WO2022099882A1 - 超声图像成像质量评价方法、装置及计算机可读存储介质 - Google Patents

超声图像成像质量评价方法、装置及计算机可读存储介质 Download PDF

Info

Publication number
WO2022099882A1
WO2022099882A1 PCT/CN2020/139646 CN2020139646W WO2022099882A1 WO 2022099882 A1 WO2022099882 A1 WO 2022099882A1 CN 2020139646 W CN2020139646 W CN 2020139646W WO 2022099882 A1 WO2022099882 A1 WO 2022099882A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
tissue
brightness
sharpness
tissue deformation
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2020/139646
Other languages
English (en)
French (fr)
Inventor
熊璟
李妍
夏泽洋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Institute of Advanced Technology of CAS
Original Assignee
Shenzhen Institute of Advanced Technology of CAS
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 Shenzhen Institute of Advanced Technology of CAS filed Critical Shenzhen Institute of Advanced Technology of CAS
Publication of WO2022099882A1 publication Critical patent/WO2022099882A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/20—Image enhancement or restoration using local operators
    • G06T5/30—Erosion or dilatation, e.g. thinning
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10132—Ultrasound image
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20036—Morphological image processing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30056—Liver; Hepatic
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30081—Prostate
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30096—Tumor; Lesion

Definitions

  • the invention belongs to the technical field of medical image imaging evaluation, and in particular relates to an ultrasonic image imaging quality evaluation method, an ultrasonic image imaging quality evaluation device and a computer-readable storage medium.
  • Ultrasound imaging technology is widely used in the medical field.
  • ultrasound image-guided therapy equipment can provide visualization information of tissue to doctors or other equipment through ultrasound imaging.
  • percutaneous interventional therapy requires the insertion of a surgical needle into the target area under the guidance of ultrasound images.
  • robot-assisted image-guided technology is also used in the ultrasound image-guided therapy equipment.
  • Robot-assisted image-guided techniques use a trained robotic system to manipulate surgical instruments while using an ultrasound probe to feed back tissue information. During the image acquisition process, it is necessary to adjust the contact state of the ultrasound probe with the tissue to obtain a clear ultrasound image.
  • an imaging quality evaluation method which includes the steps:
  • the specific steps of the evaluation method based on local tissue characteristics include:
  • the extracted foreground image F(i,j) is expressed by the following formula:
  • T is the optimal threshold determined by the maximum inter-class variance criterion, and the optimal threshold satisfies the following formula:
  • ⁇ B 2 is the between-class variance
  • ⁇ W 2 is the intra-class variance
  • P is the gray value array of all pixels in the image
  • the tissue deformation r is expressed as:
  • H(i,j) represents the segmented image containing the tissue shape
  • NH represents the number of pixels in the segmented image
  • i t is the row number of the t-th pixel
  • ⁇ i is the number of tissue pixel rows in the segmented image. average of.
  • the specific steps of using the global image-based evaluation method are:
  • the noise reduction method includes:
  • the image matrix E(i,j) after the etching operation in the step S500 is expressed as:
  • E(i,j) is the image matrix after corrosion
  • B(i,j) is the structural element
  • D B is the definition domain of the structural element
  • ⁇ is the corrosion operator
  • the image matrix D(i,j) expanded by the step S501 is expressed as:
  • D(i,j) is the expanded image matrix
  • B(i,j) is the structuring element
  • D B is the domain of the structuring element
  • the specific steps of extracting tissue edge features are: using the Sobel operator to extract tissue edge features, the obtained template S x in the horizontal direction and template S y in the vertical direction are:
  • the tissue deformation is evaluated according to the discrete degree of edge features, and the tissue deformation r is expressed as:
  • x t is the row number of the t-th pixel in the edge feature extraction image
  • ⁇ x is the average of all edge pixel row numbers in the edge feature extraction image
  • N V is the edge pixel number in the edge feature extraction image.
  • the score for evaluating the total quality of the ultrasound image is:
  • q 1 , q 2 , and q 3 are the normalized values of brightness, sharpness, and tissue deformation values, respectively, and ⁇ 1 , ⁇ 2 , and ⁇ 3 are the brightness scores, sharpness scores, and tissue deformation scores, respectively.
  • the weights, Q 1 and Q 2 are the brightness threshold and sharpness threshold of the ultrasound image, respectively.
  • the present invention also provides an ultrasonic image imaging quality evaluation device, which includes a brightness and sharpness acquisition module, a tissue deformation evaluation module and an ultrasonic image total quality evaluation module, wherein the brightness and sharpness acquisition module is connected to the tissue deformation evaluation module , the ultrasound image total quality evaluation module is respectively connected with the brightness and sharpness acquisition module and the tissue deformation evaluation module;
  • the brightness and sharpness acquisition module is used to calculate the brightness and sharpness of the ultrasound image
  • the tissue deformation evaluation module is used to select a tissue deformation evaluation method from an evaluation method based on local tissue characteristics and an evaluation method based on a global image;
