Blood vessel segmentation method and related product

Abstract

The invention relates to a blood vessel segmentation method and a related product. The blood vessel segmentation method comprises the steps that a brightness ultrasound image in a brightness mode and a contrast ultrasound image in an ultrasound contrast mode are obtained, the brightness ultrasound image and the contrast ultrasound image are ultrasound images containing target tissue of an object, and the position of the target tissue in the brightness ultrasound image is the same as the position of the target tissue in the contrast ultrasound image; segmenting the contour of the target tissue in the brightness ultrasound image to determine the position of the contour in the brightness ultrasound image; determining the position of a target tissue in the contrast ultrasound image according to the position of the contour in the brightness ultrasound image; and segmenting the blood vessel in the target tissue in the contrast ultrasound image to determine the position of the blood vessel in the contrast ultrasound image. By adopting the blood vessel segmentation method, the position of the blood vessel in the ultrasonic image in the ultrasonic contrast mode can be more accurately determined.

Description

Technical Field

[0001] The present application relates to the field of medical technology, and in particular to a blood vessel segmentation method and related products. Background Art

[0002] Ultrasound contrast imaging is a technology that uses intravenous contrast agents containing microbubbles to help doctors more clearly observe the flow of blood in organs and blood vessels when using ultrasound equipment. It can be used to detect various diseases. In ultrasound contrast-guided surgery, the doctor injects the ultrasound contrast agent into the subject's vein and uses the ultrasound equipment with the ultrasound contrast mode turned on to scan the target tissue of the subject to obtain a real-time ultrasound image in ultrasound contrast mode. The target tissue is such as the kidney. In the real-time ultrasound image, as the time of injection of the ultrasound contrast agent changes, the blood vessels in the target tissue will first be imaged and then disappear. In this process, the doctor needs to use the special characteristics of ultrasound contrast imaging to observe the distribution of blood vessels in the target tissue of the subject, that is, to determine the position of the blood vessels in the ultrasound image in ultrasound contrast mode.

[0003] Since ultrasound imaging is short, it is difficult for doctors to capture the best imaging moment of blood vessels and understand the distribution of blood vessels. Based on this, doctors need to rely on their clinical experience to judge the location of blood vessels, making it difficult to accurately determine the location of blood vessels in ultrasound images under ultrasound imaging mode. Summary of the invention

[0004] Based on this, it is necessary to provide a blood vessel segmentation method and related products to address the above technical issues, so as to more accurately determine the position of blood vessels in ultrasound images in ultrasound angiography mode. The related products include a blood vessel segmentation device, a computer device, a computer-readable storage medium, and a computer program product.

[0005] In a first aspect, the present application provides a blood vessel segmentation method, the method comprising: Acquire a brightness ultrasound image in a brightness mode and a contrast ultrasound image in an ultrasound contrast mode, wherein both the brightness ultrasound image and the contrast ultrasound image are ultrasound images containing a target tissue of an object, and a position of the target tissue in the brightness ultrasound image is the same as a position of the target tissue in the contrast ultrasound image; Segmenting the contour of the target tissue in the brightness ultrasound image to determine the position of the contour in the brightness ultrasound image; Determining the position of the target tissue in the contrast ultrasound image according to the position of the contour in the brightness ultrasound image; The blood vessels in the target tissue in the contrast-enhanced ultrasound image are segmented to determine the positions of the blood vessels in the contrast-enhanced ultrasound image.

[0006] The blood vessel segmentation method provided in the first aspect is adopted to obtain a brightness ultrasound image in brightness mode (B mode) and a contrast ultrasound image in ultrasound contrast mode. Both ultrasound images contain the target tissue of the object, and the position of the target tissue in the two ultrasound images is the same. Because the contour of the target tissue in the brightness ultrasound image in the B mode is clearer and easier to capture than the contour of the target tissue in the contrast ultrasound image in ultrasound contrast mode. Based on this, the contour of the target tissue in the brightness ultrasound image is first segmented to accurately determine the position of the contour in the brightness ultrasound image. Then, based on the fact that the position of the target tissue in the two ultrasound images is the same, the position of the target tissue in the contrast ultrasound image is accurately determined according to the position of the contour of the target tissue in the brightness ultrasound image. Furthermore, when the position of the target tissue is determined, the blood vessels in the target tissue in the contrast ultrasound image are segmented, so that the position of the blood vessels in the contrast ultrasound image can be more accurately determined.

[0007] In one of the embodiments, the number of the brightness ultrasound images and the number of the contrast ultrasound images are both at least two; the position of the target tissue in the brightness ultrasound image is the same as the position of the target tissue in the contrast ultrasound image, which means that the position of the target tissue is the same in the brightness ultrasound image and the contrast ultrasound image with the same acquisition timestamp; there is a one-to-one correspondence between the at least two brightness ultrasound images, the at least two contrast ultrasound images and the at least two acquisition timestamps; After segmenting the blood vessels in the target tissue in the contrast-enhanced ultrasound image to determine the positions of the blood vessels in the contrast-enhanced ultrasound image, the method further includes: respectively determining image brightness values ​​of at least two of the contrast ultrasound images, wherein the image brightness values ​​are used to characterize the brightness of the contrast ultrasound images, and at least two acquisition time stamps of at least two of the contrast ultrasound images are continuous; Determining a critical time period for ultrasound contrast imaging based on the image brightness value and the acquisition timestamp of each of the contrast ultrasound images, wherein the image brightness value of the contrast ultrasound image acquired in the critical time period is greater than a brightness threshold; The positions of at least two of the blood vessels in at least two target ultrasound images are merged to obtain the position of the target blood vessel; the target ultrasound image refers to a contrast ultrasound image whose acquisition timestamp is within the key time period.

[0008] In one embodiment, the step of respectively determining the image brightness values ​​of at least two of the contrast ultrasound images comprises: Converting at least two of the contrast ultrasound images into grayscale images respectively; Calculating the average brightness value of the pixel points of the target tissue in each of the grayscale images; The average brightness value is used as the image brightness value of the corresponding contrast ultrasound image.

[0009] In one embodiment, segmenting the contour of the target tissue in the brightness ultrasound image to determine the position of the contour in the brightness ultrasound image includes: Inputting the brightness ultrasound image into a contour segmentation model to obtain the position of the contour in the brightness ultrasound image; The contour segmentation model is obtained by supervised training with first training data and first supervised data, wherein the first training data is a first ultrasound image in brightness mode, and the first supervised data is a reference contour of a reference tissue marked in the first ultrasound image.

[0010] In one embodiment, segmenting the blood vessels in the target tissue in the contrast-enhanced ultrasound image to determine the position of the blood vessels in the contrast-enhanced ultrasound image includes: Setting the target tissue in the contrast ultrasound image as a region of interest; Inputting the contrast ultrasound image provided with the region of interest into a blood vessel segmentation model to obtain the position of the blood vessel in the contrast ultrasound image; The blood vessel segmentation model is obtained by supervised training with second training data and second supervisory data, the second training data is a second ultrasound image in which a reference tissue is marked as a reference region of interest, the second supervisory data is a reference blood vessel marked in the second ultrasound image, and the second ultrasound image is an ultrasound image in ultrasound contrast mode.

