CN113344842A - Blood vessel labeling method of ultrasonic image - Google Patents

Blood vessel labeling method of ultrasonic image Download PDF

Info

Publication number
CN113344842A
CN113344842A CN202110312435.XA CN202110312435A CN113344842A CN 113344842 A CN113344842 A CN 113344842A CN 202110312435 A CN202110312435 A CN 202110312435A CN 113344842 A CN113344842 A CN 113344842A
Authority
CN
China
Prior art keywords
image
blood vessel
labeling
region
vessel region
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.)
Pending
Application number
CN202110312435.XA
Other languages
Chinese (zh)
Inventor
齐鹏
陈子杰
陈禹
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.)
Tongji University
Original Assignee
Tongji University
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 Tongji University filed Critical Tongji University
Priority to CN202110312435.XA priority Critical patent/CN113344842A/en
Publication of CN113344842A publication Critical patent/CN113344842A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformation in the plane of the image
    • G06T3/40—Scaling the whole image or part thereof
    • G06T3/4023—Decimation- or insertion-based scaling, e.g. pixel or line decimation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/20—Image enhancement or restoration by the use of local operators
    • G06T5/30—Erosion or dilatation, e.g. thinning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/40—Image enhancement or restoration by the use of histogram techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/64—Analysis of geometric attributes of convexity or concavity
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20112—Image segmentation details
    • G06T2207/20132—Image cropping
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • G06T2207/30101—Blood vessel; Artery; Vein; Vascular
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • G06T2207/30204—Marker

Abstract

The invention relates to a vessel labeling method of an ultrasonic image, which comprises the following steps: acquiring an ultrasonic blood vessel image; selecting a blood vessel area of the ultrasonic blood vessel image for cutting, and interpolating to obtain the size of the image before cutting; performing histogram equalization processing; carrying out binarization treatment; carrying out corrosion and expansion operation; converting the blood vessel regions with the areas larger than the maximum blood vessel region threshold value and smaller than the minimum blood vessel region threshold value into non-blood vessel regions; converting the blood vessel region of which the ratio of the outline area to the convex hull area is smaller than the segmentation threshold of the outline-convex hull area ratio into a non-blood vessel region; converting the vascular region with the aspect ratio larger than the maximum aspect ratio and smaller than the minimum aspect ratio into a non-vascular region; and respectively setting the transparency of the image and the transparency of the original image, and superposing the images for vessel labeling. Compared with the prior art, the labeling process is semi-automatic, does not need support of a large number of data sets, does not need a large amount of pre-training time, is simpler and more convenient to operate, and is less prone to errors.