  • the tissue deformation assessment module is further configured to output the assessed tissue deformation amount to the ultrasound image total quality assessment module;
  • the overall quality evaluation module of the ultrasound image evaluates the overall quality of the ultrasound image through the acquired three parameters of image brightness, image sharpness and soft tissue deformation.
  • the present invention also provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, implements the steps of the above-mentioned ultrasonic image imaging quality evaluation method.
  • the ultrasonic image imaging quality evaluation method provided by the present invention can comprehensively evaluate the ultrasonic image by integrating the three parameters of image brightness, image sharpness and soft tissue deformation degree, which solves the deficiencies in the prior art.
  • the ultrasonic image imaging quality evaluation method can use two different soft tissue deformation evaluation methods according to the difference of soft tissue structure characteristics, so as to improve the detection efficiency and accuracy.
  • Embodiment 1 is a flowchart of a method for evaluating the imaging quality of an ultrasound image in Embodiment 1 of the present invention
  • FIG. 2 is a flow chart of a method for evaluating tissue deformation based on local tissue characteristics in Embodiment 3 of the present invention
  • Embodiment 3 is a flowchart of an evaluation method based on a global image in Embodiment 3 of the present invention.
  • Embodiment 4 is a flowchart of a noise reduction method in Embodiment 3 of the present invention.
  • FIG. 5 is a schematic structural diagram of an ultrasonic image imaging quality evaluation device in Embodiment 5 of the present invention.
  • Fig. 6 is the evaluation score curve diagram of the phantom image of the built-in cyst in the fifth embodiment of the present invention.
  • FIG. 7 is a partial ultrasound image of a phantom with a cyst embedded in the fifth embodiment of the present invention.
  • Brightness and sharpness acquisition module 1 tissue deformation assessment module 2 and ultrasound image total quality assessment module 3.
  • FIG. 1 is a flowchart of a method for evaluating the imaging quality of an ultrasound image in Embodiment 1 of the present invention. As shown in Figure 1, the present invention provides a method for evaluating the imaging quality of an ultrasound image, which includes the steps:
  • the method for evaluating the imaging quality of an ultrasonic image provided by the present invention can comprehensively evaluate the ultrasonic image by integrating the three parameters of image brightness, image sharpness and soft tissue deformation degree, which solves the deficiencies in the prior art.
  • the ultrasonic image imaging quality evaluation method can use two different soft tissue deformation evaluation methods according to the difference of soft tissue structure characteristics, so as to improve the detection efficiency and accuracy.
  • the brightness is calculated by the following formula:
  • I(i,j) represents the pixel intensity of image I at (i,j).
  • M and N represent the number of horizontal pixels and the number of vertical pixels, respectively.
  • the gradient magnitude map is constructed using the Scharr gradient operator.
  • the partial derivative g x (i,j) in the horizontal direction and the partial derivative g y (i,j) in the vertical direction are:
  • the gradient size GA(i,j) of the image is:
  • step S3 before evaluating the degree of deformation of the soft tissue, different evaluation methods need to be adopted according to the distribution characteristics of the tissues or glands in the ultrasound image.
  • the method based on local tissue features has higher sensitivity and discrimination for evaluating tissue deformation. Therefore, in order to be suitable for ultrasound images of different tissue structures, the two soft tissue deformation evaluation methods mentioned in the present invention will be determined according to the following formulas, that is, if the segmented image satisfies the following formula, the method based on local tissue characteristics will be used to evaluate tissue deformation , otherwise a global image-based approach was used to assess tissue deformation.
  • the ultrasound image imaging quality evaluation method makes an automatic decision on the selection of soft tissue evaluation methods.
  • FIG. 2 is a flowchart of a method for evaluating tissue deformation based on local tissue features in this embodiment. As shown in Figure 2, the specific steps of the tissue deformation assessment method based on local tissue features include:
  • step S40 the foreground image F(i,j) extracted from the grayscale image is expressed by the following formula:
  • T is the optimal threshold determined by the maximum inter-class variance criterion. And the optimal threshold satisfies the following formula:
  • ⁇ B 2 is the between-class variance
  • ⁇ W 2 is the intra-class variance
  • P is the gray value array of all pixels in the image.
  • step S42 when the tissue is deformed, the aggregation degree of the tissue in the vertical direction will change, so the tissue deformation can be evaluated by observing the height change of the segmented tissue shape.
  • the tissue deformation r can be expressed by the following formula:
  • H(i,j) represents the segmented image containing the tissue shape
  • NH represents the number of pixels of the segmented image
  • i t is the row number of the t-th pixel
  • ⁇ i is the average number of tissue pixel rows in the segmented image value.