[0011] In one embodiment, the step of acquiring a brightness ultrasound image in a brightness mode and a contrast ultrasound image in an ultrasound contrast mode includes: Acquire an initial brightness image and an initial contrast image with the same acquisition timestamp, wherein the initial brightness image is an ultrasound image in a brightness mode, and the initial contrast image is an ultrasound image in an ultrasound contrast mode, and both the initial brightness image and the initial contrast image are ultrasound images obtained by scanning the target tissue of the object; Preprocessing the initial brightness image to obtain the brightness ultrasound image; Preprocessing the initial contrast image to obtain the contrast ultrasound image; The brightness ultrasound image and the contrast ultrasound image obtained after preprocessing contain the same information and have the same size.

[0012] In a second aspect, the present application further provides a blood vessel segmentation device, the device comprising: an image acquisition module, used to acquire a brightness ultrasound image in a brightness mode and a contrast ultrasound image in an ultrasound contrast mode, wherein both the brightness ultrasound image and the contrast ultrasound image are ultrasound images containing a target tissue of an object, and a position of the target tissue in the brightness ultrasound image is the same as a position of the target tissue in the contrast ultrasound image; A contour segmentation module, used to segment the contour of the target tissue in the brightness ultrasound image to determine the position of the contour in the brightness ultrasound image; a position determination module, configured to determine the position of the target tissue in the contrast ultrasound image according to the position of the contour in the brightness ultrasound image; The blood vessel segmentation module is used to segment the blood vessels in the target tissue in the contrast ultrasound image to determine the position of the blood vessels in the contrast ultrasound image.

[0013] In a third aspect, the present application further provides a computer device, comprising: a memory and a processor, wherein the memory stores program instructions; when the program instructions are executed by the processor, the processor executes the method shown in the first aspect or any embodiment of the first aspect.

[0014] In a fourth aspect, the present application further provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium; when the computer program runs on one or more processors, the method shown in the first aspect or any embodiment of the first aspect is executed.

[0015] In a fifth aspect, the present application also provides a computer program product, which includes a computer program or instructions; when the computer program or instructions are run on a computer, the computer is enabled to execute the method shown in the first aspect or any embodiment of the first aspect.

[0016] It can be understood that the blood vessel segmentation device provided in the second aspect, the computer device provided in the third aspect, the computer readable storage medium provided in the fourth aspect, and the computer program product provided in the fifth aspect can be used to execute the blood vessel segmentation method shown in the first aspect of the present application or any embodiment of the first aspect. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the blood vessel segmentation method, which will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments are introduced as follows.

[0018] Figure 1One of the flowcharts of the blood vessel segmentation method provided in the embodiment of the present application; Figure 2 A brightness change curve diagram provided in an embodiment of the present application; Figure 3 The second flowchart of the blood vessel segmentation method provided in the embodiment of the present application; Figure 4 A schematic diagram of the structure of a blood vessel segmentation device provided in an embodiment of the present application; Figure 5 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to facilitate understanding of the embodiments of the present application, the embodiments of the present application will be described more fully below with reference to the relevant drawings. The preferred embodiments of the embodiments of the present application are given in the drawings. However, the embodiments of the present application can be implemented in many different forms and are not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the embodiments of the present application more thorough and comprehensive.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the technical field of the embodiments of the present application. The terms used herein in the description of the embodiments of the present application are only for the purpose of describing specific embodiments and are not intended to limit the embodiments of the present application.

[0021] The terms "first", "second", etc. used in this article are used to distinguish different objects rather than to describe a specific order. For the purpose of distinguishing different objects, "first training data" can also be described as "second training data", and "second training data" can be described as "first training data". When used, the singular expressions "one", "a", "said", "above", "the" and "this" are intended to also include plural expressions, unless the context clearly indicates otherwise. It should also be understood that the term "includes / comprising" specifies the presence of stated features, wholes, steps, operations, parts or their fusions, but does not exclude the possibility of the presence or addition of one or more other features, wholes, steps, operations, parts or their fusions.

[0022] An embodiment of the present application provides a blood vessel segmentation method. Figure 1 This is one of the flow charts of the blood vessel segmentation method provided in the embodiment of the present application. Figure 1 As shown, the blood vessel segmentation method includes the following steps S101 to S104.

[0023] S101, acquiring a brightness ultrasound image in a brightness mode and a contrast ultrasound image in an ultrasound contrast mode. Both the brightness ultrasound image and the contrast ultrasound image are ultrasound images containing a target tissue of an object, and the position of the target tissue in the brightness ultrasound image is the same as the position of the target tissue in the contrast ultrasound image.

[0024] Among them, the target tissue is an organ or tissue in the subject's body, such as a kidney. The position of the target tissue in the two ultrasound images described in the embodiment of the present application is the same, which means that the pixel points corresponding to the target tissue in the two ultrasound images are the same. If the position of the target tissue in the brightness ultrasound image is from the second row of pixels to the fifth row of pixels, then the position of the target tissue in the contrast ultrasound image is also from the second row of pixels to the fifth row of pixels. The same position includes the same orientation, size, and dimensions. It should be understood that the ultrasound image described in the embodiment of the present application contains the target tissue, which means that there are pixels in the ultrasound image for describing the target tissue, and does not mean that the ultrasound image contains the target tissue as an object, and other similar descriptions are the same.

[0025] In one embodiment, the above step S101, acquiring a brightness ultrasound image in a brightness mode and a contrast ultrasound image in an ultrasound contrast mode, includes the following steps S201 to S203.

[0026] S201, acquiring an initial brightness image and an initial contrast image with the same acquisition timestamp. The initial brightness image is an ultrasound image in a brightness mode, and the initial contrast image is an ultrasound image in an ultrasound contrast mode. Both the initial brightness image and the initial contrast image are ultrasound images obtained by scanning a target tissue of the object.

[0027] Specifically, the ultrasound probe collects ultrasound images of the target tissue of the object in the B mode and the ultrasound contrast mode at the same time, and obtains the initial brightness image in the B mode and the initial contrast image in the ultrasound contrast mode accordingly. The vascular segmentation device that executes the vascular segmentation method in the embodiment of the present application obtains the initial brightness image and the initial contrast image with the same acquisition timestamp. Acquiring two ultrasound images with the same acquisition timestamp ensures that the two ultrasound images are collected by the object at the same breathing amplitude, so as to ensure that the position of the target tissue in the two ultrasound images is the same under the influence of breathing. It should be noted that the initial brightness image and the initial contrast image may contain other information in addition to the target tissue of the object. Other information includes text information describing image attributes and other tissues in the object except the target tissue.