Description

Blood vessel labeling method of ultrasonic image
Technical Field
The invention relates to the field of vessel image annotation, in particular to a vessel annotation method of an ultrasonic image.
Background
Under the background of the rapid development of artificial intelligence image recognition technology, vein recognition technology is in a unique importance position. From vein identity authentication to full-automatic venipuncture, the development of vein image recognition technology is promoted by a high-end technology which is more efficient and safer.
At present, the mainstream blood vessel segmentation and identification method in academia is generally realized based on a convolutional neural network. The goal of vessel separation from the image is typically accomplished using a u-net network or similar network structure. Compared with the traditional image segmentation algorithm, the image recognition algorithm based on the convolutional neural network has the characteristics of high speed, high accuracy, optimization, strong robustness and the like. But also has a very significant disadvantage: a large amount of data is required for training. This means that we need a large number of well-labeled (ground route) images to support the training process of the network model. Traditionally, such image labeling is performed by professional medical personnel using professional labeling software (e.g., imagelabel). However, training a network with a high degree of accuracy usually requires a large number of pictures, which is clearly a huge amount of engineering for the annotating staff.
Chinese patent CN201810607801.2 discloses a blood vessel segmentation method based on neural network. In this method, the data set employed is derived from the DRIVE public database. Although the method can effectively segment the blood vessel, the method needs the support of a large number of data sets, and can only detect the blood vessel picture which is the same as the segmented and trained set, and the identification precision of the pictures with different brightness and contrast is greatly reduced.
Chinese patent CN202010164856.8 discloses a method for labeling blood vessels. The method obtains the vessel boundary by selecting the seed point by a labeling person and utilizing a region growing method, thereby achieving the purpose of labeling the formed vessel. However, this method is prone to errors: if the seed point is selected outside the blood vessel, the growth process is not controllable, and the labeling of the whole picture causes problems.
Disclosure of Invention
The invention aims to overcome the defects of large data set support and easy error in the prior art, and provides a blood vessel labeling method of an ultrasonic image, which has high efficiency, does not need large data set support, and is time-saving and convenient.
The purpose of the invention can be realized by the following technical scheme:
a method for labeling a blood vessel of an ultrasonic image comprises the following steps:
acquiring an ultrasonic blood vessel image;
selecting a blood vessel area of the ultrasonic blood vessel image for cutting, and interpolating to the size of the image before cutting to obtain a first image;
performing histogram equalization processing on the first image to obtain a second image;
carrying out binarization processing on the second image according to a preset binarization segmentation threshold value to obtain a third image, so that the ultrasonic blood vessel image respectively represents a blood vessel region and a non-blood vessel region through two pixel values;
carrying out corrosion and expansion operation on the third image to obtain a fourth image;
calculating the area of each blood vessel region in the fourth image, and converting the blood vessel regions with the areas larger than a preset maximum blood vessel region threshold value and smaller than a preset minimum blood vessel region threshold value into non-blood vessel regions to obtain a fifth image;
calculating the ratio of the outline area of each blood vessel region in the fifth image to the convex hull area, wherein the convex hull area is the maximum convex polygon area formed by connecting all vertexes in the blood vessel region, and converting the blood vessel region of which the ratio of the outline area to the convex hull area is smaller than a preset outline-convex hull area ratio segmentation threshold value into a non-blood vessel region to obtain a sixth image;
calculating the aspect ratio of each blood vessel region in the sixth image, and converting the blood vessel regions with the aspect ratios larger than a preset maximum aspect ratio and smaller than a preset minimum aspect ratio into non-blood vessel regions to obtain a seventh image;
and respectively setting the transparency of the seventh image and the transparency of the first image, superposing and synthesizing an eighth image, and carrying out blood vessel labeling according to the eighth image.
Further, inverting a blood vessel region which does not conform to a first removal formula in a fourth image to obtain the fifth image, wherein the first removal formula is as follows:
Scontour_min≤Scontour≤Scontour_max
in the formula, ScontourIs the area of the vessel region, Scontour_minIs a minimum vessel region threshold, Scontour_maxThe maximum blood vessel region threshold.
Further, the minimum vessel region threshold is in the range of 40 to 60, and the maximum vessel region threshold is in the range of 450 to 550.
Further, inverting a blood vessel region which does not conform to a second removal formula in the fifth image to obtain the sixth image, wherein the second removal formula is as follows:
Figure BDA0002989918980000031
in the formula, ScontourIs the area of the contour, SconvexFor convex hull area, threshold 1 is the contour-convex hull area ratio partitioning threshold.