  • FIG. 3 is a flowchart of the evaluation method based on the global image in this embodiment. As shown in Figure 3, the specific steps of adopting the evaluation method based on the global image in step S5 are:
  • the image is denoised by using a morphological algorithm.
  • FIG. 4 is a flowchart of the noise reduction method in this embodiment. As shown in Figure 4, the noise reduction method in step S20 includes:
  • the image matrix E(i,j) after the etching operation in step S500 is expressed as:
  • E(i,j) is the image matrix after corrosion
  • B(i,j) is the structuring element
  • D B is the definition domain of the structuring element
  • ⁇ is the erosion operator.
  • the expanded image matrix D(i,j) in step S501 is expressed as:
  • D(i,j) is the image matrix after expansion, is the dilation operator, B(i,j) is the structuring element, and D B is the domain of the structuring element.
  • Step S51 the specific steps of proposing tissue edge features are:
  • the template S x in the horizontal direction and the template S y in the vertical direction are:
  • step S52 can be performed, that is, the tissue deformation is evaluated according to the discrete degree of the edge feature, and the tissue deformation r is expressed as:
  • x t is the row number of the t-th pixel in the edge feature extraction image
  • ⁇ x is the average of all edge pixel row numbers in the edge feature extraction image
  • N V is the edge pixel number in the edge feature extraction image.
  • the total quality score of the ultrasound image is composed of three parts: brightness score, sharpness score and tissue deformation score. Brightness, sharpness and tissue deformation have equal weights in the total quality score.
  • the total image quality score obtained by normalizing the brightness value, sharpness value, and tissue deformation value is:
  • q 1 , q 2 , and q 3 are the normalized values of brightness, sharpness, and tissue deformation, respectively, and ⁇ 1 , ⁇ 2 , and ⁇ 3 are the weights of brightness, sharpness, and tissue deformation, respectively , Q 1 and Q 2 are the brightness threshold and sharpness threshold of the ultrasound image, respectively.
  • the image brightness and sharpness are lower than the threshold, the tissue or gland in the image is not obvious, and the tissue or gland cannot be observed in a given pixel.
  • the contour shows that the ultrasound probe is not in sufficient contact with the tissue at this time, and no deformation occurs. Therefore, only the brightness quality and sharpness quality of the ultrasound image are considered in this case.
  • FIG. 5 is a schematic structural diagram of an apparatus for evaluating the imaging quality of an ultrasound image in Embodiment 5 of the present invention.
  • the present invention further provides an ultrasonic image imaging quality evaluation apparatus, which can execute the ultrasonic image imaging quality evaluation methods described in Embodiment 1 to Embodiment 4.
  • the device includes a brightness and sharpness acquisition module 1 , a tissue deformation assessment module 2 , and an ultrasound image overall quality assessment module 3 .
  • the brightness and sharpness acquisition module 1 is connected with the tissue deformation assessment module 2 .
  • the ultrasound image total quality evaluation module 3 is respectively connected with the brightness and sharpness acquisition module 1 and the tissue deformation evaluation module 2 .
  • the brightness and sharpness acquisition module 1 is used to calculate the brightness and sharpness of the ultrasound image.
  • the tissue deformation evaluation module 2 is used to select a tissue deformation evaluation method from an evaluation method based on local tissue features and an evaluation method based on a global image.
  • the tissue deformation evaluation module 2 is also used to output the estimated tissue deformation amount to the ultrasound image total quality evaluation module 6 .
  • the overall quality evaluation module 3 of the ultrasound image evaluates the overall quality of the ultrasound image through the acquired three parameters of image brightness, image sharpness and soft tissue deformation.
  • the device was experimentally tested on ultrasound images of prostate phantoms, porcine liver tissue, and phantoms with built-in cysts, and could achieve real-time evaluation of ultrasound images with an average evaluation time of 66 milliseconds. Taking the ultrasound images of the phantom with built-in cysts as an example, a total of 20 ultrasound images were collected.
  • FIG. 6 is the evaluation score of the phantom image of the built-in cyst in this example.
  • Figure 7 is a partial ultrasound image of the cyst-embedded phantom in this example. The evaluation scores are shown in Figure 6, and Figure 7 is a part of the acquired ultrasound images. The six images in a-e shown in Figure 7 correspond to the phantom with the first cyst, the phantom with the sixth cyst, and the eighth image. Phantom No. 10 with endocyst, phantom No. 10 with endocyst, No. 12 with endocyst, and No. 15 with endocyst.
  • the phantom with built-in cyst is an ultrasound phantom with cyst tissue features inside.
  • 20 ultrasound images of the phantom were collected. From the 1st to the 20th, the ultrasonic probe and the phantom gradually increase the contact tightness from the slight contact, and then use the above evaluation method to calculate the evaluation score.
  • the ultrasound probe and the phantom are in light contact state corresponding to Fig. 7a.
  • This embodiment provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, implements the steps of the above-mentioned ultrasonic image imaging quality evaluation method.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Ultra Sonic Daignosis Equipment (AREA)

