WO2018113282A1 - 三维盆底超声图像处理方法及系统 - Google Patents

三维盆底超声图像处理方法及系统 Download PDF

Info

Publication number
WO2018113282A1
WO2018113282A1 PCT/CN2017/093456 CN2017093456W WO2018113282A1 WO 2018113282 A1 WO2018113282 A1 WO 2018113282A1 CN 2017093456 W CN2017093456 W CN 2017093456W WO 2018113282 A1 WO2018113282 A1 WO 2018113282A1
Authority
WO
WIPO (PCT)
Prior art keywords
levator
pelvic floor
muscle
ultrasound image
levator ani
Prior art date
Application number
PCT/CN2017/093456
Other languages
English (en)
French (fr)
Inventor
李萍
艾金钦
潘美玲
唐艳红
许龙
Original Assignee
深圳开立生物医疗科技股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳开立生物医疗科技股份有限公司 filed Critical 深圳开立生物医疗科技股份有限公司
Publication of WO2018113282A1 publication Critical patent/WO2018113282A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T15/003D [Three Dimensional] image rendering

Definitions

  • the present invention relates to the field of image processing technologies, and in particular, to a three-dimensional pelvic floor ultrasound image processing method and system.
  • the processing of 3D pelvic floor ultrasound images is basically manually selected, adjusted and measured by a doctor or researcher.
  • manually adjust the volume of interest (VOI) placed in the lower rim of the pubic symphysis and the rectal anal canal area.
  • Four-dimensional imaging was performed by manually playing back the Valsalva action (ie, performing a strong closing action), and manually selecting the three-dimensional levator muscle ultrasound image under the maximum Valsalva motion.
  • On the three-dimensional pelvic floor ultrasound image manually measure the anteroposterior diameter, transverse, area, thickness of the pubic visceral muscle, and the angle of the levator ani muscle.
  • the manual operation of the three-dimensional pelvic floor ultrasound image is cumbersome and the image processing efficiency is low.
  • the use of manual measurements will also bring in certain errors, affecting the accuracy of subsequent research and diagnosis.
  • the invention provides a three-dimensional pelvic floor ultrasound image processing method and system, which can automatically and accurately extract the contour of the levator muscle hole, thereby improving the efficiency and accuracy of image processing.
  • the present invention adopts the following technical solutions:
  • a three-dimensional pelvic floor ultrasound image processing method includes:
  • the three-dimensional pelvic floor ultrasound image with the largest area is taken as the three-dimensional pelvic floor ultrasound image when the maximum Valsalva is in motion, and the three-dimensional pelvic floor ultrasound image when the maximum Valsalva is in motion is used as a reference image.
  • the step of extracting the levator ani muscle contour and the levator muscle split contour from the tissue structure information comprises:
  • a local minimum value is set according to a distribution of absolute values of the gradient using a recursive algorithm
  • the segmentation result is subjected to small region merging.
  • the step of calculating the area of the corresponding levator ani muscle hole according to the contour of the levator ani muscle in the three-dimensional levator muscle ultrasound image of each frame includes:
  • the method further includes:
  • a three-dimensional pelvic floor ultrasound image processing system includes:
  • An identification unit configured to identify tissue structure information of the levator ani muscle and the levator levator muscle in the three-dimensional pelvic floor ultrasound image of each frame according to the sonographic features of the pelvic floor;
  • An extracting unit configured to extract the levator ani muscle contour and the levator levator muscle contour from the tissue structure information
  • a calculation module configured to calculate an area of the corresponding levator ani muscle hole according to the contour of the levator muscle hole in the three-dimensional pelvic floor ultrasound image of each frame;
  • the selecting unit is configured to use the three-dimensional pelvic floor ultrasound image with the largest area as the three-dimensional pelvic floor ultrasound image when the maximum Valsalva is in motion, and use the three-dimensional pelvic floor ultrasound image when the maximum Valsalva is in motion as a reference image.
  • the extraction unit is for:
  • a local minimum value is set according to a distribution of absolute values of the gradient using a recursive algorithm
  • the extracting module is further configured to perform small area merging on the segmentation result when the watershed segmentation region is excessively segmented.
  • the calculation module is further configured to calculate an area of the levator levator hole according to a corresponding single pixel area ratio according to the number of pixels in the levator muscle splitting contour.
  • the method further includes:
  • a measuring module configured to measure an levator ani muscle and an levator levator hole of the reference image to obtain measurement parameters of the levator ani muscle and the levator levator hole, wherein the measurement parameter comprises: an anal lift The anteroposterior diameter, transverse diameter, area, thickness and angle of the visceral muscle of the phalanges;
  • the measuring module is further configured to determine the levator ani muscle damage according to the continuity characteristic of the levator ani muscle contour.
  • the embodiment of the present invention obtains the time when the examinee is in the action of Valsalva Multi-frame three-dimensional pelvic floor ultrasound image; according to the acoustic image of the pelvic floor, the tissue structure information of the levator ani muscle and the levator ani muscle in the three-dimensional pelvic floor ultrasound image of each frame is identified; the levator ani muscle is extracted from the tissue structure information Contour and levator muscle hole contour; calculate the area of the corresponding levator ani muscle hole according to the contour of the levator ani muscle in the three-dimensional pelvic floor ultrasound image of each frame; the largest three-dimensional pelvic floor ultrasound image as the maximum Valsalva action A three-dimensional pelvic floor ultrasound image and a three-dimensional pelvic floor ultrasound image with maximum Valsalva motion as a reference image.
  • the above-mentioned three-dimensional pelvic floor ultrasound image processing method and system can automatically and accurately extract the contour of the levator ani muscle hole, and the operation is simple and convenient, thereby improving the efficiency and accuracy of
  • FIG. 1 is a flowchart of a method of a first embodiment of a three-dimensional pelvic floor ultrasound image processing method according to an embodiment of the present invention
  • FIG. 2 is a schematic cross-sectional view of a three-dimensional pelvic floor ultrasound image provided in an embodiment of the present invention
  • FIG. 3 is a flow chart of extracting an outline of an levator ani muscle and a contour of an levator ani muscle in a specific embodiment of the present invention
  • FIG. 4 is a schematic diagram of measurement parameters of an levator ani muscle of a three-dimensional pelvic floor ultrasound image provided in an embodiment of the present invention
  • FIG. 5 is a structural block diagram of a three-dimensional pelvic floor ultrasound image processing system provided in an embodiment of the present invention.
  • FIG. 6 is a block diagram showing the structure of a three-dimensional pelvic floor ultrasound image processing system according to another embodiment of the present invention.