Further, the contour-convex hull area ratio segmentation threshold is within a range of 0.8 to 0.9.
Further, inverting a blood vessel region in the sixth image that does not conform to a third removal formula to obtain the seventh image, where the third removal formula is:
Figure BDA0002989918980000032
in the formula, contourrheightBeing the longitudinal height of the vessel region, contourrwidthIs the transverse width of the vessel region, thresholdminFor minimum aspect ratio, thresholdmaxIs the maximum aspect ratio.
Further, the minimum aspect ratio is in the range of 0.45 to 0.55 and the maximum aspect ratio is in the range of 1.15 to 1.25.
Further, the interpolation adopts a bicubic interpolation method.
Further, the calculation expression of the histogram equalization is as follows:
Figure BDA0002989918980000033
in the formula, SkIs a new value of the original image after histogram equalization of the value with the gray scale value k,
Figure BDA0002989918980000034
and expressing the proportion of the pixel point with the gray value i in the total number of the image pixel points.
Further, the eighth image is subjected to vessel labeling by an expert method.
Compared with the prior art, the invention has the following advantages:
(1) compared with the pure manual labeling, the labeling process is semi-automatic, after a blood vessel picture acquired under a near ultrasonic probe is input, the algorithm outputs all parts which are possibly blood vessels, and after the part determined as the blood vessel is selected by a labeling person, a binary image is output, wherein white is the segmented blood vessel, and black is a non-blood vessel part. The image is the marking data, can be directly used for training the neural network, and the efficiency can be improved by more than 500%.
(2) Compared with a convolutional neural network, the method does not need the support of a large number of data sets during labeling, does not need a large amount of pre-training time, and is more time-saving and light in weight.
(3) The labeling method has another advantage over manual labeling that the boundary is logically distinguished and is finally obtained through a series of processing modes such as morphological processing. In the process of convolutional neural network training, the characteristics are easier to learn by the network, so that the model accuracy is improved.
Drawings
Fig. 1 is a flowchart of a method for vessel labeling in an ultrasound image according to an embodiment of the present invention;
FIG. 2 is a result of histogram equalization of an image according to an embodiment of the present invention;
FIG. 3 is a result of thresholding an image according to an embodiment of the invention;
FIG. 4 is a diagram illustrating a seventh image after a series of morphological processing is performed on the image according to an embodiment of the invention.
FIG. 5 is a diagram illustrating a seventh image after a series of morphological processing is performed on the image according to an embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings or orientations or positional relationships that the products of the present invention are conventionally placed in use, and are only for convenience of describing the present invention and simplifying the description, but do not indicate or imply that the device or element to which the description refers must have a specific orientation, be constructed in a specific orientation, and be operated, and thus should not be construed as limiting the present invention.
It should be noted that the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.
Furthermore, the terms "horizontal", "vertical" and the like do not imply that the components are required to be absolutely horizontal or pendant, but rather may be slightly inclined. For example, "horizontal" merely means that the direction is more horizontal than "vertical" and does not mean that the structure must be perfectly horizontal, but may be slightly inclined.
Example 1
The embodiment provides a method for vessel annotation of an ultrasonic image, which comprises the following steps:
firstly, image acquisition and preprocessing
S101: acquiring a vein longitudinal section image of the forearm as an ultrasonic blood vessel image through an ultrasonic probe;
s102: and (4) cutting and zooming the ultrasonic blood vessel image. The main part of the ultrasound blood vessel image is cut into 70 × 70 pixels, and the image is interpolated into 120 × 120 pixels by a bicubic interpolation method (this step is to make the pixel values of the image consistent with the input of the neural network, and meanwhile, the erosion operation (erode) and the dilation operation (dilate) are beneficial to be used later, because the erosion and dilation algorithms have poor performance on the image with too small pixels), so as to obtain a first image, and the reference image is shown in fig. 2.
Bicubic interpolation:
Figure BDA0002989918980000051
s103: performing histogram equalization processing on the first image to obtain a second image; because the gray value of the ultrasound image is generally small, the gray value can be uniformly distributed by adopting the histogram equalization operation, so that the color value difference between the blood vessel and the background is increased, the threshold can be easily selected in the process of threshold segmentation, and a reference graph is shown in fig. 3.
Histogram equalization:
Figure BDA0002989918980000052
wherein SkRepresenting the new value of k in the original image after histogram equalization,
Figure BDA0002989918980000053
and expressing the proportion of the pixel point with the gray value i in the total number of the image pixel points.
Second, threshold segmentation and adaptive screening