Abstract

本发明提供了一种超声图像成像质量评价方法、装置及计算机可读存储介质，其方法包括步骤：S1、计算超声图像的亮度和锐度；S2、根据亮度值和锐度值判断给定像素内是否能观测到腺体或组织形态；S3、根据图像中组织分布结构来匹配相应的组织形变评估方法，若组织中存在符合预设条件的腺体结构，则执行S4；否则执行S5；S4、采用基于局部组织特征的评估方法评估组织变形，然后执行S6；S5、采用基于全局图像的评估方法评估组织变形，然后执行S6；S6、基于图像亮度、图像锐度和软组织形变三个参数评估超声图像的总质量。该方法能够融合图像亮度、图像锐度以及软组织形变程度3个参数对超声图像进行综合评估。

Description

超声图像成像质量评价方法、装置及计算机可读存储介质 技术领域
本发明属于医疗影像成像评价技术领域，尤其涉及一种超声图像成像质量评价方法、一种超声图像成像质量评价装置及计算机可读存储介质。
背景技术
超声影像技术在医疗领域应用广泛。其中，超声影像引导治疗设备可以通过超声成像向医生或其他设备提供组织的可视化信息。例如，经皮介入治疗中需要在超声影像的引导下将手术针插入到目标区域。
另外，在超声影像引导治疗设备上还运用了机器人辅助的影像引导技术。机器人辅助的影像引导技术是使用经过训练的机器人系统操作手术器械，同时利用超声探头反馈组织信息。在图像采集过程中，需调整超声探头与组织接触状态，以获得清晰的超声图像。
因此，在机器人辅助的超声引导过程中，对超声图像的图像质量进行评估很有必要。
然而，目前已有的图像质量评估只考虑图像成像质量，并不能评估图像中软组织的形变程度。在实际的临床病例中，显然近距离探测可获得清晰的图像，但是超声探头与组织过近接触会引起软组织形变。这种现象会降低靶向穿刺的精确度。为了避免这种情况，临床中需要在软组织形变程度最小的前提下获得清晰的超声图像。
因此，急需设计一种能够考虑软组织形变的实时超声图像质量评估方法，用于机器人辅助超声影像引导治疗。
发明内容
有鉴于此，为了解决现有超声图像成像评价方法只能对图像的成像质量进行评估而不能量化图像中的软组织形变程度而导致的靶向穿刺的精度不足的问题，本发明提供一种超声图像成像质量评价方法，其包括步骤：
S1、计算超声图像的亮度和锐度；
S2、根据亮度值和锐度值判断给定像素内是否能观测到腺体或组织形态；
S3、根据图像中组织分布结构来匹配相应的组织形变评估方法，若组织中存在符合预设条件的腺体结构，则执行S4；否则执行S5；
S4、采用基于局部组织特征的评估方法评估组织变形，然后执行S6；
S5、采用基于全局图像的评估方法评估组织变形，然后执行S6；
S6、基于图像亮度、图像锐度和软组织形变三个参数评估超声图像的总质量。
较佳地，所述步骤S4中，所述基于局部组织特征的评估方法的具体步骤包括：
S40、对超声图像进行提取前景操作，排除背景特征；
S41、对前景图像中的连通域进行标记和提取，并保留最大连通域，即分割目标组织形状；
S42、执行聚集特性分析，获取组织形变量。
较佳地，在所述步骤S40中，提取的前景图像F(i,j)通过如下公式表示：
Figure PCTCN2020139646-appb-000001
上式中，T为最大类间方差准则确定的最佳阈值，且最佳阈值满足下式：
Figure PCTCN2020139646-appb-000002
上式中，σ B 2为类间方差，σ W 2为类内方差,P为图像所有像素点的灰度值数组；
在所述步骤S42中，组织形变r表示为：
Figure PCTCN2020139646-appb-000003
上式中，H(i,j)表示含有组织形状的分割图像，N H表示分割图像的像素个数，i t为第t个像素的行数，μ i为分割图像中组织像素点行数的平均值。
较佳地，所述步骤S5中，所述采用基于全局图像的评估方法的具体步骤为：
S50、将图像采用形态学算法进行降噪；
S51、提取组织边缘特征；
S52、执行聚集特性分析，获取组织形变量。
较佳地，所述步骤S50中，所述降噪的方法包括：
S500，对图像进行腐蚀操作，用以消除斑点噪声；
S501，对图像进行膨胀运算，将图像和结构元素进行卷积运算，填补图像间隙。
较佳地，经所述步骤S500腐蚀操作后的图像矩阵E(i,j)表示为：
Figure PCTCN2020139646-appb-000004
上式中，E(i,j)为腐蚀后的图像矩阵，B(i,j)是结构元素，D B是结构元素的定义域，⊙是腐蚀运算符；
经所述步骤S501膨胀后的图像矩阵D(i,j)表示为：
Figure PCTCN2020139646-appb-000005
上式中，D(i,j)为膨胀后的图像矩阵，
Figure PCTCN2020139646-appb-000006
是膨胀运算符，B(i,j)是结构元素，D B是结构元素的定义域。
较佳地，在所述步骤S51中，提取组织边缘特征的具体步骤为：利用Sobel算子提取组织边缘特征，所获取的水平方向的模板S x和垂直方向的模板S y为：
Figure PCTCN2020139646-appb-000007
Figure PCTCN2020139646-appb-000008
在所述步骤S52中，根据边缘特征的离散程度评估组织变形，组织形变r表示为：
Figure PCTCN2020139646-appb-000009
其中，x t为边缘特征提取图像中第t个像素的行数，μ x为边缘特征提取图像中所有边缘像素点行数的平均值,N V为边缘特征提取图像中边缘像素点个数。
较佳地，所述步骤S6中，评价超声图像的总质量的分数为：
Figure PCTCN2020139646-appb-000010
其中，q 1、q 2、q 3分别为亮度值、锐度值和组织变形值归一化后的数值，λ 1、λ 2、λ 3分别为亮度分数、锐度分数、组织形变分数所占权重，Q 1和Q 2分别为超声图像的亮度阈值和锐度阈值。