  • FIG. 1 is a flowchart of a method for a first embodiment of a three-dimensional pelvic floor ultrasound image processing method according to an embodiment of the present invention. As shown, the method includes:
  • Step 101 Acquire a multi-frame three-dimensional pelvic floor ultrasound image of the examinee when the Valsalva moves.
  • the initial scan section shows the median sagittal section of the pelvic floor of the examinee, and the median sagittal section of the pelvic floor clearly shows the pubic symphysis, urethra, vagina, rectum, and anal canal.
  • the examinee is performing a Valsalva action (ie, performing a strong closed call action)
  • the scan is turned on.
  • the user needs to observe the two-dimensional image of the probe scan, move or rotate the probe until the pubic symphysis (medical fixed position) is approximately 45 degrees from the horizontal line, and fix the probe to enter the 4D scan mode.
  • Collect 3D pelvic floor tissue information and obtain 3D pelvic floor ultrasound images when multi-frame (Zhang) Valsalva action.
  • the number of frames can be set according to actual conditions, for example, it can be 10, 20, 30 or 60.
  • Step 102 Identify tissue structure information of the levator ani muscle and the levator levator muscle in each frame of the three-dimensional pelvic floor ultrasound image according to the characteristics of the sound image.
  • each three-dimensional pelvic floor ultrasound image Before identifying the tissue structure information of the levator ani muscle and the levator ani muscle in each three-dimensional pelvic floor ultrasound image, adjust the volume of interest frame area so that each three-dimensional pelvic floor ultrasound image shows the levator ani muscle plane Complete information.
  • the volumetric region of interest of each frame of the three-dimensional pelvic floor ultrasound image is automatically adjusted, so that the three-dimensional pelvic floor ultrasound image shows the complete information of the levator ani muscle plane, clearly showing the pubic symphysis, two The lateral puberulent visceral muscle and its internal urethra, vagina, rectum and other organizational structures, and exclude other irrelevant information, to avoid adverse effects on subsequent image processing effects.
  • the bilateral levator ani muscle and the pubic symphysis form a diamond-shaped levator ani muscle in the resting state under the resting state.
  • the structure in the levator sac is in the order from the front to the back without echo.
  • the ultrasonic echo of the levator ani muscle is continuous, the bilateral symmetry is basically symmetrical, the levator ani muscle is brighter in the ultrasound image, and the contour area of the levator ani muscle is the bright and dark junction area.
  • the levator ani muscle refers to the U-shaped area between the inner contour and the outer contour
  • the levator muscle split hole refers to the inner contour line of the U shape and the area surrounded by the upper edge of the U-shaped inner contour line.
  • step 103 the contour of the levator ani muscle and the contour of the levator ani muscle are extracted from the tissue structure information.
  • the levator ani muscle contour includes an inner contour and an outer contour.
  • the contour of the levator ani muscle ie, the U-shaped inner contour, can be obtained.
  • the levator ani muscles are brighter in the three-dimensional pelvic floor image, and the levator ani muscle contour area is the light and dark intersection area, in the present embodiment, the levator muscle inner contour and the outer contour are extracted by the watershed algorithm.
  • Step 104 Calculate the area of the corresponding levator ani muscle hole according to the contour of the levator ani muscle in each three-dimensional pelvic floor ultrasound image.
  • the area of the levator ani muscle hole in the three-dimensional pelvic floor ultrasound image of each frame is calculated, and multiple area values are obtained, and multiple area values are obtained.
  • step 105 the three-dimensional pelvic floor ultrasound image with the largest area is taken as the three-dimensional pelvic floor ultrasound image when the maximum Valsalva is operated, and the three-dimensional pelvic floor ultrasound image when the maximum Valsalva is operated is used as the reference image.
  • the three-dimensional pelvic floor ultrasound image corresponding to the maximum area value obtained in the above step 104 is obtained, and the three-dimensional pelvic floor ultrasound image is taken as the three-dimensional pelvic floor ultrasound image when the maximum Valsalva is operated.
  • the three-dimensional pelvic floor ultrasound image when the maximum Valsalva is operated that is, the three-dimensional pelvic floor ultrasound image corresponding to the maximum area value is used as the reference image, and the reference image is used as a reference for subsequent detection and measurement.
  • the three-dimensional pelvic floor ultrasound image processing method of the above embodiment obtains a multi-frame three-dimensional pelvic floor ultrasound image of the examinee in the action of Valsalva; and identifies the levator ani muscle and the levator ani muscle in each frame of the three-dimensional pelvic floor ultrasound image according to the acoustic image characteristics.
  • Tissue structure information of the rupture hole extracting the contour of the levator ani muscle and the contour of the levator ani muscle from the tissue structure information; calculating the area of the levator ani muscle hole according to the contour of the levator ani muscle; taking the largest 3D pelvic floor ultrasound image as the maximum Valsalva action
  • a three-dimensional pelvic floor ultrasound image and a three-dimensional pelvic floor ultrasound image with maximum Valsalva motion as a reference image.
  • the above three-dimensional pelvic floor ultrasound map Like the processing method, it can automatically and accurately extract the outline of the levator ani muscle and the levator ani muscle, and the operation is simple and convenient, thereby improving the efficiency and accuracy of the pelvic
  • the step of extracting the levator ani muscle contour and the levator ani muscle split contour from the tissue structure information comprises:
  • Step 302 filtering the three-dimensional pelvic floor ultrasound image to obtain a filtered image.
  • Gaussian filtering is performed on the three-dimensional levator muscle ultrasound image f(x, y) to obtain a Gaussian filtered image.
  • a corrosion expansion process or an opening and closing process
  • Step 304 calculating a gradient of the filtered image.
  • grad(.) represents the gradient
  • h represents the pixel spacing of the gradient (generally set to 1)
  • f 1 (x, y) represents the gray value of the pixel (x, y)
  • f 1 (x+h) , y) represents the gray value of the pixel point (x+h, y)
  • f 1 (xh, y) represents the gray value of the pixel point (xh, y).
  • step 306 the absolute value of the gradient is obtained, and the maximum value and the minimum value of the absolute value are filtered.
  • the absolute value D of the plurality of gradient values obtained in the above steps is obtained, and the plurality of absolute values D are sorted, and the maximum value and the minimum value of the absolute values are selected and defined as g_max and g_min, respectively.
  • Step 308 in the process of increasing from the minimum value to the maximum value, the local minimum value is calculated using an iterative algorithm according to the distribution of the absolute values of the gradients.
  • X(g) as the union of the catchment basin when the water level is at g; at g+1, a connected component T(g+1) is a new local minimum, or an existing X(g) A basin expansion.
  • T(g+1) is a basin extension of an existing X(g)