[0028] S202, preprocessing the initial brightness image to obtain a brightness ultrasound image.

[0029] S203, preprocessing the initial contrast image to obtain a contrast ultrasound image.

[0030] The brightness ultrasound image and the contrast ultrasound image obtained after preprocessing contain the same information and have the same size.

[0031] Furthermore, the blood vessel segmentation device separates the acquired initial brightness image and initial contrast image, and preprocesses the initial brightness image and the initial contrast image respectively. Wherein, the preprocessing includes image cropping. Specifically, the initial brightness image is cropped to remove the above-mentioned other information contained in the initial brightness image, and a brightness ultrasound image of the target tissue of the object is obtained. The initial contrast image is cropped to remove the above-mentioned other information contained in the initial contrast image, and a contrast ultrasound image of the target tissue of the object is obtained. Wherein, the information and size contained in the brightness ultrasound image and the contrast ultrasound image obtained after cropping are the same. The same size means that the length, width and other dimensions of the two are the same. The information contained is the same, which means that the information described by the pixel points of the two is the same. It should be understood that the information and size contained in the brightness ultrasound image and the contrast ultrasound image with the same acquisition timestamp are also the same, so that the position of the target tissue in the brightness ultrasound image and the position of the target tissue in the contrast ultrasound image can be guaranteed to be the same. In other words, the difference between the brightness ultrasound image and the contrast ultrasound image is that the ultrasound imaging features in the B mode and the ultrasound contrast mode are inconsistent. Therefore, by preprocessing the initial brightness image and the initial contrast image with the same acquisition timestamp, a brightness ultrasound image and a contrast ultrasound image are obtained, both of which contain the target tissue of the object and have the same position of the target tissue therein.

[0032] In this embodiment, the initial brightness image and the initial contrast image acquired by the ultrasonic probe are preprocessed to remove information other than the target tissue in the initial brightness image and the initial contrast image, so as to obtain a brightness ultrasound image and a contrast ultrasound image of the target tissue of the object. Based on the purpose of determining the position of the blood vessels in the target tissue in the ultrasound image under the ultrasound contrast mode of the present blood vessel segmentation method, the above-mentioned preprocessing of the ultrasound image can not only ensure that the target tissue is included in the brightness ultrasound image and the contrast ultrasound image to achieve this purpose. Moreover, the resolution of the brightness ultrasound image and the contrast ultrasound image to be processed can be reduced to improve the processing efficiency of the ultrasound image. Furthermore, the interference of information other than the target tissue in the process of achieving the purpose can be reduced, which is conducive to more accurately determining the position of the blood vessels in the target tissue in the ultrasound image under the ultrasound contrast mode.

[0033] S102, segmenting the contour of the target tissue in the brightness ultrasound image to determine the position of the contour in the brightness ultrasound image.

[0034] In order to determine the position of the blood vessels in the target tissue in the ultrasound image in the ultrasound contrast mode, it is necessary to first determine the position of the target tissue in the ultrasound image in the ultrasound contrast mode. Because the contour of the target tissue in the brightness ultrasound image in the B mode is clearer and easier to capture than the contour of the target tissue in the contrast ultrasound image in the ultrasound contrast mode. Based on this, the contour of the target tissue in the brightness ultrasound image can be segmented first to accurately determine the position of the contour in the brightness ultrasound image. Furthermore, based on the fact that the position of the target tissue in the brightness ultrasound image and the contrast ultrasound image is the same, the position of the target tissue in the contrast ultrasound image can be accurately determined according to the position of the contour of the target tissue in the brightness ultrasound image.

[0035] It can be understood that when the initial brightness image is preprocessed to obtain a brightness ultrasound image, by reducing the resolution of the brightness ultrasound image, the segmentation efficiency can be improved when the contour of the target tissue in the brightness ultrasound image is segmented. By reducing the interference of information other than the target tissue, the segmentation accuracy can be improved when the contour of the target tissue in the brightness ultrasound image is segmented, which is conducive to more accurately determining the position of the contour of the target tissue in the brightness ultrasound image, and further conducive to more accurately determining the position of the target tissue and the blood vessels in the target tissue in the contrast ultrasound image.

[0036] In one embodiment, the above step S102, segmenting the contour of the target tissue in the brightness ultrasound image to determine the position of the contour in the brightness ultrasound image, includes the following steps: The brightness ultrasound image is input into the contour segmentation model to obtain the position of the contour in the brightness ultrasound image.

[0037] The contour segmentation model is obtained by supervised training with first training data and first supervised data, the first training data is a first ultrasound image in brightness mode, and the first supervised data is a reference contour of a reference tissue marked in the first ultrasound image.

[0038] The process of segmenting the contour of the target tissue in the brightness ultrasound image can be completed by using the contour segmentation model. The training process of the contour segmentation model is as follows.

[0039] First, accurate data. The data to be prepared include at least the first training data and the first supervision data. The first training data is the first ultrasound image in the B mode, and the first ultrasound image contains the reference tissue. The reference tissue should be consistent with the target tissue. For example, when the target tissue is the kidney, the reference tissue is also the kidney. Exemplarily, the first ultrasound image is an ultrasound image obtained after preprocessing and removing information other than the reference tissue. The first ultrasound image obtained after preprocessing is used as the first training data, and by reducing the resolution of the first ultrasound image, it is beneficial to make the model converge faster during the training process. It can be understood that in the case where the first ultrasound image is an ultrasound image obtained after preprocessing, the brightness ultrasound image should also be an ultrasound image obtained after preprocessing. The first supervision data is the reference contour of the reference tissue marked by a professional in the first ultrasound image. Then. The model is trained with reference supervision using the first training data and the first supervision data, and the training is stopped after the model converges to obtain a contour segmentation model. In addition, a corresponding test data set can be prepared to test the trained contour segmentation model to determine the prediction accuracy of the contour segmentation model.

[0040] In this embodiment, the contour of the target tissue in the brightness ultrasound image is segmented using the trained contour segmentation model, so that the position of the contour of the target tissue in the brightness ultrasound image can be accurately determined.

[0041] S103, determining the position of the target tissue in the contrast ultrasound image according to the position of the contour in the brightness ultrasound image.

[0042] As described above, based on the fact that the position of the target tissue in the brightness ultrasound image and the contrast ultrasound image is the same, the position of the target tissue in the contrast ultrasound image can be accurately determined according to the position of the contour of the target tissue in the brightness ultrasound image.

[0043] S104, segmenting the blood vessels in the target tissue in the contrast-enhanced ultrasound image to determine the positions of the blood vessels in the contrast-enhanced ultrasound image.

[0044] When the position of the target tissue in the contrast ultrasound image is accurately determined, segmenting the blood vessels in the target tissue in the contrast ultrasound image can more accurately determine the position of the blood vessels in the target tissue in the contrast ultrasound image.