S201: carrying out binarization processing on the second image according to a preset binarization segmentation threshold value to obtain a third image, so that the ultrasonic blood vessel image respectively represents a blood vessel region and a non-blood vessel region through two pixel values;
for the image data of the same source during the acquisition, the threshold value is selected only by determining before the first image is processed, and the segmentation threshold value is not required to be changed when the subsequent images are processed, as shown in fig. 4.
Threshold segmentation:
Figure BDA0002989918980000054
the value of the segmentation threshold a is within the range of 160 to 200, preferably 180.
S202: carrying out corrosion and expansion operation on the third image to obtain a fourth image; this step is to eliminate a small white area in the image, and to effectively eliminate interference factors such as noise. While smoothing the vessel edges.
In this embodiment, the erosion and dilation operations are performed on the image using filters of 5 × 5 and 3 × 3, respectively. An excessively large filter may cause the vessel edges to deform; a filter that is too small may not effectively remove noise.
S203: and removing the too large or too small connected region in the image. The vessel diameter area in the human body is generally within a defined range. Too large or too small a connected region is often a non-vascular dark region (e.g., muscle). Therefore, this area needs to be deleted in image processing.
The specific operation is as follows: and acquiring the outlines of all white blocks on the fourth image, calculating the inner area of the outlines, and setting a maximum blood vessel area threshold value and a minimum blood vessel area threshold value. Removing the contour region (i.e. reversing the inner region of the contour) which is greater than the maximum blood vessel region threshold and less than the minimum blood vessel region threshold, i.e. reversing the blood vessel region which does not conform to the first removal formula in the fourth image, and obtaining a fifth image, wherein the first removal formula is as follows:
Scontour_min≤Scontour≤Scontour_max
in the formula, ScontourIs the area of the vessel region, Scontour_minIs a minimum vessel region threshold, Scontour_maxThe maximum blood vessel region threshold.
The minimum blood vessel region threshold is within the range of 40 to 60, the maximum blood vessel region threshold is within the range of 450 to 550, the minimum blood vessel region threshold is preferably 50, and the maximum blood vessel region threshold is preferably 500.
S204: the region in the fifth image where the ratio of the outline area to the convex hull area (the maximum convex polygon area in the image connected by the vertices) is too small is removed. The radial cross section of the vessel is usually circular or elliptical, approximating the morphology with a ratio of vessel contour area to convex hull area close to 1. When the ratio is less than the threshold, therefore, the contour may not be an edge of a blood vessel or an edge of a blood vessel is unclear, the present embodiment removes such a contour region,
specifically, a blood vessel region which does not conform to a second removal formula in the fifth image is inverted to obtain a sixth image, wherein the second removal formula is as follows:
Figure BDA0002989918980000061
in the formula, ScontourIs the area of the contour, SconvexFor convex hull area, threshold 1 is the contour-convex hull area ratio partitioning threshold.
The contour-convex hull area ratio segmentation threshold is in the range of 0.8 to 0.9, preferably 0.85.
S205: contour regions with too large or too small aspect ratios in the image are removed. The aspect ratio of the vessel is typically around 0.7 due to slight compression when acquiring the ultrasound image. Thus, regions with aspect ratios above a maximum threshold or below a minimum threshold are removed,
specifically, a blood vessel region in the sixth image that does not conform to the third removal formula is inverted to obtain a seventh image, and the third removal formula is:
Figure BDA0002989918980000062
in the formula, contourrheightBeing the longitudinal height of the vessel region, contourrwidthIs the transverse width of the vessel region, thresholdminFor minimum aspect ratio, thresholdmaxIs the maximum aspect ratio.
The minimum aspect ratio is in the range of 0.45 to 0.55, preferably 0.5, and the maximum aspect ratio is in the range of 1.15 to 1.25, preferably 1.2.
A reference image of the final seventh image is shown in fig. 5.
Thirdly, manual screening
S301: respectively setting the transparency of the seventh image and the transparency of the first image, and synthesizing an eighth image in an overlapping manner, wherein in the embodiment, the pictures are synthesized with the respective transparencies of 50%, so that the effect of circling the outline on the original image is achieved;
s302: respectively displaying the eighth image and the original ultrasonic blood vessel image on a GUI interface, so that the marking personnel can conveniently and manually screen the rest contours;
s303: manually selecting contours which are not part of the blood vessel in the image displayed on the GUI interface by a label operator according to experience, and clicking to invert the contour regions;
s304: after the annotation personnel finishes selecting, selecting an output image, namely outputting the overlay image and the annotation image;
s305: and outputting the image after the labeling, and outputting the labeled image which can be directly used for neural network training.
The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations could be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions available to those skilled in the art through logic analysis, reasoning and limited experiments based on the prior art according to the concept of the present invention should be within the scope of protection defined by the claims.