  • MIN(g) is the local minimum that occurs when the height is g
  • Y(g+1, X(g)) is the height of g+1.
  • a collection of X(g) points at the same time.
  • Step 310 forming a watershed of the filtered grayscale image according to the local minimum value, and extracting the contour of the levator ani muscle and the contour of the levator levator muscle from the tissue structure information of the levator ani muscle according to the watershed.
  • the iterative algorithm is iterated according to the above steps. After the threshold is set, the iteration is stopped. Calculate the basin corresponding to each local minimum and obtain the boundary of the basin.
  • the boundary of the basin is the watershed boundary, which is the watershed.
  • the boundary of the watershed is the resulting contour.
  • the inner and outer contours of the levator ani muscle and the contour of the upper apex of the levator ani muscle can be obtained by calculating the watershed of the preset height.
  • the inner contour of the levator ani muscle and the contour of the upper apex of the levator ani muscle split the entire contour of the levator ani muscle. For example, according to the gray feature of the current image, when the preset height is 20, the watershed (boundary) is the inner contour; when the preset height is 35, the watershed (boundary) is the outer contour.
  • the segmentation result is subjected to small region merging.
  • the segmentation process if the watershed segmentation region is over-segmented, the segmentation result is merged into small regions. For example, if the difference in the average gray value in the region is within the preset range, it belongs to the same region.
  • the above-mentioned watershed algorithm can accurately extract the contour of the levator ani muscle, and the operation is simple and convenient, and the efficiency is improved.
  • the step of calculating the area of the levator ani muscle hole according to the contour of the levator ani muscle in the three-dimensional levator muscle ultrasound image of each frame includes:
  • the area of the levator ani muscle hole is calculated according to the corresponding single pixel area ratio according to the number of pixels in the contour of the levator musculature.
  • the method further includes:
  • the levator ani muscle and the levator ani muscle hole of the reference image are measured to obtain measurement parameters, wherein the measurement parameters include: anteroposterior diameter, transverse diameter, area of the levator levator muscle, thickness and angle of the visceral visceral muscle.
  • the anteroposterior diameter of the levator ani muscle (M, the distance between the midpoint of the medial edge of the pubic symphysis and the medial edge of the pubic visceral muscle at the posterior rectum), transverse (N, The maximum distance between the inner edges of the pubic visceral muscles), the area (the area between the medial edge of the pubic symphysis and the medial edge of the pubic visceral muscle), and the thickness of the pubic visceral muscle (T, the medial diameter of the medial segment of the pubic visceral muscle ), angle (R, The angle formed by the two sides of the pubic visceral muscle in the back of the rectum is centered on the vertical line.
  • the angle at this time is the angle R).
  • the inner and outer contours obtained in the previous step are searched for the vertical spacing between the inner and outer contours, which is the thickness T of the pubic visceral muscle.
  • the continuity feature of the levator ani muscle contour and the contour feature of the levator ani muscle are further obtained.
  • FIG. 5 is a structural block diagram of a first embodiment of a three-dimensional pelvic floor ultrasound image processing system provided in an embodiment of the present invention.
  • the system 500 includes:
  • the acquiring unit 501 is configured to acquire a multi-frame three-dimensional pelvic floor ultrasound image when the examinee moves in the Valsalva;
  • the identification unit 502 is configured to identify tissue structure information of the levator ani muscle and the levator levator muscle in each frame of the three-dimensional pelvic floor ultrasound image according to the acoustic image of the pelvic floor;
  • An extracting unit 503, configured to extract an levator ani muscle contour and an levator levator muscle contour from the tissue structure information
  • the calculating module 504 is configured to calculate an area of the corresponding levator levator muscle hole according to the contour of the levator ani muscle hole in the three-dimensional pelvic floor ultrasound image of each frame;
  • the selecting unit 505 is configured to use the three-dimensional pelvic floor ultrasound image with the largest area as the three-dimensional pelvic floor ultrasound image in the maximum Valsalva motion, and the three-dimensional pelvic floor ultrasound image in the maximum Valsalva motion as the reference image.
  • the extraction unit 503 is configured to:
  • the local minimum value is set according to the distribution of the absolute values of the gradients using a recursive algorithm
  • the extraction module 503 is further configured to: when the watershed segmentation region is excessively segmented, perform small region merging on the segmentation result.
  • the calculation module 504 is further configured to calculate the area of the levator levator hole according to the corresponding single pixel area ratio according to the number of pixels in the levator muscle splitting contour.
  • system 500 further includes:
  • the measuring module 506 is configured to measure the levator ani muscle and the levator levator hole of the reference image to obtain measurement parameters of the levator ani muscle and the levator levator hole, wherein the measurement parameters include: anteroposterior diameter and transverse direction of the levator ani muscle hole Diameter, area, thickness and angle of visceral muscle of the phalanx.
  • the measuring module 506 is further configured to determine the levator ani muscle damage according to the continuity feature of the levator ani muscle contour.
  • the levator ani muscle hole loses the typical "U" shape or "V" shape.
  • the levator ani muscle injury can be considered, and the degree of levator ani muscle damage can be evaluated based on the measurement results. For example, in general, during the maximum Valsalva exercise, the area of the levator ani muscle is less than 25 cm 2 is normal, 30-34.9 cm 2 is mild expansion, 35-39.9 cm 2 is moderate expansion, and greater than 40 cm 2 is severe expansion.
  • the diagnostic grades vary slightly from country to country.
  • the three-dimensional pelvic floor ultrasound image processing system 500 of the present embodiment is used to implement the aforementioned three-dimensional pelvic floor ultrasound image processing method. Therefore, the specific embodiment in the three-dimensional pelvic floor ultrasound image processing system 500 can be seen in the foregoing three-dimensional pelvic floor ultrasound image processing method.
  • the embodiment part, for example, the obtaining unit 501, the identifying unit 502, the extracting unit 503, the calculating module 504, and the selecting unit 505 are respectively used to implement steps 101, 102, 103, 104 and 105 in the above-mentioned three-dimensional pelvic floor ultrasound image processing method. Therefore, the specific implementation manners can be referred to the description of the respective partial embodiments, and details are not described herein again.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Graphics (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Ultra Sonic Daignosis Equipment (AREA)
  • Image Processing (AREA)