[0045] It can be understood that when the initial contrast image is preprocessed to obtain a contrast ultrasound image, by reducing the resolution of the contrast ultrasound image, the segmentation efficiency can be improved when the blood vessels in the target tissue in the contrast ultrasound image are segmented. By reducing the interference of information other than the target tissue, the segmentation accuracy can be improved when the blood vessels in the target tissue in the contrast ultrasound image are segmented, which is conducive to more accurately determining the position of the blood vessels in the target tissue in the contrast ultrasound image.

[0046] In one embodiment, the above step S104, segmenting the blood vessels in the target tissue in the contrast ultrasound image to determine the position of the blood vessels in the contrast ultrasound image, includes the following steps S301 to S302.

[0047] S301, setting the target tissue in the contrast ultrasound image as a region of interest.

[0048] S302: Input the contrast-enhanced ultrasound image with the region of interest set into a blood vessel segmentation model to obtain the position of the blood vessel in the contrast-enhanced ultrasound image.

[0049] The vascular segmentation model is obtained by supervised training with second training data and second supervisory data, the second training data is a second ultrasound image in which the reference tissue is marked as a reference region of interest, the second supervisory data is a reference blood vessel marked in the second ultrasound image, and the second ultrasound image is an ultrasound image in ultrasound contrast mode.

[0050] The process of segmenting the blood vessels in the target tissue in the contrast ultrasound image can be completed using the blood vessel segmentation model. Specifically, the target tissue in the contrast ultrasound image is set as the region of interest (ROI). The region of interest is an image area selected from the image, which is the focus of image analysis and is used to represent the area in the image that needs to be analyzed and processed. In this embodiment, the target tissue is set as the region of interest so that the subsequent analysis and processing of the contrast ultrasound image is only performed on the target tissue, which can reduce the processing time and improve the processing accuracy. Then, the contrast ultrasound image with the region of interest is input into the blood vessel segmentation model, and the blood vessel segmentation model is used to segment the blood vessels in the region of interest. The training process of the blood vessel segmentation model is as follows.

[0051] First, accurate data. The data to be prepared include at least the second training data and the second supervisory data. The second training data is a second ultrasound image in which the reference tissue is marked as a reference region of interest, and the second ultrasound image is an ultrasound image in ultrasound contrast mode. The reference tissue should be consistent with the target tissue. For example, when the target tissue is the kidney, the reference tissue is also the kidney. The purpose of the reference tissue in the second ultrasound image being marked as a reference region of interest by a professional is to make the model only process the reference region of interest in the second ultrasound image, that is, only process the reference tissue. Exemplarily, the second ultrasound image is an ultrasound image obtained after preprocessing and removing information other than the reference tissue. The second ultrasound image obtained after preprocessing is used as the second training data, and by reducing the resolution of the second ultrasound image, it is conducive to making the model converge faster during the training process. It can be understood that in the case where the second ultrasound image is an ultrasound image obtained after preprocessing, the contrast ultrasound image should also be an ultrasound image obtained after preprocessing. The second supervisory data is the reference blood vessel marked by a professional in the second ultrasound image, and the reference blood vessel is located in the reference tissue. And then. The model is trained with parameter supervision using the second training data and the second supervision data, and the training is stopped after the model converges to obtain a blood vessel segmentation model. In addition, a corresponding test data set can be prepared to test the trained blood vessel segmentation model to determine the prediction accuracy of the blood vessel segmentation model.

[0052] In this embodiment, the blood vessels in the target tissue in the contrast-enhanced ultrasound image are segmented using the trained blood vessel segmentation model, so that the position of the blood vessels in the target tissue in the contrast-enhanced ultrasound image can be determined more accurately.

[0053] By means of the blood vessel segmentation method in the embodiment of the present application, the blood vessels in the target tissue in the contrast ultrasound image can be segmented more accurately, and the position of the blood vessels in the ultrasound image in the ultrasound contrast mode can be determined more accurately. Exemplarily, the above blood vessel segmentation method can be regarded as a blood vessel segmentation method based on a cascade neural network. Among them, the cascade neural network includes a contour segmentation model and a blood vessel segmentation model, and the cascade neural network can fuse multimodal ultrasound image data, and the multimodal ultrasound image data includes an ultrasound image in the B mode and an ultrasound image in the ultrasound contrast mode. Specifically, a brightness ultrasound image in the B mode and a contrast ultrasound image in the ultrasound contrast mode are obtained, and both the brightness ultrasound image and the contrast ultrasound image are ultrasound images of the target tissue of the object, and the position of the target tissue in the brightness ultrasound image is the same as the position of the target tissue in the contrast ultrasound image. Exemplarily, the brightness ultrasound image and the contrast ultrasound image are ultrasound images obtained after preprocessing and removing information other than the reference tissue. The brightness ultrasound image is input into the contour segmentation model in the cascade neural network to obtain the position of the contour of the target tissue in the brightness ultrasound image. According to the position of the contour in the brightness ultrasound image, the position of the target tissue in the contrast ultrasound image is determined. The target tissue in the contrast ultrasound image is set as a region of interest, and the contrast ultrasound image with the region of interest is input into the blood vessel segmentation model in the cascade neural network to obtain the position of the blood vessels in the target tissue in the contrast ultrasound image. Through the above-mentioned cascade neural network, a deep learning method is used to segment the contour of the target tissue in the ultrasound image under the B mode step by step, and the blood vessels in the target tissue in the ultrasound image under the ultrasound contrast mode are segmented. In this way, the blood vessels in the target tissue in the ultrasound image under the ultrasound contrast mode are segmented more accurately and quickly, and the position of the blood vessels in the target tissue in the ultrasound image under the ultrasound contrast mode is determined more accurately and quickly.

[0054] It should be noted that the above-mentioned cascade neural network can be obtained by respectively training the contour segmentation model and the blood vessel segmentation model. The training process of the contour segmentation model and the blood vessel segmentation model is as described above. When the contour segmentation model and the blood vessel segmentation model are obtained through training, a cascade neural network including the contour segmentation model and the blood vessel segmentation model can be obtained. In this embodiment, the training process of the cascade neural network will not be described in detail.

