image. Filtering of an original image is performed, according as an additional weighted value reflecting the estimated amount of noise is applied in a surrounding pixel in the range of the acquired second anisotropic filter kernel (S600-S800). [Reference numerals] (S100) Image is obtained; (S200) Amount of noises in each pixel is estimated; (S300) Signal coherence and structural direction of each pixel are identified; (S400) Anisotropic smoothing filter kernel of each pixel is identified; (S500) Filtered anisotropic smoothing image in a middle stage is generated; (S600) Anisotropic smoothing filter kernel of each pixel is identified from the filtered anisotropic smoothing image in a middle stage; (S700) Anisotropic smoothing filter, which has an additional weight reflecting the amount of noises for neighboring pixels within the kernel's range, is applied to the original image; (S800) Final image is generated
Abstract translation: 目的：提供一种降低医学图像噪声的方法，以通过使用医学图像采集的物理原理估计图像中每像素的噪声幅度来利用噪声消除中的图像的像素中的估计的噪声幅度 处理。 构成：通过使用先前存储的参考表（S200）来估计图像的每个像素位置中的噪声量。 从每个像素位置提取第一信号的第一信号相干性和结构方向（S300）。 通过使用第一各向异性平滑核（S400，S500）进行滤波来获取中间步骤的各向异性平滑图像。 从各向异性平滑图像的每个像素提取第二信号的结构方向和第二信号相干性。 根据所获取的第二各向异性过滤器核心的范围内的周围像素中的反映估计噪声量的附加加权值，执行原始图像的滤波（S600-S800）。 [附图标记]（S100）获得图像; （S200）估计每个像素中的噪声量; （S300）识别每个像素的信号相干性和结构方向; （S400）识别每个像素的各向异性平滑滤波器核; （S500）生成中间阶段的滤色各向异性平滑图像; （S600）从中间阶段的滤波后的各向异性平滑图像中识别各像素的各向异性平滑滤波器核; （S700）将各向异性平滑滤波器应用于原始图像，其具有反映内核范围内相邻像素的噪声量的附加权重; （S800）生成最终图像
-
公开(公告)号：KR1020110039897A
公开(公告)日：2011-04-20
申请号：KR1020090096935
申请日：2009-10-12
Applicant: 서울대학교산학협력단
Abstract: PURPOSE: A method for detecting clustered microcalcifications on a digital mammogram is provided to early diagnose a breast cancer by detecting the preprocessed image by using a Gaussian smoothing filter and a LoG(Laplacian of Gaussian) filter. CONSTITUTION: A breast region is detected on a mammogram(S1). The detected breast region is preprocessed(S2). A candidate group of clustered microcalcifications is detected(S3). The microcalcification is detected(S4). A preprocess is performed by a Gaussian smoothing filter and a LoG filter.
Abstract translation: 目的：通过使用高斯平滑滤波器和LoG（拉普拉斯高斯）滤波器检测预处理图像，提供了一种用于检测数字乳房X线照片上的聚集微钙化的方法，以早期诊断乳腺癌。 构成：乳房X线照片检测到乳房区域（S1）。 检测到的乳房区域是预处理的（S2）。 检测聚类微钙化的候选组（S3）。 检测到微钙化（S4）。 预处理由高斯平滑滤波器和LoG滤波器执行。
-
公开(公告)号：KR101006189B1
公开(公告)日：2011-01-07
申请号：KR1020080118276
申请日：2008-11-26
Applicant: 서울대학교산학협력단
IPC: A61B6/03
CPC classification number: A61B6/583
Abstract: 본 발명은, CT(Computed Tomogr aphy)촬영장치로 기관지 팬텀을 촬영하여 기관지 팬텀 기관지 벽의 신호강도 프로파일을 산출하는 단계; 상기 기관지 팬텀을 실측하여 이상적인 벽의 신호강도 프로파일을 산출하는 단계; 상기 기관지 팬텀 벽의 신호강도 프로파일과 상기 이상적인 벽의 신호강도 프로파일을 3차원 신호복원하여 상기 CT촬영장치의 점확산함수(Point Spread Function)를 추정하는 단계; 기관지 벽 데이터를 알고자하는 환자 기관지 벽의 신호강도 프로파일을 산출하는 단계; 상기 환자의 기관지 벽의 신호강도 프로파일과 상기 CT촬영장치의 점확산함수를 상기 3차원 신호복원하여 상기 환자의 기관지 벽 데이터를 측정하는 단계; 및 상기 환자의 기관지 벽 데이터를 표현하는 단계를 포함하는 것을 특징으로 하는 전산화단층촬영 영상에서 기관지 벽 데이터를 측정하는 방법을 제공한다.