Abstract

一种三维盆底超声图像处理方法及系统。该方法包括:获取被检查者在Valsalva动作时的多帧三维盆底超声图像;根据盆底的声像学特征识别每帧三维盆底超声图像中的肛提肌以及肛提肌裂孔的组织结构信息;从组织结构信息中提取肛提肌轮廓和肛提肌裂孔轮廓;分别根据每帧所述三维盆底超声图像中肛提肌裂孔的轮廓计算其对应的肛提肌裂孔的面积;将面积最大的三维盆底超声图像作为最大Valsalva动作时的三维盆底超声图像,并将最大Valsalva动作时的三维盆底超声图像作为参考图像,由此其能够自动、准确地提取肛提肌裂孔的轮廓,从而提高图像处理的效率和准确度。

Description

三维盆底超声图像处理方法及系统
本申请要求于2016年12月22日提交中国专利局、申请号为201611197301.3、发明名称为“三维盆底超声图像处理方法及系统”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本发明涉及图像处理技术领域,尤其涉及一种三维盆底超声图像处理方法及系统。
背景技术
目前对三维盆底超声图像的处理基本是由医生或者研究人员手动选取、调整和测量。例如:手动调节感兴趣容积区域(VOI),放置于耻骨联合内下缘及直肠肛管角区域。手动回放Valsalva动作(即进行强力闭呼动作)下进行四维成像,手动选择最大Valsalva动作下的三维肛提肌超声图像。在三维盆底超声图像上手动测量肛提肌裂孔前后径、横经、面积、耻骨内脏肌的厚度、夹角等信息。手动处理三维盆底超声图像的操作繁琐,图像处理效率低。另外,采用手动测量还会带入一定的误差,影响后续研究和诊断的准确性。
发明内容
本发明提供了一种三维盆底超声图像处理方法及系统,该方法及系统能够自动、准确地提取肛提肌裂孔的轮廓,从而提高图像处理的效率和准确度。
为实现上述设计,本发明采用以下技术方案:
第一方面,一种三维盆底超声图像处理方法,包括:
获取被检查者在Valsalva动作时的多帧三维盆底超声图像;
根据盆底的声像学特征识别每帧所述三维盆底超声图像中的肛提肌以及肛提肌裂孔的组织结构信息;
从所述组织结构信息中提取所述肛提肌的轮廓和所述肛提肌裂孔的轮廓;
分别根据每帧所述三维盆底超声图像中所述肛提肌裂孔轮廓计算其对应的所述肛提肌裂孔的面积;
将所述面积最大的三维盆底超声图像作为最大Valsalva动作时的三维盆底超声图像,并将所述最大Valsalva动作时的三维盆底超声图像作为参考图像。
在一个实施例中,所述从所述组织结构信息中提取所述肛提肌轮廓和所述肛提肌裂孔轮廓的步骤,包括:
对所述三维盆底超声图像的灰度图像进行滤波,得到滤波后的灰度图像;
计算所述滤波后的灰度图像的梯度;
求取所述梯度的绝对值,选出所述绝对值的最大值和最小值;
从所述最小值递增到所述最大值的过程中,根据所述梯度的绝对值的分布使用递归算法设定局部极小值;
根据所述局部极小值形成所述滤波处理后的灰度图像的分水岭;
根据所述分水岭从所述肛提肌和所述肛提肌裂孔的组织结构信息中提取所述肛提肌的轮廓和所述肛提肌裂孔的轮廓。
在一个实施例中,当所述分水岭分割的区域出现过度分割时,则对分割结果进行小区域合并。
在一个实施例中,所述分别根据每帧所述三维肛提肌超声图像中所述肛提肌裂孔轮廓计算其对应的肛提肌裂孔的面积的步骤包括:
根据所述肛提肌裂孔轮廓的像素点数按照对应单像素面积比,计算所述肛提肌裂孔的面积。
在一个实施例中,所述将所述最大Valsalva动作时的三维肛提肌超声图像作为参考图像的步骤之后,还包括:
对所述参考图像的肛提肌和肛提肌裂孔进行测量,以获取所述肛提肌和所述肛提肌裂孔的测量参数,其中,所述测量参数包括:肛提肌裂孔的前后径、横径、面积、趾骨内脏肌的厚度和夹角。
第二方面,一种三维盆底超声图像处理系统,包括:
获取单元,用于获取被检查者在Valsalva动作时的多帧三维盆底超声图 像;
识别单元,用于根据盆底的声像学特征识别每帧所述三维盆底超声图像中的肛提肌以及肛提肌裂孔的组织结构信息;
提取单元,用于从所述组织结构信息中提取所述肛提肌轮廓和所述肛提肌裂孔轮廓;
计算模块,用于分别根据每帧所述三维盆底超声图像中所述肛提肌裂孔轮廓计算其对应的所述肛提肌裂孔的面积;
选取单元,用于将所述面积最大的三维盆底超声图像作为最大Valsalva动作时的三维盆底超声图像,并将所述最大Valsalva动作时的三维盆底超声图像作为参考图像。
在一个实施例中,所述提取单元用于:
对所述三维盆底超声图像的灰度图像进行滤波,得到滤波后的灰度图像;
计算所述滤波后的灰度图像的梯度;
求取所述梯度的绝对值,选出所述绝对值的最大值和最小值;
从所述最小值递增到所述最大值的过程中,根据所述梯度的绝对值的分布使用递归算法设定局部极小值;
根据所述局部极小值形成所述滤波处理后的灰度图像的分水岭;
根据所述分水岭从所述肛提肌和所述肛提肌裂孔的组织结构信息中提取所述肛提肌的轮廓和所述肛提肌裂孔的轮廓。
在一个实施例中,所述提取模块还用于:当所述分水岭分割的区域出现过度分割时,则对分割结果进行小区域合并。
在一个实施例中,所述计算模块还用于根据所述肛提肌裂孔轮廓内的像素点数按照对应单像素面积比,计算所述肛提肌裂孔的面积。
在一个实施例中,还包括:
测量模块,用于对所述参考图像的肛提肌和肛提肌裂孔进行测量,以获取所述肛提肌和所述肛提肌裂孔的测量参数,其中,所述测量参数包括:肛提肌裂孔的前后径、横径、面积、趾骨内脏肌的厚度和夹角;
测量模块,还用于根据所述肛提肌轮廓的连续性特征判定肛提肌损伤。
本发明的有益效果为:本发明实施例获取被检查者在Valsalva动作时的 多帧三维盆底超声图像;根据盆底的声像学特征识别每帧所述三维盆底超声图像中的肛提肌以及肛提肌裂孔的组织结构信息;从组织结构信息中提取肛提肌轮廓和肛提肌裂孔轮廓;分别根据每帧所述三维盆底超声图像中肛提肌裂孔轮廓计算其对应的肛提肌裂孔的面积;将面积最大的三维盆底超声图像作为最大Valsalva动作时的三维盆底超声图像,并将最大Valsalva动作时的三维盆底超声图像作为参考图像。上述的三维盆底超声图像处理方法及系统,能够自动、准确地提取肛提肌裂孔的轮廓,且操作简单便捷,从而提高了盆底图像处理的效率和准确度。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是本发明具体实施方式中提供的一种三维盆底超声图像处理方法的第一实施例的方法流程图;