[0055] By using the blood vessel segmentation method in the embodiment of the present application, based on the premise that the position of the target tissue in the brightness ultrasound image in the B mode is the same as the position in the contrast ultrasound image in the ultrasound contrast mode, the contour of the target tissue in the brightness ultrasound image is accurately segmented to obtain the position of the contour in the brightness ultrasound image, and then the position of the target tissue in the contrast ultrasound image can be determined. Furthermore, the blood vessels in the target tissue in the contrast ultrasound image are segmented, so that the segmentation results of the blood vessels in the target tissue in the ultrasound image in the ultrasound contrast mode can be obtained quickly and more accurately. That is, the position of the blood vessels in the ultrasound image in the ultrasound contrast mode can be determined quickly and more accurately. The blood vessel segmentation method is applied to ultrasound contrast guided surgery, by acquiring the brightness ultrasound image in the B mode and the contrast ultrasound image in the ultrasound contrast mode in real time, and segmenting the blood vessels in the target tissue in the contrast ultrasound image based on the brightness ultrasound image and the contrast ultrasound image, the position of the blood vessels in the ultrasound image in the ultrasound contrast mode can be determined in real time and more accurately. Furthermore, by marking the position of blood vessels in ultrasound images under ultrasound contrast mode such as contrast ultrasound images or initial contrast images in real time, it is helpful for doctors to quickly and more accurately determine the distribution of blood vessels in the target tissue. Making the vascular imaging display retained is also beneficial to delay the time of clinical judgment, improve the accuracy of doctors in clinical judgment, and reduce surgical risks. In ultrasound contrast-guided puncture surgery, marking the position of blood vessels in ultrasound images under ultrasound contrast mode is also beneficial for more accurate planning of puncture paths and more accurate puncture when the puncture surface scanned by the ultrasound probe is fixed. In the case where the present blood vessel segmentation method is a blood vessel segmentation method based on a cascade neural network, and the cascade neural network includes a contour segmentation model and a blood vessel segmentation model, the present blood vessel segmentation method is based on the cascade neural network, which realizes more accurate and rapid segmentation of blood vessels in target tissues in ultrasound images under ultrasound contrast mode, and fills the gap in the algorithm for segmenting blood vessels in target tissues in ultrasound images under ultrasound contrast mode.

[0056] In one of the embodiments, the number of brightness ultrasound images and contrast ultrasound images is at least two; the position of the target tissue in the brightness ultrasound image is the same as the position of the target tissue in the contrast ultrasound image, which means that the position of the target tissue is the same in the brightness ultrasound image and the contrast ultrasound image with the same acquisition timestamp; there is a one-to-one correspondence between the at least two brightness ultrasound images, the at least two contrast ultrasound images and the at least two acquisition timestamps.

[0057] Based on the above-mentioned limitations on brightness ultrasound images and contrast ultrasound images, the blood vessel segmentation method further includes the following steps S401 to S407. Among them, steps S401 to S404 correspond one-to-one to steps S101 to S104 in the aforementioned embodiment. The relevant discussion and beneficial effects of steps S401 to S404 in this embodiment can refer to the relevant discussion and beneficial effects of steps S101 to S104 in the aforementioned embodiment, and will not be repeated in this embodiment.

[0058] S401, acquiring at least two brightness ultrasound images in a brightness mode and at least two contrast ultrasound images in an ultrasound contrast mode. The at least two brightness ultrasound images and the at least two contrast ultrasound images are both ultrasound images containing target tissues of the object; in the brightness ultrasound images and the contrast ultrasound images with the same acquisition timestamp, the positions of the target tissues are the same, and there is a one-to-one correspondence between the at least two brightness ultrasound images, the at least two contrast ultrasound images, and the at least two acquisition timestamps.

[0059] Specifically, it is assumed that two brightness ultrasound images A1 and A2 and two contrast ultrasound images a1 and a2 are acquired. The one-to-one correspondence between at least two brightness ultrasound images, at least two contrast ultrasound images and at least two acquisition timestamps means that the brightness ultrasound image A1 and one of the two contrast ultrasound images a1 and a2 have the same acquisition timestamp t1; the brightness ultrasound image A2 and the other of the two contrast ultrasound images a1 and a2 have the same acquisition timestamp t2. Among them, the acquisition timestamp t1 is different from the acquisition timestamp t2. The position of the target tissue in the brightness ultrasound image and the contrast ultrasound image with the same acquisition timestamp is the same. Based on this, the position of the target tissue in the contrast ultrasound image of the target can be determined according to the position of the contour of the target tissue in any brightness ultrasound image. Among them, the contrast ultrasound image of the target is a contrast ultrasound image with the same acquisition timestamp as any of the brightness ultrasound images.

[0060] Exemplarily, the above step S401, at least two brightness ultrasound images in brightness mode and at least two contrast ultrasound images in ultrasound contrast mode, includes the following steps.

[0061] At least two initial brightness images and at least two initial contrast images are acquired. The initial brightness image is an ultrasound image in a brightness mode, and the initial contrast image is an ultrasound image in an ultrasound contrast mode. Both the initial brightness image and the initial contrast image are ultrasound images obtained by scanning a target tissue of the object. There is a one-to-one correspondence between the at least two initial brightness images, the at least two initial contrast images, and the at least two acquisition timestamps.

[0062] Preprocessing is performed on at least two initial brightness images to obtain at least two brightness ultrasound images.

[0063] Preprocessing is performed on at least two initial contrast-enhanced images to obtain at least two contrast-enhanced ultrasound images.

[0064] Among the at least two brightness ultrasound images and the at least two contrast ultrasound images obtained after preprocessing, the brightness ultrasound images and the contrast ultrasound images with the same acquisition timestamp contain the same information and have the same size.

[0065] Specifically, the ultrasound probe performs ultrasound scanning on the target tissue of the object in the B mode and the ultrasound contrast mode at the same time, and correspondingly obtains at least two initial brightness images in the B mode and at least two initial contrast images in the ultrasound contrast mode. Among them, the ultrasound video acquired by the ultrasound probe can be in two modes, and at least two initial brightness images and at least two initial contrast images are obtained by dividing the video frames of each ultrasound video. The blood vessel segmentation device that executes the blood vessel segmentation method in the embodiment of the present application obtains at least two initial brightness images and at least two initial contrast images. There is a one-to-one correspondence between the at least two initial brightness images, the at least two initial contrast images and the at least two acquisition timestamps.

[0066] S402, segmenting the contour of the target tissue in each brightness ultrasound image to determine the position of the contour in each brightness ultrasound image.

[0067] Exemplarily, the process of segmenting the contour of the target tissue in each brightness ultrasound image can be completed by using a contour segmentation model.

[0068] S403: Determine the position of the target tissue in each contrast ultrasound image according to the position of the contour in each brightness ultrasound image.

[0069] Specifically, the position of the target tissue in the contrast ultrasound image of the target is determined according to the position of the contour of the target tissue in any one brightness ultrasound image. The contrast ultrasound image of the target is a contrast ultrasound image with the same acquisition timestamp as any one brightness ultrasound image. Based on this, the position of the target tissue in each contrast ultrasound image can be determined.

[0070] S404: segment the blood vessels in the target tissue in each contrast-enhanced ultrasound image to determine the position of the blood vessels in each contrast-enhanced ultrasound image.

[0071] Exemplarily, the process of segmenting the blood vessels in the target tissue in each contrast ultrasound image can be completed using a blood vessel segmentation model.