Claims (10)

1. A method for labeling a blood vessel of an ultrasonic image is characterized by comprising the following steps:
acquiring an ultrasonic blood vessel image;
selecting a blood vessel area of the ultrasonic blood vessel image for cutting, and interpolating to the size of the image before cutting to obtain a first image;
performing histogram equalization processing on the first image to obtain a second image;
carrying out binarization processing on the second image according to a preset binarization segmentation threshold value to obtain a third image, so that the ultrasonic blood vessel image respectively represents a blood vessel region and a non-blood vessel region through two pixel values;
carrying out corrosion and expansion operation on the third image to obtain a fourth image;
calculating the area of each blood vessel region in the fourth image, and converting the blood vessel regions with the areas larger than a preset maximum blood vessel region threshold value and smaller than a preset minimum blood vessel region threshold value into non-blood vessel regions to obtain a fifth image;
calculating the ratio of the outline area of each blood vessel region in the fifth image to the convex hull area, wherein the convex hull area is the maximum convex polygon area formed by connecting all vertexes in the blood vessel region, and converting the blood vessel region of which the ratio of the outline area to the convex hull area is smaller than a preset outline-convex hull area ratio segmentation threshold value into a non-blood vessel region to obtain a sixth image;
calculating the aspect ratio of each blood vessel region in the sixth image, and converting the blood vessel regions with the aspect ratios larger than a preset maximum aspect ratio and smaller than a preset minimum aspect ratio into non-blood vessel regions to obtain a seventh image;
and respectively setting the transparency of the seventh image and the transparency of the first image, superposing and synthesizing an eighth image, and carrying out blood vessel labeling according to the eighth image.
2. The method for vessel labeling of an ultrasound image according to claim 1, wherein the fifth image is obtained by inverting a vessel region in the fourth image that does not conform to a first removal formula:
Scontour_min≤Scontour≤Scontour_max
in the formula, ScontourIs the area of the vessel region, Scontour_minIs a minimum vessel region threshold, Scontour_maxThe maximum blood vessel region threshold.
3. The method for vessel labeling in an ultrasound image of claim 2, wherein the minimum vessel region threshold is in the range of 40 to 60, and the maximum vessel region threshold is in the range of 450 to 550.
4. The method for vessel labeling of an ultrasound image according to claim 1, wherein the sixth image is obtained by inverting a vessel region in the fifth image that does not conform to a second removal formula:
Figure FDA0002989918970000021
in the formula, ScontourIs the area of the contour, SconvexFor convex hull area, threshold 1 is the contour-convex hull area ratio partitioning threshold.
5. The method for vessel labeling of an ultrasound image according to claim 4, wherein the contour-convex hull area ratio segmentation threshold is in a range of 0.8 to 0.9.
6. The method for vessel labeling of an ultrasound image according to claim 1, wherein the seventh image is obtained by inverting a vessel region in the sixth image that does not conform to a third removal formula:
Figure FDA0002989918970000022
in the formula, contourrheightBeing the longitudinal height of the vessel region, contourrwidthIs the transverse width of the vessel region, thresholdminFor minimum aspect ratio, thresholdmaxIs the maximum aspect ratio.
7. The method for vessel labeling of an ultrasound image according to claim 6, wherein the minimum aspect ratio is in the range of 0.45 to 0.55 and the maximum aspect ratio is in the range of 1.15 to 1.25.
8. The method for vessel labeling of an ultrasound image as claimed in claim 1, wherein said interpolation is bicubic interpolation.
9. The method for vessel labeling of an ultrasound image as claimed in claim 1, wherein the calculation expression of the histogram equalization is:
Figure FDA0002989918970000023
in the formula, SkIs a new value of the original image after histogram equalization of the value with the gray scale value k,
Figure FDA0002989918970000024
and expressing the proportion of the pixel point with the gray value i in the total number of the image pixel points.
10. The method for vessel labeling in an ultrasound image according to claim 1, wherein the vessel labeling is performed in the eighth image by expert method.
CN202110312435.XA 2021-03-24 2021-03-24 Blood vessel labeling method of ultrasonic image Pending CN113344842A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110312435.XA CN113344842A (en) 2021-03-24 2021-03-24 Blood vessel labeling method of ultrasonic image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110312435.XA CN113344842A (en) 2021-03-24 2021-03-24 Blood vessel labeling method of ultrasonic image