图2是本发明具体实施方式中提供的三维盆底超声图像的横切面示意图;
图3是本发明具体实施方式中提供的提取肛提肌轮廓和肛提肌裂孔轮廓的流程图;
图4是本发明具体实施方式中提供的一种三维盆底超声图像的肛提肌测量参数示意图;
图5是本发明具体实施方式中提供的三维盆底超声图像处理系统的结构框图;
图6是本发明具体实施方式中提供的另一实施例的三维盆底超声图像处理系统的结构框图。
具体实施方式
为使本发明解决的技术问题、采用的技术方案和达到的技术效果更加清 楚,下面将结合附图对本发明实施例的技术方案作进一步的详细描述,显然,所描述的实施例仅仅是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
请参考图1,其是本发明具体实施方式中提供的一种三维盆底超声图像处理方法的第一实施例的方法流程图。如图所示,该方法包括:
步骤101,获取被检查者在Valsalva动作时的多帧三维盆底超声图像。
当用户使用超声设备扫查被检查者时,起始扫查切面显示被检查者的盆底正中矢状切面,盆底正中矢状切面能够清晰地显示耻骨联合、尿道、阴道、直肠和肛管。当被检查者在进行Valsalva动作(即进行强力闭呼动作)时,开启扫查。扫查时用户需观察探头扫查的二维图像,移动或旋转探头,直到耻骨联合处(医学固定位置)与水平线大致成45度时,固定住探头后进入4D扫查模式。采集三维盆底组织信息,获取多帧(张)Valsalva动作时的三维盆底超声图像。帧数可以根据实际情况进行设置,例如可以为10、20、30或60等。
步骤102,根据声像学特征识别每帧三维盆底超声图像中肛提肌和肛提肌裂孔的组织结构信息。
在识别每一幅三维盆底超声图像中肛提肌和肛提肌裂孔的组织结构信息之前,调整感兴趣容积框区域,以使每一幅三维盆底超声图像上显示肛提肌轴平面的完整信息。
根据肛提肌轴平面的信息自动调整每一帧三维盆底超声图像的感兴趣容积框区域,以使三维盆底超声图像上显示肛提肌轴平面的完整信息,清晰地显示耻骨联合、两侧的耻骨内脏肌以及其内的尿道、阴道、直肠等组织结构,且排除其他无关信息,避免对后续的图像处理效果造成不良影响。
在正常的盆底超声图像上,在静息状态下双侧肛提肌与耻骨联合在双侧耻骨下方共同形成菱形的肛提肌裂孔,肛提肌裂孔内的结构从前至后依次为无回声的尿道、“U”形或“H”形阴道横断面及圆形直肠横断面。肛提肌的超声回声连续性较好,双侧基本对称,肛提肌在超声图像中表现较亮,肛提肌裂孔的轮廓区域为亮暗交接区域。根据上述肛提肌和肛提肌裂孔特有的声像学特征, 在三维超声图像中识别肛提肌和肛提肌裂孔的组织结构,并提取肛提肌和肛提肌裂孔的组织结构信息。如图2所示,肛提肌是指内轮廓与外轮廓之间的U形区域;肛提肌裂孔是指U形的内轮廓线及U形的内轮廓线的上边沿包围起来的区域。当扫查过程中得到了U形区域的位置,用户就要识别肛提肌和肛提肌裂孔。当图像不够清晰,比如U环图像比较暗,这样的标准图像就叫不能很好的让用户识别其组织结构。
步骤103,从组织结构信息中提取肛提肌轮廓和肛提肌裂孔轮廓。
由上述步骤102可知,肛提肌轮廓包括内轮廓和外轮廓。在本实施例中,从组织结构信息中提取肛提肌轮廓后,便可得到肛提肌裂孔轮廓,即U型内轮廓。
由于肛提肌在三维盆底图像中表现较亮,肛提肌轮廓区域为亮暗交接区域,在本实施例中,采用分水岭算法提取肛提肌内轮廓和外轮廓。
步骤104,分别根据每帧三维盆底超声图像中肛提肌裂孔轮廓计算其对应的肛提肌裂孔的面积。
根据上述步骤获取的肛提肌裂孔轮廓,计算每帧三维盆底超声图像中肛提肌裂孔的面积,得到多个面积值,并对得到多个面积值排序。
步骤105,将面积最大的三维盆底超声图像作为最大Valsalva动作时的三维盆底超声图像,并将最大Valsalva动作时的三维盆底超声图像作为参考图像。
获取上述步骤104得到的最大面积值对应的三维盆底超声图像,将该三维盆底超声图像作为最大Valsalva动作时的三维盆底超声图像。将最大Valsalva动作时的三维盆底超声图像即最大面积值对应的三维盆底超声图像作为参考图像,以参考图像为进行后续的检测和测量的基准。
上述实施例的三维盆底超声图像处理方法,获取被检查者在Valsalva动作时的多帧三维盆底超声图像;根据声像学特征识别每帧三维盆底超声图像中肛提肌和肛提肌裂孔的组织结构信息;从组织结构信息中提取肛提肌轮廓和肛提肌裂孔轮廓;根据肛提肌裂孔轮廓计算肛提肌裂孔的面积;将面积最大的三维盆底超声图像作为最大Valsalva动作时的三维盆底超声图像,并将最大Valsalva动作时的三维盆底超声图像作为参考图像。上述的三维盆底超声图 像处理方法,能够自动、准确地提取肛提肌和肛提肌裂孔的轮廓,且操作简单便捷,从而提高了盆底图像处理的效率和准确度。
可选地,在一个实施例中,从组织结构信息中提取肛提肌轮廓和肛提肌裂孔轮廓的步骤包括:
步骤302,对三维盆底超声图像进行滤波,得到滤波后的图像。
在本实施例中,对三维肛提肌超声图像f(x,y)进行高斯滤波,得到高斯滤波后的图像。另外,为了进一步减少噪声的影响,还可以对对高斯滤波后的图像进行腐蚀膨胀处理(或开闭处理)。
步骤304,计算滤波后图像的梯度。
求腐蚀膨胀处理后的图像的梯度图像,公式如下:
Figure PCTCN2017093456-appb-000001
其中,grad(.)表示求梯度,h表示求梯度的像素间隔(一般设为1),f1(x,y)表示像素点(x,y)的灰度值,f1(x+h,y)表示像素点(x+h,y)的灰度值,f1(x-h,y)表示像素点(x-h,y)的灰度值。
步骤306,求取梯度的绝对值,筛选出绝对值的最大值和最小值。
求取上述步骤获取的多个梯度值的绝对值D,并对多个绝对值D排序,筛选出绝对值的最大值与最小值,分别定义为g_max与g_min。
步骤308,从最小值递增到最大值的过程中,根据梯度的绝对值的分布使用迭代算法计算局部极小值。