전산화단층촬영, 기관지, 벽 데이터-
公开(公告)号：KR100996050B1
公开(公告)日：2010-11-22
申请号：KR1020080110654
申请日：2008-11-07
Applicant: 서울대학교산학협력단 , 주식회사 인트로메딕
Abstract: 본 발명은 환자가 의료기관이 아닌 임의에 장소에 있더라도 환자나 의료진이 원격지에서 실시간으로 캡슐 내시경 영상을 확인할 수 있도록 하여 보다 안전하게 검사를 시행할 수 있을 뿐만 아니라 언제 어디서나 간편하게 진료서비스를 제공받을 수 있고, 병변을 자동으로 검출하여 캡슐 내시경 영상의 판독시간과 판독료를 줄일 수 있는 캡슐 내시경 영상을 이용한 U-Health 기반의 자동병변 검출 시스템에 관한 것이다.
캡슐 내시경, 병변검출, 출혈검출, 원격진료, U-health-
公开(公告)号：KR1020100059488A
公开(公告)日：2010-06-04
申请号：KR1020080118276
申请日：2008-11-26
Applicant: 서울대학교산학협력단
IPC: A61B6/03
CPC classification number: A61B6/583
Abstract: PURPOSE: A method for measuring bronchus wall data in the computed tomography image is provided to accurately measure the bronchus wall data by using a three dimensional signal restoring method after analyzing the morphological information of bronchus in three dimension CT image. CONSTITUTION: A bronchus phantom profile is calculated by photographing bronchus phantom using a CT(Computed Tomography) camera(S110). A signal strength profile of an ideal wall is calculated by measuring the bronchus phantom(S130). The point spread function of a CT camera is estimated(S150). The profile of a patient which requires the bronchus wall data is calculated(S170). The bronchus wall data of the patient is measured(S190). The bronchus wall data of the patient is marked(S210).
Abstract translation: 目的：提供一种计算机断层摄影图像中支气管壁数据测量的方法，用于在分析三维CT图像中支气管形态学信息后，采用三维信号恢复方法准确测量支气管壁数据。 构成：使用CT（计算机断层摄影）照相机（S110）拍摄支气管体模来计算支气管幻影剖面。 通过测量支气管体模计算理想壁的信号强度分布（S130）。 估计CT摄像机的点扩散功能（S150）。 计算需要支气管壁数据的患者的轮廓（S170）。 测量患者的支气管壁数据（S190）。 标记患者的支气管壁数据（S210）。
-
Patent Agency Ranking
公开(公告)号：KR1020170044046A
公开(公告)日：2017-04-24
申请号：KR1020160132934
申请日：2016-10-13
Applicant: 서울대학교산학협력단
Abstract: CT 영상기반의신체크기특이적피폭선량및 화질지수제공방법에관한것이며, CT 영상기반의신체크기특이적피폭선량및 화질지수제공방법은, 입력된 CT 영상으로부터피검자의신체영역을식별하는단계, 상기식별된신체영역에기초하여상기피검자의신체크기를고려한신체크기특이적피폭선량을산출하는단계, 상기식별된신체영역에기초하여상기 CT 영상에대한화질지수를산출하는단계, 및상기신체크기특이적피폭선량및 상기화질지수를디스플레이하는단계를포함할수 있다.