Publications (1)

Publication Number Publication Date
CN113344842A true CN113344842A (en) 2021-09-03

Family

ID=77467830

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110312435.XA Pending CN113344842A (en) 2021-03-24 2021-03-24 Blood vessel labeling method of ultrasonic image

Country Status (1)

Country Link
CN (1) CN113344842A (en)

Citations (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060241463A1 (en) * 2004-12-08 2006-10-26 Yio-Wha Shau Quantitative non-invasive method for detecting degree of malignancy in tumors and application thereof
CN105844256A (en) * 2016-04-07 2016-08-10 广州盈可视电子科技有限公司 Panorama video frame image processing method and device
CN106355599A (en) * 2016-08-30 2017-01-25 上海交通大学 Non-fluorescent eye fundus image based automatic segmentation method for retinal blood vessels
CN106407917A (en) * 2016-09-05 2017-02-15 山东大学 Dynamic scale distribution-based retinal vessel extraction method and system
CN106815853A (en) * 2016-12-14 2017-06-09 海纳医信（北京）软件科技有限责任公司 To the dividing method and device of retinal vessel in eye fundus image
CN108615051A (en) * 2018-04-13 2018-10-02 博众精工科技股份有限公司 Diabetic retina image classification method based on deep learning and system
CN109166124A (en) * 2018-11-20 2019-01-08 中南大学 A kind of retinal vascular morphologies quantization method based on connected region
CN109886938A (en) * 2019-01-29 2019-06-14 深圳市科曼医疗设备有限公司 A kind of ultrasound image blood vessel diameter method for automatic measurement
CN110136092A (en) * 2019-05-21 2019-08-16 北京三快在线科技有限公司 Image processing method, device and storage medium
CN110717888A (en) * 2019-09-06 2020-01-21 天津大学 Automatic identification method for intravascular Optical Coherence Tomography (OCT) vessel wall inner contour
CN110929728A (en) * 2020-02-18 2020-03-27 南京景三医疗科技有限公司 Image region-of-interest dividing method, image segmentation method and device
CN111415335A (en) * 2020-03-11 2020-07-14 北京深睿博联科技有限责任公司 Blood vessel labeling method and device
CN111681276A (en) * 2020-06-11 2020-09-18 北京大恒普信医疗技术有限公司 Method and device for determining ratio of arteriovenous diameter in fundus image and electronic equipment
CN112001928A (en) * 2020-07-16 2020-11-27 北京化工大学 Retinal vessel segmentation method and system
CN112215800A (en) * 2020-09-14 2021-01-12 北京航空航天大学 Machine learning-based overlapped chromosome recognition and segmentation method
CN112308866A (en) * 2020-11-04 2021-02-02 Oppo广东移动通信有限公司 Image processing method, image processing device, electronic equipment and storage medium