定义一个从g_min到g_max的水位g不断递增的递归过程,这个过程中每个不同的局部最小的汇水盆地都不断扩展。定义X(g)为水位在g时候的汇水盆地集合的并;在g+1层,一个连通分量T(g+1)是一个新的局部最小,或者是一个已经存在的X(g)的一个盆地扩展。若T(g+1)为一个已经存在的X(g)的一个盆地扩展,按邻接关系计算高度为g+1的每个点与各汇水盆的距离,如果一个点与两个以上的盆地等距离,则不属于任何盆地,否则属于与它最近的盆地,这样就产生新的X(g+1),计算公式如下:
Figure PCTCN2017093456-appb-000002
其中,MIN(g)为高度为g时出现的局部最小,Y(g+1,X(g))为高度为g+1 同时属于X(g)点的集合。
步骤310,根据局部极小值形成滤波处理后的灰度图像的分水岭,根据该分水岭从肛提肌的组织结构信息中提取肛提肌的轮廓和肛提肌裂孔的轮廓。
按照上述步骤的迭代算法迭代,达到设置阈值后,停止迭代。计算每一个局部最小值对应的盆地,获取盆地的边界,盆地的边界即为分水岭的边界即为分水岭。分水岭的边界即为所得轮廓。
通过计算预设高度的分水岭可以得到肛提肌的内轮廓和外轮廓以及肛提肌裂孔的上围盖的轮廓。肛提肌的内轮廓和肛提肌裂孔的上围盖的轮廓组成了肛提肌裂孔的整个轮廓。例如:根据当前图像的灰度特征,当预设高度为20时分水岭(边界)为内轮廓;当预设高度为35时分水岭(边界)为外轮廓。
进一步地,在一个实施例中,当分水岭分割的区域出现过度分割时,则对分割结果进行小区域合并。
在分割过程中,如果分水岭分割的区域出现了过度分割情况,则对分割结果进行小区域合并,例如将区域内平均灰度值的差别在预设范围内则属于同一区域。
上述的采用分水岭算法能准确地提取肛提肌裂孔的轮廓,且操作简单便捷,提高了效率。
在一个实施例中,分别根据每帧三维肛提肌超声图像中肛提肌裂孔轮廓计算肛提肌裂孔的面积的步骤包括:
根据肛提肌裂孔轮廓内的像素点数按照对应单像素面积比,计算肛提肌裂孔的面积。
在一个实施例中,将最大Valsalva动作时的三维肛提肌超声图像作为参考图像的步骤之后,还包括:
对参考图像的肛提肌和肛提肌裂孔进行测量,以获取测量参数,其中,测量参数包括:肛提肌裂孔的前后径、横径、面积、趾骨内脏肌的厚度和夹角。
在本实施例中,如图4所示,对肛提肌裂孔前后径(M,耻骨联合内侧缘中点与耻骨内脏肌在直肠后方汇合处内侧缘之间的距离)、横经(N,耻骨内脏肌的两侧支内缘之间的最大距离)、面积(耻骨联合内侧缘与耻骨内脏肌内侧缘之间的面积)、耻骨内脏肌的厚度(T,耻骨内脏肌两侧支中段内径)、夹角(R, 耻骨内脏肌两侧支在直肠后方形成的角度,以横经垂直线为中心线,当角度两侧线与轮廓最外侧相切,此时的角度即为夹角R)进行描迹。同时由上一步获取的内外轮廓,搜索得到内外轮廓之间的垂直间距,此为耻骨内脏肌的厚度T。
在使用分水岭算法提取肛提肌轮廓后,进一步获取肛提肌轮廓的连续性特征,以及肛提肌裂孔轮廓特征。
在本实施例中,包括进一步判断提肌轮廓的连续性特征中是否出现双侧肛提肌出现连续性局部中断或完全中断,肛提肌裂孔的轮廓特征是否为典型的“U”形或“V”形。
请参考图5,其是本发明具体实施方式中提供的一种三维盆底超声图像处理系统的第一实施例的结构方框图。如图所示,该系统500包括:
获取单元501,用于获取被检查者在Valsalva动作时的多帧三维盆底超声图像;
识别单元502,用于根据盆底的声像学特征识别每帧三维盆底超声图像中的肛提肌以及肛提肌裂孔的组织结构信息;
提取单元503,用于从组织结构信息中提取肛提肌轮廓和肛提肌裂孔轮廓;
计算模块504,用于分别根据每帧所述三维盆底超声图像中肛提肌裂孔的轮廓计算其对应的肛提肌裂孔的面积;
选取单元505,用于将面积最大的三维盆底超声图像作为最大Valsalva动作时的三维盆底超声图像,并将最大Valsalva动作时的三维盆底超声图像作为参考图像。
在一个实施例中,提取单元503用于:
对三维盆底超声图像的灰度图像进行滤波,得到滤波后的灰度图像;
计算滤波后的灰度图像的梯度;
求取梯度的绝对值,选出绝对值的最大值和最小值;
从最小值递增到所述最大值的过程中,根据梯度的绝对值的分布使用递归算法设定局部极小值;
根据局部极小值形成滤波处理后的灰度图像的分水岭;
根据分水岭从肛提肌和肛提肌裂孔的组织结构信息中提取肛提肌轮廓和 肛提肌裂孔轮廓。
在一个实施例中,提取模块503还用于:当分水岭分割的区域出现过度分割时,则对分割结果进行小区域合并。
在一个实施例中,计算模块504还用于根据肛提肌裂孔轮廓内的像素点数按照对应单像素面积比,计算肛提肌裂孔的面积。
在一个实施例中,如图6所示,系统500还包括:
测量模块506,用于对参考图像的肛提肌和肛提肌裂孔进行测量,以获取肛提肌和肛提肌裂孔的测量参数,其中,测量参数包括:肛提肌裂孔的前后径、横径、面积、趾骨内脏肌的厚度和夹角。
测量模块506,还用于根据肛提肌轮廓的连续性特征判定肛提肌损伤。
根据提取的肛提肌轮廓的连续性判断肛提肌是否存在损伤,若双侧肛提肌出现连续性局部中断或完全中断,肛提肌裂孔失去典型的“U”形或“V”形,即可考虑肛提肌损伤,同时根据测量结果来评估肛提肌损伤的程度。如:一般情况下,在最大Valsalva运动时,肛提肌裂孔面积小于25cm2为正常,30-34.9cm2为轻度扩张,35-39.9cm2为中度扩张,大于40cm2为重度扩张。不同的国家及地区的诊断分级略有差异。
本实施例的三维盆底超声图像处理系统500用于实现前述的三维盆底超声图像处理方法,因此三维盆底超声图像处理系统500中的具体实施方式可见前文中的三维盆底超声图像处理方法的实施例部分,例如,获取单元501、识别单元502、提取单元503、计算模块504和选取单元505,分别用于实现上述三维盆底超声图像处理方法中步骤101,102,103,104和105,所以,其具体实施方式可以参照相应的各个部分实施例的描述,在此不再赘述。
以上结合具体实施例描述了本发明的技术原理。这些描述只是为了解释本发明的原理,而不能以任何方式解释为对本发明保护范围的限制。基于此处的解释,本领域的技术人员不需要付出创造性的劳动即可联想到本发明的其它具体实施方式,这些方式都将落入本发明的保护范围之内。