Abstract translation: 涉及基于身体尺寸具体剂量和图像质量指数的方法，以提供CT图像，提供了物理尺寸具体剂量CT图像为基础的图像质量指数的方法，包括：识别从输入CT图像的主题的一个物理区域， 计算用于基于所述基于所识别的物理区域计算身体尺寸具体剂量考虑患者，识别体区域的身体尺寸CT图像质量指标，并且主体尺寸 一个特定的曝光剂量和图像质量指数。
-
公开(公告)号：KR101697501B1
公开(公告)日：2017-01-18
申请号：KR1020150104212
申请日：2015-07-23
Applicant: 서울대학교산학협력단
Inventor： 김종효
CPC classification number: G06T11/006 , G06T5/002 , G06T11/008 , G06T2207/10081 , G06T2207/20048
Abstract: 잡음저감방법은, 입력된원본 CT이미지로부터합성사이노그램을생성하는단계와, 생성된상기합성사이노그램으로부터잡음성분을획득하는단계와, 상기잡음성분에기초하여잡음성분 CT 이미지를생성하는단계및 상기잡음성분 CT 이미지에기초하여상기원본 CT 이미지의잡음을저감하는단계를포함할수 있다.
Abstract translation: 提供了一种去噪方法，包括：从输入的原始CT图像生成合成正弦图; 从所生成的合成正弦图获取噪声分量; 基于噪声分量生成噪声分量CT图像; 并基于噪声分量CT图像降低原始CT图像中的噪声。
-
公开(公告)号：KR101133503B1
公开(公告)日：2012-04-05
申请号：KR1020110041182
申请日：2011-04-29
Applicant: 서울대학교산학협력단
Inventor： 김종효
Abstract: PURPOSE: An integrated optical and x-ray ct system and a method of reconstructing the data thereof are provided to enable the precise observation of the inner structure of a subject by reconstructing a color image of the exterior and a 3D image of the inner structure of the subject. CONSTITUTION: The edge points of a subject are extracted from a 3D CT image(S110). 3D coordinates are acquired(S112). The coordinates captured by a camera image surface is acquired(S114). The color component of the corresponding coordinates is acquired(S116). The coordinates of the edge point and the color component of the coordinates are acquired(S118). The color information of the edge point is determined based on color information acquired from various photography points(S120). The color information for all the edge points is determined(S122). A 3D color fusion CT image is produced by fusing the color information(S124).
Abstract translation: 目的：提供一种集成的光学和x射线ct系统及其数据的重建方法，以通过重建外部的彩色图像和内部结构的3D图像来精确地观察对象的内部结构 主题。 构成：从3D CT图像中提取被摄体的边缘点（S110）。 获取3D坐标（S112）。 获取由相机图像表面捕获的坐标（S114）。 获取相应坐标的颜色分量（S116）。 获取坐标的边缘点和颜色分量的坐标（S118）。 基于从各种拍摄点获取的颜色信息确定边缘点的颜色信息（S120）。 确定所有边缘点的颜色信息（S122）。 通过融合颜色信息产生3D彩色融合CT图像（S124）。
-
-
公开(公告)号：KR1020110039896A
公开(公告)日：2011-04-20
申请号：KR1020090096934
申请日：2009-10-12
Applicant: 서울대학교산학협력단
Abstract: PURPOSE: A method for automatically measuring breast density is provided to early diagnose a breast cancer by measuring breast density using the intensity of a boundary and an observer model. CONSTITUTION: A histogram is generated according to a pixel value of a breast region(S1). A profile of mean and dispersion is generated as a probability medium variable of the breast region(S2). A boundary intensity profile of the fat region and a mammary duct region(S3). The fat region and the mammary duct region are optimally separated by calculating separability(S4). The density of the breast is calculated(S5).
Abstract translation: 目的：提供一种自动测量乳腺密度的方法，以便通过使用边界和观察者模型的强度来测量乳腺密度来早期诊断乳腺癌。 构成：根据乳房区域的像素值生成直方图（S1）。 产生平均和分散的轮廓作为乳房区域的概率介质变量（S2）。 脂肪区域和乳腺导管区域的边界强度分布（S3）。 脂肪区域和乳腺管区域通过计算分离性来最佳地分离（S4）。 计算乳房密度（S5）。
-
-
-
-
-
-
-
-
-
Search Results
25
records
(took 0.026 seconds)