Patent Citations (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060241463A1 (en) * 2004-12-08 2006-10-26 Yio-Wha Shau Quantitative non-invasive method for detecting degree of malignancy in tumors and application thereof
CN105844256A (en) * 2016-04-07 2016-08-10 广州盈可视电子科技有限公司 Panorama video frame image processing method and device
CN106355599A (en) * 2016-08-30 2017-01-25 上海交通大学 Non-fluorescent eye fundus image based automatic segmentation method for retinal blood vessels
CN106407917A (en) * 2016-09-05 2017-02-15 山东大学 Dynamic scale distribution-based retinal vessel extraction method and system
CN106815853A (en) * 2016-12-14 2017-06-09 海纳医信（北京）软件科技有限责任公司 To the dividing method and device of retinal vessel in eye fundus image
CN108615051A (en) * 2018-04-13 2018-10-02 博众精工科技股份有限公司 Diabetic retina image classification method based on deep learning and system
CN109166124A (en) * 2018-11-20 2019-01-08 中南大学 A kind of retinal vascular morphologies quantization method based on connected region
CN109886938A (en) * 2019-01-29 2019-06-14 深圳市科曼医疗设备有限公司 A kind of ultrasound image blood vessel diameter method for automatic measurement
CN110136092A (en) * 2019-05-21 2019-08-16 北京三快在线科技有限公司 Image processing method, device and storage medium
CN110717888A (en) * 2019-09-06 2020-01-21 天津大学 Automatic identification method for intravascular Optical Coherence Tomography (OCT) vessel wall inner contour
CN110929728A (en) * 2020-02-18 2020-03-27 南京景三医疗科技有限公司 Image region-of-interest dividing method, image segmentation method and device
CN111415335A (en) * 2020-03-11 2020-07-14 北京深睿博联科技有限责任公司 Blood vessel labeling method and device
CN111681276A (en) * 2020-06-11 2020-09-18 北京大恒普信医疗技术有限公司 Method and device for determining ratio of arteriovenous diameter in fundus image and electronic equipment
CN112001928A (en) * 2020-07-16 2020-11-27 北京化工大学 Retinal vessel segmentation method and system
CN112215800A (en) * 2020-09-14 2021-01-12 北京航空航天大学 Machine learning-based overlapped chromosome recognition and segmentation method
CN112308866A (en) * 2020-11-04 2021-02-02 Oppo广东移动通信有限公司 Image processing method, image processing device, electronic equipment and storage medium

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
周茂: "基于深度学习的医学CT图像分割技术研究", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 *
周茂: "基于深度学习的医学CT图像分割技术研究", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》, no. 1, 15 January 2021 (2021-01-15) *
郭笑妍 等: "形状约束下活动轮廓模型冠脉血管图像多尺度分割", 《中国图象图形学报》 *
郭笑妍 等: "形状约束下活动轮廓模型冠脉血管图像多尺度分割", 《中国图象图形学报》, vol. 21, no. 7, 31 July 2016 (2016-07-31), pages 923 - 932 *

Similar Documents

Publication Publication Date Title
CN110706246B (en) Blood vessel image segmentation method and device, electronic equipment and storage medium
CN108198184B (en) Method and system for vessel segmentation in contrast images
CN110287956B (en) Automatic matching method and device for blood vessel central lines
CN108154519A (en) Dividing method, device and the storage medium of eye fundus image medium vessels
US9881381B2 (en) Image processing apparatus, image processing method, and computer-readable recording medium
CN112258488A (en) Medical image focus segmentation method
US11783488B2 (en) Method and device of extracting label in medical image
CN110706233A (en) Retina fundus image segmentation method and device
CN111882566B (en) Blood vessel segmentation method, device, equipment and storage medium for retina image
CN111062331B (en) Image mosaic detection method and device, electronic equipment and storage medium
CN113643353B (en) Measurement method for enhancing resolution of vascular caliber of fundus image
CN111815663A (en) Hepatic vessel segmentation system based on Hessian matrix and gray scale method
CN116152500A (en) Full-automatic tooth CBCT image segmentation method based on deep learning
CN109872337B (en) Eye fundus image optic disc segmentation method based on rapid mean shift
CN111340773A (en) Retinal image blood vessel segmentation method
CN113139929A (en) Gastrointestinal tract endoscope image preprocessing method comprising information screening and fusion repairing
CN113344842A (en) Blood vessel labeling method of ultrasonic image
CN115631194B (en) Method, device, equipment and medium for identifying and detecting intracranial aneurysm
CN114693622B (en) Plaque erosion automatic detection system based on artificial intelligence
CN111062909A (en) Method and equipment for judging benign and malignant breast tumor
CN113889238B (en) Image identification method and device, electronic equipment and storage medium
CN108629780A (en) Tongue image dividing method based on color decomposition and threshold technology
CN111860173B (en) Remote sensing image ground feature element extraction method and system based on weak supervision
CN113096117A (en) Ectopic ossification CT image segmentation method, three-dimensional reconstruction method and device
Ardizzone et al. Blood vessels and feature points detection on retinal images

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20210903