[0072] S405: Determine the image brightness values ​​of at least two contrast ultrasound images respectively. The image brightness value is used to represent the brightness of the contrast ultrasound image, and at least two acquisition time stamps of the at least two contrast ultrasound images are continuous.

[0073] The at least two acquisition time stamps of the at least two contrast ultrasound images are continuous, which means that the at least two contrast ultrasound images are continuously acquired within a time period. If the at least two contrast ultrasound images are obtained after preprocessing the at least two initial contrast images, then the at least two initial contrast images are continuously acquired within a time period.

[0074] In one embodiment, the above step S405, respectively determining the image brightness values ​​of at least two contrast ultrasound images, includes the following steps S501 to S503.

[0075] S501, convert at least two contrast ultrasound images into grayscale images respectively.

[0076] S502, calculating the average brightness value of the pixel points of the target tissue in each grayscale image.

[0077] S503: Using the average brightness value as the image brightness value of the corresponding contrast-enhanced ultrasound image.

[0078] In order to reduce processing time and improve processing accuracy, based on the purpose of determining the position of blood vessels in the target tissue in the contrast ultrasound image, this embodiment can set the target tissue as a region of interest. Only the characteristics of the target tissue in the contrast ultrasound image are focused on, and only the target tissue is analyzed and processed. Based on this, the brightness of the contrast ultrasound image to which the target tissue belongs is represented by the brightness of the target tissue, and the brightness value of the target tissue is represented by the image brightness value of the contrast ultrasound image to which the target tissue belongs.

[0079] Specifically, at least two contrast ultrasound images are converted into grayscale images respectively, and the average brightness value of the pixel points of the target tissue in each grayscale image is calculated. The average brightness value of the pixel points is used to represent the brightness value of the corresponding target tissue. Furthermore, the average brightness value of the pixel points of each target tissue is used as the image brightness value of the contrast ultrasound image to which each target tissue belongs.

[0080] S406: Determine a critical time period for ultrasound contrast imaging based on the image brightness value and acquisition timestamp of each contrast ultrasound image. The image brightness value of the contrast ultrasound image acquired in the critical time period is greater than a brightness threshold.

[0081] The key time period of ultrasound contrast imaging can be understood as the time period when ultrasound contrast imaging is relatively clear, or as the best time period for vascular development in contrast ultrasound images. The higher the image brightness value of the contrast ultrasound image, the clearer the ultrasound contrast imaging is, and the clearer the vascular development in the target tissue in the contrast ultrasound image is. Based on this, the time period when ultrasound contrast imaging is relatively clear can be determined by the image brightness values ​​of at least two contrast ultrasound images acquired at consecutive acquisition timestamps, that is, the key time period of ultrasound contrast imaging is determined. Specifically, the brightness threshold is determined according to actual conditions, scene requirements, clinical experience, etc. When the image brightness value of the contrast ultrasound image is greater than the brightness threshold, the clarity of ultrasound contrast imaging meets the requirements, and the vascular development in the target tissue in the contrast ultrasound image is clearer. Based on this, the key time period of ultrasound contrast imaging is determined by at least two consecutive acquisition timestamps corresponding to at least two contrast ultrasound images whose image brightness values ​​are greater than the brightness threshold.

[0082] Exemplarily, a brightness variation curve graph may be drawn based on the image brightness value and acquisition timestamp of each contrast ultrasound image. Figure 2 This is a brightness change curve diagram provided in the embodiment of the present application. Figure 2 As shown in the figure, the horizontal axis of the brightness change curve is time, that is, the acquisition timestamp. The vertical axis is brightness, that is, the image brightness value. The brightness change curve in the brightness change curve is used to analyze the image sequence of the continuous ultrasound contrast imaging mode. Based on the brightness change curve, the brightness threshold is used as the standard to determine the time of the target corresponding to the brightness greater than the brightness threshold. The time period formed by the target time is the key time period of ultrasound contrast imaging. Figure 2 As shown, the brightness pointed to by the horizontal line in the brightness change curve diagram represents the brightness threshold; the time period formed by the time pointed to by the two vertical lines in the brightness change curve diagram represents the key time period.

[0083] It should be understood that in actual situations, due to the instability of ultrasound contrast imaging, errors in image brightness value calculation, etc., the image brightness values ​​of continuously acquired contrast ultrasound images do not continuously increase or decrease. Figure 2 As shown in FIG. 1 , the brightness of the contrast ultrasound images collected in a continuous time period fluctuates in the process of steadily increasing or steadily decreasing. Based on this, among the contrast ultrasound images collected in the critical time period, there are still contrast ultrasound images whose brightness is lower than the brightness threshold, which is a normal phenomenon. In the process of determining the critical time period based on the brightness threshold, this phenomenon can be ignored.

[0084] In contrast ultrasound images acquired during this critical period, the blood vessels of the target tissue are clearly visualized. Figure 2As shown in the figure, the brightness of the contrast ultrasound image collected during the arterial phase increases steadily, and the vascular development of the target tissue gradually becomes clearer. The brightness of the contrast ultrasound image collected during the venous phase decreases steadily, and the vascular development of the target tissue gradually becomes blurred. During the critical phase, that is, during the critical time period, the brightness of the contrast ultrasound image collected is greater than the brightness threshold, and the vascular development of the target tissue is clearer.

[0085] S407, merging the positions of at least two blood vessels in at least two target ultrasound images to obtain the position of the target blood vessel. The target ultrasound image refers to a contrast ultrasound image whose acquisition timestamp is within the key time period.

[0086] Since the blood vessels of the target tissue are more clearly displayed in the contrast ultrasound image within the critical time period, that is, in the target ultrasound image, the blood vessels in the target tissue in the target ultrasound image are segmented to obtain relatively complete blood vessels. In other words, the position of relatively complete blood vessels can be determined. Furthermore, by merging the positions of at least two blood vessels in at least two target ultrasound images collected continuously, the position of the target blood vessel can be obtained. The position of this target blood vessel is a complete and accurate position of the blood vessel. This step can also be understood as merging the blood vessels segmented from at least two target ultrasound images respectively, so that the distribution of complete and accurate blood vessels in the target tissue can be obtained. It should be noted that the merging described in this embodiment can be understood as taking a union of data. Merging the blood vessels means taking a union of the blood vessels to obtain a complete and accurate blood vessel.

[0087] For example, the blood vessels are divided into five parts, 1 to 5, and it is assumed that three contrast ultrasound images are collected within the critical time period. Based on the vascular development, the blood vessels in the three contrast ultrasound images are segmented respectively. Through the first contrast ultrasound image, the blood vessels 1, 2, 3 and 4 parts are segmented, and the positions of the blood vessels 1, 2, 3 and 4 parts are determined; through the second contrast ultrasound image, the blood vessels 1, 2, 3 and 5 parts are segmented, and the positions of the blood vessels 1, 2, 3 and 5 parts are determined; through the third contrast ultrasound image, the blood vessels 2, 3, 4 and 5 parts are segmented, and the positions of the blood vessels 2, 3, 4 and 5 parts are determined. Then, the positions of the blood vessels in the three contrast ultrasound images are merged. That is, the blood vessels segmented from the three contrast ultrasound images are merged. Specifically, the blood vessels 1, 2, 3 and 4 parts, the blood vessels 1, 2, 3 and 5 parts, and the blood vessels 2, 3, 4 and 5 parts are merged, and then the blood vessels 1 to 5, a total of five parts, can be obtained. That is, the complete and accurate blood vessels are segmented, and the positions of the complete and accurate blood vessels can be determined.