本发明还提供一种超声图像成像质量评价装置，其包括亮度和锐度获取模块、 组织变形评估模块和超声图像总质量评价模块，所述亮度和锐度获取模块与所述组织变形评估模块连接，所述超声图像总质量评价模块分别与所述亮度和锐度获取模块和所述组织变形评估模块连接；
所述亮度和锐度获取模块用于计算超声图像的亮度和锐度；
所述组织变形评估模块用于从采用基于局部组织特征的评估方法和采用基于全局图像的评估方法中选取组织形变评估方法；
所述组织变形评估模块还用于向所述超声图像总质量评价模块输出其所评估的组织变形量；
所述超声图像总质量评价模块通过获取到的图像亮度、图像锐度和软组织形变三个参数评估超声图像的总质量。
本发明还提供一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行时实现上述超声图像成像质量评价方法的步骤。
本发明实施例与现有技术相比存在的有益效果是：
1、本发明提供的超声图像成像质量评价方法能够融合图像亮度、图像锐度以及软组织形变程度3个参数对超声图像进行综合评估，解决了现有技术中的不足。
2、该超声图像成像质量评价方法能够针对软组织结构特征的差异，使用两种不同的软组织形变评估方法，使检测效率和准确性提高。
附图说明
为了更清楚地说明本发明实施例中的技术方案，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性的前提下，还可以根据这些附图获得其他的附图。
图1是本发明实施例一中超声图像成像质量评价方法的流程图；
图2是本发明实施例三中基于局部组织特征的组织形变评估方法的流程图；
图3是本发明实施例三中基于全局图像的评估方法的流程图；
图4是本发明实施例三中的降噪方法的流程图；
图5本发明实施例五中超声图像成像质量评价装置的结构示意图；
图6是本发明实施例五中内置囊肿的体模图像的评价分数曲线图；
图7是本发明实施例五中内置囊肿的体模的部分超声图像。
附图标记：
亮度和锐度获取模块1、组织变形评估模块2和超声图像总质量评价模块3。
具体实施方式
以下描述中，为了说明而不是为了限定，提出了诸如特定系统结构、技术之类的具体细节，以便透彻理解本发明实施例。然而，本领域的技术人员应当清楚，在没有这些具体细节的其它实施例中也可以实现本发明。在其它情况中，省略对众所周知的系统、装置、电路以及方法的详细说明，以免不必要的细节妨碍本发明的描述。
为了说明本发明所述的技术方案，下面通过具体实施例来进行说明。
应理解，上述实施例中各步骤的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本发明实施例的实施过程 构成任何限定。
实施例一
图1是本发明实施例一中超声图像成像质量评价方法的流程图。如图1所示，本发明提供了一种超声图像成像质量评价方法，其包括步骤：
S1、计算超声图像的亮度和锐度；
S2、根据亮度值和锐度值判断给定像素内是否能观测到腺体或组织形态；
S3、根据图像中组织分布结构来匹配相应的组织形变评估方法，若组织中存在符合预设条件的腺体结构，则执行S4；否则执行S5；
S4、采用基于局部组织特征的评估方法评估组织变形，然后执行S6；
S5、采用基于全局图像的评估方法评估组织变形，然后执行S6；
S6、基于图像亮度、图像锐度和软组织形变三个参数评估超声图像的总质量。
本发明实施例与现有技术相比存在的有益效果是：
1、本发明提供的一种超声图像成像质量评价方法能够融合图像亮度、图像锐度以及软组织形变程度3个参数对超声图像进行综合评估，解决了现有技术中的不足。
2、该超声图像成像质量评价方法能够针对软组织结构特征的差异，使用两种不同的软组织形变评估方法，使检测效率和准确性提高。
实施例二
本实施例与实施例一的不同之处在于：
步骤S1中的计算超声图像的亮度，其亮度通过如下公式计算：
Figure PCTCN2020139646-appb-000011
其中，I(i,j)表示图像I在(i,j)处的像素强度。M和N分别代表横向像素数量和纵向像素数量。
采用Scharr梯度算子构造梯度幅值映射。水平方向的偏导数g x(i,j)和垂直方向的偏导数g y(i,j)为：
Figure PCTCN2020139646-appb-000012
Figure PCTCN2020139646-appb-000013
图像的梯度大小GA(i,j)为：
Figure PCTCN2020139646-appb-000014
实施例三
本实施例与实施例一的不同之处在于：
在步骤S3中，在对软组织进行形变程度评估前，需要根据超声图像中组织或腺体分布的特点采取不同的评估方法。对于有明显腺体结构的超声图像，采用基于局部组织特征的方法评估组织形变具有更高的灵敏性和区分度。因此，为适用于不同组织结构的超声图形，本发明所提及的两种软组织形变评估方法将根据如下公式进行决策，即若分割图像满足如下式，则采用基于局部组织特征的方法评估组织变形，否则采用基于全局图像的方法评估组织变形。该超声图像成像质量评价方法对软组织评价方法的选择进行自动决策。
Figure PCTCN2020139646-appb-000015
在步骤S4中，基于局部组织特征的组织形变评估方法只关注组织形态变化。图2是本实施例中基于局部组织特征的组织形变评估方法的流程图。如图2所示， 基于局部组织特征的组织形变评估方法的具体步骤包括：
S40、对超声图像进行提取前景操作，排除背景特征；
S41、对前景图像中的连通域进行标记和提取，并保留最大连通域，即分割目标组织形状；
S42、执行聚集特性分析，获取组织形变量。
步骤S40中，从灰度图像中提取的前景图像F(i,j)通过如下公式表示：