Claims (10)

  1. 一种三维盆底超声图像处理方法,其特征在于,包括:
    获取被检查者在Valsalva动作时的多帧三维盆底超声图像;
    根据盆底的声像学特征识别每帧所述三维盆底超声图像中的肛提肌以及肛提肌裂孔的组织结构信息;
    从所述组织结构信息中提取所述肛提肌轮廓和所述肛提肌裂孔轮廓;
    分别根据每帧所述三维盆底超声图像中所述肛提肌裂孔轮廓计算其对应的所述肛提肌裂孔的面积;
    将所述面积最大的三维盆底超声图像作为最大Valsalva动作时的三维盆底超声图像,并将所述最大Valsalva动作时的三维盆底超声图像作为参考图像。
  2. 根据权利要求1所述的方法,其特征在于,所述从所述组织结构信息中提取所述肛提肌轮廓和所述肛提肌裂孔轮廓的步骤,包括:
    对所述三维盆底超声图像的灰度图像进行滤波,得到滤波后的灰度图像;
    计算所述滤波后的灰度图像的梯度;
    求取所述梯度的绝对值,选出所述绝对值的最大值和最小值;
    从所述最小值递增到所述最大值的过程中,根据所述梯度的绝对值的分布使用递归算法设定局部极小值;
    根据所述局部极小值形成所述滤波处理后的灰度图像的分水岭;
    根据所述分水岭从所述肛提肌和所述肛提肌裂孔的组织结构信息中提取所述肛提肌轮廓和所述肛提肌裂孔轮廓。
  3. 根据权利要求2所述的方法,其特征在于,当所述分水岭分割的区域出现过度分割时,则对分割结果进行小区域合并。
  4. 根据权利要求1所述的方法,其特征在于,所述分别根据每帧所述三维肛提肌超声图像中所述肛提肌裂孔轮廓计算其对应的肛提肌裂孔的面积的步骤包括:
    根据所述肛提肌裂孔轮廓内的像素点数按照对应单像素面积比,计算所述肛提肌裂孔的面积。
  5. 根据权利要求1所述的方法,其特征在于,所述将所述最大Valsalva动作时的三维肛提肌超声图像作为参考图像的步骤之后,还包括:
    对所述参考图像的肛提肌和肛提肌裂孔进行测量,以获取所述肛提肌和所述肛提肌裂孔的测量参数,其中,所述测量参数包括:肛提肌裂孔的前后径、横径、面积、趾骨内脏肌的厚度和夹角。
  6. 一种三维盆底超声图像处理系统,其特征在于,包括:
    获取单元,用于获取被检查者在Valsalva动作时的多帧三维盆底超声图像;
    识别单元,用于根据盆底的声像学特征识别每帧所述三维盆底超声图像中的肛提肌以及肛提肌裂孔的组织结构信息;
    提取单元,用于从所述组织结构信息中提取所述肛提肌轮廓和所述肛提肌裂孔轮廓;
    计算模块,用于分别根据每帧所述三维盆底超声图像中所述肛提肌裂孔轮廓计算其对应的所述肛提肌裂孔的面积;
    选取单元,用于将所述面积最大的三维盆底超声图像作为最大Valsalva动作时的三维盆底超声图像,并将所述最大Valsalva动作时的三维盆底超声图像作为参考图像。
  7. 根据权利要求6所述的系统,其特征在于,所述提取单元用于:
    对所述三维盆底超声图像的灰度图像进行滤波,得到滤波后的灰度图像;
    计算所述滤波后的灰度图像的梯度;
    求取所述梯度的绝对值,选出所述绝对值的最大值和最小值;
    从所述最小值递增到所述最大值的过程中,根据所述梯度的绝对值的分布使用递归算法设定局部极小值;
    根据所述局部极小值形成所述滤波处理后的灰度图像的分水岭;
    根据所述分水岭从所述肛提肌和所述肛提肌裂孔的组织结构信息中提取所述肛提肌轮廓和所述肛提肌裂孔轮廓。
  8. 根据权利要求7所述的系统,其特征在于,所述提取模块还用于:当所述分水岭分割的区域出现过度分割时,则对分割结果进行小区域合并。
  9. 根据权利要求6所述的系统,其特征在于,所述计算模块还用于根据所述肛提肌裂孔轮廓内的像素点数按照对应单像素面积比,计算所述肛提肌裂孔的面积。
  10. 根据权利要求6所述的系统,其特征在于,还包括:
    测量模块,用于对所述参考图像的肛提肌和肛提肌裂孔进行测量,以获取所述肛提肌和所述肛提肌裂孔的测量参数,其中,所述测量参数包括:肛提肌裂孔的前后径、横径、面积、趾骨内脏肌的厚度和夹角;
    测量模块,还用于根据所述肛提肌轮廓的连续性特征判定肛提肌损伤。
PCT/CN2017/093456 2016-12-22 2017-07-19 三维盆底超声图像处理方法及系统 WO2018113282A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201611197301.3A CN106683159B (zh) 2016-12-22 2016-12-22 三维盆底超声图像处理方法及系统
CN201611197301.3 2016-12-22