[0088] For example, Figure 3This is a second flow chart of the blood vessel segmentation method provided in the embodiment of the present application. Taking the target tissue as kidney as an example, a feasible execution process of the blood vessel segmentation method is as follows: Figure 3 As shown. First, the original image is obtained, that is, at least two initial brightness images and at least two initial contrast images are obtained. The initial brightness image is an ultrasound image in B mode, and the initial contrast image is an ultrasound image in ultrasound contrast mode. Both the initial brightness image and the initial contrast image are ultrasound images obtained by scanning the kidneys of the object. There is a one-to-one correspondence between at least two initial brightness images, at least two initial contrast images and at least two acquisition timestamps. At least two acquisition timestamps are continuous. Then, at least two initial brightness images and at least two initial contrast images are preprocessed respectively to obtain a brightness ultrasound image in B mode and a contrast ultrasound image in ultrasound contrast mode. The at least two brightness ultrasound images and at least two contrast ultrasound images obtained after preprocessing are both ultrasound images containing the kidneys of the object. There is a one-to-one correspondence between at least two brightness ultrasound images, at least two contrast ultrasound images and at least two acquisition timestamps. The kidneys are located in the same position in the brightness ultrasound image and the contrast ultrasound image with the same acquisition timestamp.

[0089] Furthermore, at least two brightness ultrasound images are respectively input into the contour segmentation model, and the contour segmentation model is used to segment the kidney contour in the brightness ultrasound image to obtain the position of the kidney contour in each brightness ultrasound image. Based on the same kidney position in the brightness ultrasound image and the contrast ultrasound image with the same acquisition timestamp, the position of the kidney in the contrast ultrasound image is determined according to the position of the kidney contour in the brightness ultrasound image. Furthermore, a region of interest is set. That is, the kidney in each contrast ultrasound image is set as the region of interest of the corresponding contrast ultrasound image. Subsequently, each contrast ultrasound image with a region of interest is input into the vascular segmentation model, and the renal blood vessels in the contrast ultrasound image are segmented using the vascular segmentation model to obtain the position of the renal blood vessels in each contrast ultrasound image. Compared with relying on the clinical experience of doctors to judge the position of the renal blood vessels, the position of the renal blood vessels in the ultrasound image in the ultrasound contrast mode is more accurately determined at this time.

[0090] Further, for each contrast ultrasound image with a region of interest, the brightness value of the region of interest is determined. The brightness value of the region of interest, that is, the brightness value of the kidney, is used to represent the image brightness value of the contrast ultrasound image to which the kidney belongs. Furthermore, based on the image brightness value and acquisition timestamp of each contrast ultrasound image, a brightness change curve is fitted and drawn. The brightness change curve is analyzed to determine the key time period of ultrasound contrast imaging. Among them, the image brightness value of the contrast ultrasound image acquired in the key time period is greater than the brightness threshold. This key time period is the best time period for renal vascular development in the contrast ultrasound image. Finally, multiple contrast ultrasound images with acquisition timestamps within the key time period are selected, and the multiple contrast ultrasound images are used as at least two target ultrasound images, and the positions of at least two renal blood vessels in the at least two target ultrasound images are merged to obtain a merged result. This merged result is a complete and accurate position of the renal blood vessels in the ultrasound image in the ultrasound contrast mode. At this time, the position of the renal blood vessels in the ultrasound image in the ultrasound contrast mode is accurately determined. Using the blood vessel segmentation method in this example, by acquiring the original image in real time, the position of the blood vessel in the ultrasound image in the ultrasound contrast mode can be accurately determined in real time.

[0091] By using the blood vessel segmentation method in this embodiment, based on the image brightness value and acquisition timestamp of each contrast ultrasound image, the key time period of ultrasound contrast imaging is determined, that is, the optimal time period for blood vessel development in the contrast ultrasound image is determined. Further, multiple contrast ultrasound images with acquisition timestamps within the key time period are determined, and the positions of the blood vessels of the target tissue in the multiple contrast images are merged, so that the complete and accurate positions of the blood vessels in the target tissue can be obtained. That is, the position of the blood vessels in the target tissue in the ultrasound image under ultrasound contrast mode can be accurately determined. The blood vessel segmentation method is applied to ultrasound contrast guided surgery, by acquiring at least two brightness ultrasound images in B mode and at least two contrast ultrasound images in ultrasound contrast mode in real time, and segmenting the blood vessels in the target tissue in the contrast ultrasound image based on the brightness ultrasound image and the contrast ultrasound image, the position of the blood vessels in the ultrasound image under ultrasound contrast mode can be determined in real time and accurately. Further, in ultrasound images under ultrasound contrast mode such as contrast ultrasound images or initial contrast images, the complete and accurate positions of the blood vessels are marked in real time, which is conducive to doctors to quickly and accurately judge the distribution of blood vessels in the target tissue. The retention of vascular imaging display is also conducive to delaying the time of clinical judgment, improving the accuracy of doctors' clinical judgment, and reducing surgical risks. In ultrasound angiography-guided puncture surgery, the complete and accurate location of the blood vessels is marked in the ultrasound image under ultrasound angiography mode. When the puncture surface scanned by the ultrasound probe is fixed, it is also conducive to more accurate planning of the puncture path and more accurate puncture.

[0092] It should be understood that the blood vessel segmentation method in this embodiment can also be regarded as a blood vessel segmentation method based on a cascade neural network. Among them, the cascade neural network includes a contour segmentation model and a blood vessel segmentation model, and the cascade neural network can fuse multi-modal ultrasound image data, and the multi-modal ultrasound image data includes ultrasound images in B mode and ultrasound images in ultrasound contrast mode. The blood vessel segmentation method based on the cascade neural network in this embodiment is not only based on the cascade neural network, but also realizes accurate and rapid segmentation of blood vessels in the target tissue in the ultrasound image in the ultrasound contrast mode, and fills the gap in the algorithm for segmenting blood vessels in the target tissue in the ultrasound image in the ultrasound contrast mode. In addition, a judgment mechanism for the key time period of ultrasound contrast imaging is introduced, which can determine the key time period of ultrasound contrast imaging based on the image brightness value and acquisition timestamp of each contrast ultrasound image. Then, based on the key time period, the positions of the blood vessels in the contrast ultrasound images collected in the key time period are merged to obtain a complete and accurate position of the blood vessels.