Figure PCTCN2020139646-appb-000016
式中，T为最大类间方差准则确定的最佳阈值。且最佳阈值满足下式：
Figure PCTCN2020139646-appb-000017
其中σ B 2为类间方差，σ W 2为类内方差,P为图像所有像素点的灰度值数组。
步骤S42中，当组织产生形变时，组织竖直方向的聚集程度会发生变化，因此可以通过观测分割出的组织形状高度变化评估组织形变。组织形变r可由下式表示：
Figure PCTCN2020139646-appb-000018
其中，H(i,j)表示含有组织形状的分割图像，N H表示分割图像的像素个数，i t为第t个像素的行数，μ i为分割图像中组织像素点行数的平均值。
图3是本实施例中基于全局图像的评估方法的流程图。如图3所示，在步骤S5中采用基于全局图像的评估方法的具体步骤为：
S50、将图像采用形态学算法进行降噪。
S51、提出组织边缘特征；
S52、执行聚集特性分析，获取组织形变量。
图4是本实施例中的降噪方法的流程图。如图4所示，步骤S20中的降噪方法包括：
S500，对图像进行腐蚀操作，用以消除斑点噪声；
S501，对图像进行膨胀运算，将图像和结构元素进行卷积运算，填补图像间隙。
经步骤S500腐蚀操作后的图像矩阵E(i,j)表示为：
Figure PCTCN2020139646-appb-000019
其中，E(i,j)为腐蚀后的图像矩阵，B(i,j)是结构元素，D B是结构元素的定义域，⊙是腐蚀运算符。
经步骤S501膨胀后的图像矩阵D(i,j)表示为：
Figure PCTCN2020139646-appb-000020
其中，D(i,j)为膨胀后的图像矩阵，
Figure PCTCN2020139646-appb-000021
是膨胀运算符，B(i,j)是结构元素，D B是结构元素的定义域。
步骤S51、提出组织边缘特征的具体步骤为：
对图像进行降噪处理后，利用Sobel算子提取组织边缘特征，水平方向的模板S x和垂直方向的模板S y为：
Figure PCTCN2020139646-appb-000022
Figure PCTCN2020139646-appb-000023
获得边缘特征提取图像后可以执行步骤S52，即根据边缘特征的离散程度评估组织变形，组织形变r表示为：
Figure PCTCN2020139646-appb-000024
其中，x t为边缘特征提取图像中第t个像素的行数，μ x为边缘特征提取图像中所有边缘像素点行数的平均值,N V为边缘特征提取图像中边缘像素点个数。
实施例四
本实施例与实施例一的不同之处在于：
超声图像的总质量分数由亮度分数、锐度分数和组织形变分数三部分组成。亮度、锐度和组织形变在总质量分数中所占的权重相等。将亮度值、锐度值和组织形变值归一化后可得图像质量的总分数为：
Figure PCTCN2020139646-appb-000025
q 1、q 2、q 3分别为亮度值、锐度值和组织变形值归一化后的数值，λ 1、λ 2、λ 3分别为亮度分数、锐度分数、组织形变分数所占权重，Q 1和Q 2分别为超声图像的亮度阈值和锐度阈值，当图像亮度和锐度低于该阈值时，图像中组织或腺体不明显，在给定像素内无法观测组织或腺体轮廓，表明此时超声探头并未与组织充分接触，没有形变产生。因此，这种情况下只考虑超声图像的亮度质量和锐度质量。
实施例五
本实施例与实施例一至四的不同之处在于：
图5是本发明实施例五中超声图像成像质量评价装置的结构示意图。如图5所示，本发明还提供一种超声图像成像质量评价装置，其能够执行实施例一至实 施例四所述的超声图像成像质量评价方法。该装置包括亮度和锐度获取模块1、组织变形评估模块2、超声图像总质量评价模块3。亮度和锐度获取模块1与组织变形评估模块2连接。超声图像总质量评价模块3分别与亮度和锐度获取模块1和组织变形评估模块2连接。
亮度和锐度获取模块1用于计算超声图像的亮度和锐度。
组织变形评估模块2用于从采用基于局部组织特征的评估方法和采用基于全局图像的评估方法中选取组织形变评估方法。
组织变形评估模块2还用于向超声图像总质量评价模块6输出其所评估的组织变形量。
超声图像总质量评价模块3通过获取到的图像亮度、图像锐度和软组织形变三个参数评估超声图像的总质量。
该装置在前列腺体模、猪肝脏组织、内置囊肿的体模的超声图像进行了实验测试，可以实现超声图像的实时评估，平均评估时间为66毫秒。以内置囊肿的体模的超声图像为例，共采集20张超声图像。
图6是本实施例中内置囊肿的体模图像的评价分数。图7是本实施例中内置囊肿的体模的部分超声图像。评估分数如图6所示，图7是采集的部分超声图像，图7中所示的a-e中六张图像依次对应第一号内置囊肿的体模、第六号内置囊肿的体模、第八号内置囊肿的体模、第十号内置囊肿的体模、第十二号内置囊肿的体模和第十五号内置囊肿的体模。
内置囊肿的体模是一种超声体模，内部有囊肿组织特征，此处采集了该体模的超声图像20张。从第1张至第20张，超声探头与体模从轻微接触开始逐渐增加接触紧密程度，然后采用上述评价方法计算评价分数。超声探头与体模从轻微 接触状态对应图7a。
从图6可以看出，随着接触增加，图像亮度和锐度都逐渐增加；接触增加到一定程度时，如图7b所示，图像中可以观测到囊肿组织轮廓。随后接触越来越紧密，组织开始发生形变，组织形变分数逐渐减小，表明组织形变程度逐渐增大，如图7c-图7f所示。
实施例六
本实施例提供一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行时实现上述超声图像成像质量评价方法的步骤。
以上所述实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围，均应包含在本发明的保护范围之内。