Publications (1)

Publication Number Publication Date
WO2018113282A1 true WO2018113282A1 (zh) 2018-06-28

Family

ID=58871227

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/093456 WO2018113282A1 (zh) 2016-12-22 2017-07-19 三维盆底超声图像处理方法及系统

Country Status (2)

Country Link
CN (1) CN106683159B (zh)
WO (1) WO2018113282A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111012326A (zh) * 2018-10-09 2020-04-17 深圳市理邦精密仪器股份有限公司 盆底校准方法、设备及计算机可读存储介质
CN114366186A (zh) * 2022-01-17 2022-04-19 中国中医科学院广安门医院 一种肛瘘手术中的挂线打洞辅助系统和方法
EP4299010A1 (en) * 2022-06-27 2024-01-03 Koninklijke Philips N.V. Obtaining ultrasound images of a moving organ

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106683159B (zh) * 2016-12-22 2020-02-18 深圳开立生物医疗科技股份有限公司 三维盆底超声图像处理方法及系统
CN108062749B (zh) * 2017-12-12 2020-04-21 深圳大学 肛提肌裂孔的识别方法、装置和电子设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080086204A1 (en) * 2006-10-06 2008-04-10 Rankin J Scott Intra-annular mounting frame for aortic valve repair
CN103340628A (zh) * 2013-06-28 2013-10-09 中国科学院深圳先进技术研究院 心脏实时电影成像图像处理方法和系统
CN103955912A (zh) * 2014-02-14 2014-07-30 西安电子科技大学 自适应窗的胃部ct图像淋巴结跟踪检测系统及方法
WO2016025323A1 (en) * 2014-08-10 2016-02-18 Autonomix Medical, Inc. Ans assessment systems, kits, and methods
US20160275678A1 (en) * 2015-03-18 2016-09-22 University Of South Florida Image-based automated measurement model to predict pelvic organ prolapse
CN106683159A (zh) * 2016-12-22 2017-05-17 深圳开立生物医疗科技股份有限公司 三维盆底超声图像处理方法及系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080086204A1 (en) * 2006-10-06 2008-04-10 Rankin J Scott Intra-annular mounting frame for aortic valve repair
CN103340628A (zh) * 2013-06-28 2013-10-09 中国科学院深圳先进技术研究院 心脏实时电影成像图像处理方法和系统
CN103955912A (zh) * 2014-02-14 2014-07-30 西安电子科技大学 自适应窗的胃部ct图像淋巴结跟踪检测系统及方法
WO2016025323A1 (en) * 2014-08-10 2016-02-18 Autonomix Medical, Inc. Ans assessment systems, kits, and methods
US20160275678A1 (en) * 2015-03-18 2016-09-22 University Of South Florida Image-based automated measurement model to predict pelvic organ prolapse
CN106683159A (zh) * 2016-12-22 2017-05-17 深圳开立生物医疗科技股份有限公司 三维盆底超声图像处理方法及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
RAMSEY, S.D.: "A REAL-TIME ULTRASONIC B-SCANIDOPPLER ARTERY-IMAGING SYSTEM", ULTRASONICS SYMPOSIUM PROCEEDINGS, 31 December 1975 (1975-12-31), XP055606352 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111012326A (zh) * 2018-10-09 2020-04-17 深圳市理邦精密仪器股份有限公司 盆底校准方法、设备及计算机可读存储介质
CN111012326B (zh) * 2018-10-09 2022-07-05 深圳市理邦精密仪器股份有限公司 盆底校准方法、设备及计算机可读存储介质
CN114366186A (zh) * 2022-01-17 2022-04-19 中国中医科学院广安门医院 一种肛瘘手术中的挂线打洞辅助系统和方法
EP4299010A1 (en) * 2022-06-27 2024-01-03 Koninklijke Philips N.V. Obtaining ultrasound images of a moving organ

Also Published As

Publication number Publication date
CN106683159B (zh) 2020-02-18
CN106683159A (zh) 2017-05-17

Similar Documents

Publication Publication Date Title
WO2018113282A1 (zh) 三维盆底超声图像处理方法及系统
US8699766B2 (en) Method and apparatus for extracting and measuring object of interest from an image
Loizou A review of ultrasound common carotid artery image and video segmentation techniques
US8798344B2 (en) Image processing apparatus, image processing method and computer-readable recording device
CN110448335B (zh) 一种基于超声图像的胎儿头围全自动测量方法和装置
US20110196236A1 (en) System and method of automated gestational age assessment of fetus
KR101121396B1 (ko) 2차원 초음파 영상에 대응하는 2차원 ct 영상을 제공하는 시스템 및 방법
CN100462054C (zh) 胎儿图像性别部位识别屏蔽方法
US20070167760A1 (en) Ultrasound imaging system and method for forming a 3d ultrasound image of a target object
CN105261015B (zh) 基于Gabor滤波器的眼底图像血管自动分割方法
Hiremath et al. Follicle detection in ultrasound images of ovaries using active contours method
WO2011022783A1 (en) Feature detection and measurement in retinal images
CN106570871A (zh) 模糊c均值颈动脉超声图像内中膜厚度测量方法及系统
Ni Ni et al. Anterior chamber angle shape analysis and classification of glaucoma in SS-OCT images
CN108846827B (zh) 一种基于多圆快速分割眼底视盘的方法
Kus et al. Fully automated gradient based breast boundary detection for digitized X-ray mammograms
CN108305247A (zh) 一种基于ct图像灰度值检测组织硬度的方法
CN102663749A (zh) 基于相位信息的超声图像病灶分割技术
Sampath et al. Fuzzy connectedness based segmentation of fetal heart from clinical ultrasound images
Purnama et al. Follicle detection on the usg images to support determination of polycystic ovary syndrome
CN106485712A (zh) 肛提肌轴平面图像处理方法及装置
Hou et al. A Study on Weak Edge Detection of COVID-19's CT Images Based on Histogram Equalization and Improved Canny Algorithm
Massich et al. Automatic seed placement for breast lesion segmentation on US images
Tarroni et al. Fully-automated identification and segmentation of aortic lumen from fetal ultrasound images
Hadzieva et al. Is the fractal dimension of the contour-lines a reliable tool for classification of medical images?

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17884544

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 15.10.2019)

122 Ep: pct application non-entry in european phase

Ref document number: 17884544

Country of ref document: EP

Kind code of ref document: A1