Claims (10)

  1. 一种超声图像成像质量评价方法，其特征在于，其包括步骤：
    S1、计算超声图像的亮度和锐度；
    S2、根据亮度值和锐度值判断给定像素内是否能观测到腺体或组织形态；
    S3、根据图像中组织分布结构来匹配相应的组织形变评估方法，若组织中存在符合预设条件的腺体结构，则执行S4；否则执行S5；
    S4、采用基于局部组织特征的评估方法评估组织变形，然后执行S6；
    S5、采用基于全局图像的评估方法评估组织变形，然后执行S6；
    S6、基于图像亮度、图像锐度和软组织形变三个参数评估超声图像的总质量。
  2. 如权利要求1所述的一种超声图像成像质量评价方法，其特征在于，所述步骤S4中，所述基于局部组织特征的评估方法的具体步骤包括：
    S40、对超声图像进行提取前景操作，排除背景特征；
    S41、对前景图像中的连通域进行标记和提取，并保留最大连通域，即分割目标组织形状；
    S42、执行聚集特性分析，获取组织形变量。
  3. 如权利要求2所述的一种超声图像成像质量评价方法，其特征在于，在所述步骤S40中，提取的前景图像F(i,j)通过如下公式表示：
    Figure PCTCN2020139646-appb-100001
    上式中，T为最大类间方差准则确定的最佳阈值，且最佳阈值满足下式：
    Figure PCTCN2020139646-appb-100002
    上式中，σ B 2为类间方差，σ W 2为类内方差,P为图像所有像素点的灰度值数组；
    在所述步骤S42中，组织形变r表示为：
    Figure PCTCN2020139646-appb-100003
    上式中，H(i,j)表示含有组织形状的分割图像，N H表示分割图像的像素个数，i t为第t个像素的行数，μ i为分割图像中组织像素点行数的平均值。
  4. 如权利要求1所述的一种超声图像成像质量评价方法，其特征在于，所述步骤S5中，所述采用基于全局图像的评估方法的具体步骤为：
    S50、将图像采用形态学算法进行降噪；
    S51、提取组织边缘特征；
    S52、执行聚集特性分析，获取组织形变量。
  5. 如权利要求4所述的一种超声图像成像质量评价方法，其特征在于，所述步骤S50中，所述降噪的方法包括：
    S500，对图像进行腐蚀操作，用以消除斑点噪声；
    S501，对图像进行膨胀运算，将图像和结构元素进行卷积运算，填补图像间隙。
  6. 如权利要求5所述的一种超声图像成像质量评价方法，其特征在于，经所述步骤S500腐蚀操作后的图像矩阵E(i,j)表示为：
    Figure PCTCN2020139646-appb-100004
    上式中，E(i,j)为腐蚀后的图像矩阵，B(i,j)是结构元素，D B是结构元素的定义域，⊙是腐蚀运算符；
    经所述步骤S501膨胀后的图像矩阵D(i,j)表示为：
    Figure PCTCN2020139646-appb-100005
    上式中，D(i,j)为膨胀后的图像矩阵，
    Figure PCTCN2020139646-appb-100006
    是膨胀运算符，B(i,j)是结构元素， D B是结构元素的定义域。
  7. 如权利要求4至6任一项所述的一种超声图像成像质量评价方法，其特征在于，在所述步骤S51中，提取组织边缘特征的具体步骤为：利用Sobel算子提取组织边缘特征，所获取的水平方向的模板S x和垂直方向的模板S y为：
    Figure PCTCN2020139646-appb-100007
    Figure PCTCN2020139646-appb-100008
    在所述步骤S52中，根据边缘特征的离散程度评估组织变形，组织形变r表示为：
    Figure PCTCN2020139646-appb-100009
    其中，x t为边缘特征提取图像中第t个像素的行数，μ x为边缘特征提取图像中所有边缘像素点行数的平均值,N V为边缘特征提取图像中边缘像素点个数。
  8. 如权利要求1至6任一项所述的一种超声图像成像质量评价方法，其特征在于，所述步骤S6中，评价超声图像的总质量的分数为：
    Figure PCTCN2020139646-appb-100010
    其中，q 1、q 2、q 3分别为亮度值、锐度值和组织变形值归一化后的数值，λ 1、λ 2、λ 3分别为亮度分数、锐度分数、组织形变分数所占权重，Q 1和Q 2分别为超声图像的亮度阈值和锐度阈值。
  9. 一种超声图像成像质量评价装置，其特征在于，其包括亮度和锐度获取 模块、组织变形评估模块和超声图像总质量评价模块，所述亮度和锐度获取模块与所述组织变形评估模块连接，所述超声图像总质量评价模块分别与所述亮度和锐度获取模块和所述组织变形评估模块连接；
    所述亮度和锐度获取模块用于计算超声图像的亮度和锐度；
    所述组织变形评估模块用于从采用基于局部组织特征的评估方法和采用基于全局图像的评估方法中选取组织形变评估方法；
    所述组织变形评估模块还用于向所述超声图像总质量评价模块输出其所评估的组织变形量；
    所述超声图像总质量评价模块通过获取到的图像亮度、图像锐度和软组织形变三个参数评估超声图像的总质量。
  10. 一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现如权利要求1至8任一项所述超声图像成像质量评价方法的步骤。
PCT/CN2020/139646 2020-11-13 2020-12-25 超声图像成像质量评价方法、装置及计算机可读存储介质 Ceased WO2022099882A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011267333.2A CN112258503B (zh) 2020-11-13 2020-11-13 超声图像成像质量评价方法、装置及计算机可读存储介质
CN202011267333.2 2020-11-13

Publications (1)

Publication Number Publication Date
WO2022099882A1 true WO2022099882A1 (zh) 2022-05-19

Family

ID=74265516

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/139646 Ceased WO2022099882A1 (zh) 2020-11-13 2020-12-25 超声图像成像质量评价方法、装置及计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN112258503B (zh)
WO (1) WO2022099882A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115082487A (zh) * 2022-08-23 2022-09-20 深圳华声医疗技术股份有限公司 超声图像切面质量评价方法、装置、超声设备及存储介质
CN116242317A (zh) * 2022-11-24 2023-06-09 中国空间技术研究院 一种敏捷光学卫星线阵动中成像方法
CN120198787A (zh) * 2025-05-26 2025-06-24 中国人民解放军海军航空大学 面向边缘结构清晰度的sar图像质量评价方法
CN120748635A (zh) * 2025-09-05 2025-10-03 西安医学院第一附属医院 一种交互式麻醉穿刺辅助指导方法及系统

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113674241B (zh) * 2021-08-17 2024-07-23 Oppo广东移动通信有限公司 选帧方法、装置、计算机设备和存储介质
CN114881970A (zh) * 2022-05-10 2022-08-09 无锡祥生医疗科技股份有限公司 超声图像质量评估方法、装置和存储介质
CN115644933B (zh) * 2022-11-17 2023-08-22 深圳微创踪影医疗装备有限公司 导管冲刷控制方法、装置、计算机设备、存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102067176A (zh) * 2008-06-18 2011-05-18 皇家飞利浦电子股份有限公司 结合局部运动监测、校正和评估的辐射成像
CN107753061A (zh) * 2017-11-03 2018-03-06 飞依诺科技(苏州)有限公司 超声弹性成像的自动优化方法及系统
CN108471997A (zh) * 2015-10-28 2018-08-31 美敦力导航股份有限公司 用于在最小化患者的x射线剂量的同时维持图像质量的装置和方法
CN111513754A (zh) * 2019-09-16 2020-08-11 深圳迈瑞生物医疗电子股份有限公司 一种超声成像设备、超声图像的质量评估方法

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6405138B2 (ja) * 2014-07-15 2018-10-17 オリンパス株式会社 画像処理装置、画像処理方法、及び画像処理プログラム
CN104732227B (zh) * 2015-03-23 2017-12-26 中山大学 一种基于清晰度和亮度评估的车牌快速定位方法
CN105261013B (zh) * 2015-09-25 2018-04-13 孙高磊 一种扫描图像质量综合评价方法及评价系统
KR101781009B1 (ko) * 2016-08-31 2017-10-23 노틸러스효성 주식회사 낡은 지폐 식별 방법
CN109584198B (zh) * 2017-09-26 2022-12-23 浙江宇视科技有限公司 一种人脸图像质量评价方法、装置及计算机可读存储介质
CN109544540B (zh) * 2018-11-28 2020-12-25 东北大学 一种基于图像分析技术的糖尿病视网膜图像质量检测方法
CN110728651A (zh) * 2019-09-03 2020-01-24 北京爱博同心医学科技有限公司 基于增强现实的管状视野图像变形检测方法及眼镜

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102067176A (zh) * 2008-06-18 2011-05-18 皇家飞利浦电子股份有限公司 结合局部运动监测、校正和评估的辐射成像
CN108471997A (zh) * 2015-10-28 2018-08-31 美敦力导航股份有限公司 用于在最小化患者的x射线剂量的同时维持图像质量的装置和方法
CN107753061A (zh) * 2017-11-03 2018-03-06 飞依诺科技(苏州)有限公司 超声弹性成像的自动优化方法及系统
CN111513754A (zh) * 2019-09-16 2020-08-11 深圳迈瑞生物医疗电子股份有限公司 一种超声成像设备、超声图像的质量评估方法

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115082487A (zh) * 2022-08-23 2022-09-20 深圳华声医疗技术股份有限公司 超声图像切面质量评价方法、装置、超声设备及存储介质
CN115082487B (zh) * 2022-08-23 2022-12-13 深圳华声医疗技术股份有限公司 超声图像切面质量评价方法、装置、超声设备及存储介质
CN116242317A (zh) * 2022-11-24 2023-06-09 中国空间技术研究院 一种敏捷光学卫星线阵动中成像方法
CN120198787A (zh) * 2025-05-26 2025-06-24 中国人民解放军海军航空大学 面向边缘结构清晰度的sar图像质量评价方法
CN120748635A (zh) * 2025-09-05 2025-10-03 西安医学院第一附属医院 一种交互式麻醉穿刺辅助指导方法及系统

Also Published As

Publication number Publication date
CN112258503A (zh) 2021-01-22
CN112258503B (zh) 2023-11-14

Similar Documents

Publication Publication Date Title
WO2022099882A1 (zh) 超声图像成像质量评价方法、装置及计算机可读存储介质
CN111127425B (zh) 基于视网膜眼底图像的目标检测定位方法和装置
CN119963681B (zh) 基于卷积神经网络的mri医学图像校正方法、系统和计算机可读存储介质
US10453191B2 (en) Automated intravascular plaque classification
CN109544540B (zh) 一种基于图像分析技术的糖尿病视网膜图像质量检测方法
CN112529894A (zh) 一种基于深度学习网络的甲状腺结节的诊断方法
US8831311B2 (en) Methods and systems for automated soft tissue segmentation, circumference estimation and plane guidance in fetal abdominal ultrasound images
US20110196236A1 (en) System and method of automated gestational age assessment of fetus
CN109087310B (zh) 睑板腺纹理区域的分割方法、系统、存储介质及智能终端
US20120099771A1 (en) Computer aided detection of architectural distortion in mammography
CN107392928A (zh) 基于MRF‑Snake分割的乳腺癌诊断系统及方法
CN117409002A (zh) 一种用于创伤的视觉识别检测系统及其检测方法
Kanca et al. Learning hand-crafted features for K-NN based skin disease classification
US20210251601A1 (en) Method for ultrasound imaging and related equipment
CN116912823A (zh) 一种基于深度学习的肾脏病理切片细胞识别方法及系统
CN118845225A (zh) 一种微创穿刺手术术中图像血肿状态识别方法及存储介质
CN117952962B (zh) 一种骨密度检测图像处理方法及系统
CN119722659A (zh) 一种基于ai辅助的甲状旁腺识别方法及装置
Somasundaram et al. Fetal head localization and fetal brain segmentation from MRI using the center of gravity
Pavani et al. Quality metric for parasternal long axis b-mode echocardiograms
CN119251192A (zh) 一种超声影像的图像特征提取方法、装置、设备及介质
CN117671237B (zh) 基于深度学习的机器人视网膜下注射针的检测与定位方法
CN116934776B (zh) 一种fish显微图像信号点区域提取方法、装置、设备及介质
CN119205659A (zh) 用于脓毒症患者胃肠超声数据的处理方法
CN116012283B (zh) 一种全自动超声图像量测方法、设备及存储介质

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: 20961435

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20961435

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 20961435

Country of ref document: EP

Kind code of ref document: A1

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 14.12.2023)

122 Ep: pct application non-entry in european phase

Ref document number: 20961435

Country of ref document: EP

Kind code of ref document: A1