[0093] The embodiment of the present application also provides a blood vessel segmentation device. Figure 4 This is a schematic diagram of the structure of the blood vessel segmentation device provided in the embodiment of the present application. Figure 4 As shown, the blood vessel segmentation device 400 includes an image acquisition module 401, a contour segmentation module 402, a position determination module 403 and a blood vessel segmentation module 404. The image acquisition module 401 is used to acquire a brightness ultrasound image in a brightness mode and a contrast ultrasound image in an ultrasound contrast mode. Both the brightness ultrasound image and the contrast ultrasound image are ultrasound images containing the target tissue of the object. The position of the target tissue in the brightness ultrasound image is the same as the position of the target tissue in the contrast ultrasound image. The contour segmentation module 402 is used to segment the contour of the target tissue in the brightness ultrasound image to determine the position of the contour in the brightness ultrasound image. The position determination module 403 is used to determine the position of the target tissue in the contrast ultrasound image according to the position of the contour in the brightness ultrasound image. The blood vessel segmentation module 404 is used to segment the blood vessels in the target tissue in the contrast ultrasound image to determine the position of the blood vessels in the contrast ultrasound image.

[0094] In one embodiment, the number of brightness ultrasound images and contrast ultrasound images is at least two; the position of the target tissue in the brightness ultrasound image is the same as the position of the target tissue in the contrast ultrasound image, which means that the position of the target tissue is the same in the brightness ultrasound image and the contrast ultrasound image with the same acquisition timestamp; there is a one-to-one correspondence between at least two brightness ultrasound images, at least two contrast ultrasound images and at least two acquisition timestamps. The above-mentioned blood vessel segmentation device 400 also includes a brightness value determination module, a time period determination module and a blood vessel determination module. The brightness value determination module is used to determine the image brightness value of at least two contrast ultrasound images respectively, and the image brightness value is used to characterize the brightness of the contrast ultrasound image, and at least two acquisition timestamps of at least two contrast ultrasound images are continuous. The time period determination module is used to determine the key time period of ultrasound contrast imaging based on the image brightness value and acquisition timestamp of each contrast ultrasound image, and the image brightness value of the contrast ultrasound image acquired in the key time period is greater than the brightness threshold. The blood vessel determination module is used to merge the positions of at least two blood vessels in at least two target ultrasound images to obtain the position of the target blood vessel; the target ultrasound image refers to the contrast ultrasound image with an acquisition timestamp located in the key time period.

[0095] In one of the embodiments, the brightness value determination module is further used to convert at least two contrast ultrasound images into grayscale images respectively; calculate the average brightness value of the pixel points of the target tissue in each grayscale image; and use the average brightness value as the image brightness value of the corresponding contrast ultrasound image.

[0096] In one embodiment, the above-mentioned contour segmentation module 402 is also used to input the brightness ultrasound image into the contour segmentation model to obtain the position of the contour in the brightness ultrasound image; wherein the contour segmentation model is obtained by supervised training with first training data and first supervision data, the first training data is the first ultrasound image in the brightness mode, and the first supervision data is the reference contour of the reference tissue marked in the first ultrasound image.

[0097] In one of the embodiments, the above-mentioned blood vessel segmentation module 404 is also used to set the target tissue in the contrast ultrasound image as a region of interest; input the contrast ultrasound image with the region of interest into the blood vessel segmentation model to obtain the position of the blood vessel in the contrast ultrasound image; wherein the blood vessel segmentation model is obtained by supervised training with second training data and second supervisory data, the second training data is a second ultrasound image in which the reference tissue is marked as a reference region of interest, the second supervisory data is a reference blood vessel marked in the second ultrasound image, and the second ultrasound image is an ultrasound image under ultrasound contrast mode.

[0098] In one embodiment, the image acquisition module 401 is further used to acquire an initial brightness image and an initial contrast image with the same acquisition timestamp, wherein the initial brightness image is an ultrasound image in a brightness mode, and the initial contrast image is an ultrasound image in an ultrasound contrast mode, and both the initial brightness image and the initial contrast image are ultrasound images obtained by scanning the target tissue of the object; the initial brightness image is preprocessed to obtain a brightness ultrasound image; the initial contrast image is preprocessed to obtain a contrast ultrasound image; wherein the brightness ultrasound image and the contrast ultrasound image obtained after the preprocessing contain the same information and size.

[0099] For explanations of the terms, please refer to the related descriptions in the aforementioned embodiment of the blood vessel segmentation method, which will not be elaborated here.

[0100] It should be noted that the specific execution process of the above-mentioned blood vessel segmentation device 400 can be found in Figures 1 to 3 The specific description of the illustrated embodiment will not be repeated here.

[0101] Each module in the above-mentioned blood vessel segmentation device 400 can be implemented in whole or in part by software, hardware or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0102] The embodiment of the present application also provides a computer device. Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present application. Figure 5 As shown, illustratively, the computer device 500 may include a processor 501, a communication interface 502, a communication bus 503, and a memory 504. Specifically, the computer device 500 may include: At least one processor 501, such as a CPU, at least one communication interface 502, a memory 504, and at least one communication bus 503. The communication bus 503 is used to realize the connection and communication between these components. The communication interface 502 may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface or a Bluetooth interface, etc.). The memory 504 may be a high-speed RAM memory, or a non-volatile memory (non-volatile memory), such as at least one disk storage. The memory 504 may also be at least one storage device located away from the aforementioned processor 501. Figure 5 As shown, the memory 504 as a computer storage medium may include an operating system and program instructions.

[0103] Exemplarily, the processor 501 may be used to implement the above Figure 4The steps or methods performed by the image acquisition module 401, the contour segmentation module 402, the position determination module 403 and the blood vessel segmentation module 404 in FIG.

[0104] It is understandable that the above method is only an example, and the processor 501 and other modules in the above computer device 500 may also cooperate to perform the above method. Figure 4 The steps or methods performed by the image acquisition module 401, the contour segmentation module 402, the position determination module 403 and the blood vessel segmentation module 404 are not limited in this document.

[0105] exist Figure 5 In the computer device 500 shown, the processor 501 may be used to load program instructions stored in the memory 504 and specifically perform the following operations: Acquire a brightness ultrasound image in a brightness mode and a contrast ultrasound image in an ultrasound contrast mode, wherein both the brightness ultrasound image and the contrast ultrasound image are ultrasound images containing a target tissue of the object, and a position of the target tissue in the brightness ultrasound image is the same as a position of the target tissue in the contrast ultrasound image; Segmenting the contour of the target tissue in the brightness ultrasound image to determine the position of the contour in the brightness ultrasound image; determining the position of the target tissue in the contrast ultrasound image according to the position of the contour in the brightness ultrasound image; The blood vessels in the target tissue in the contrast ultrasound image are segmented to determine the positions of the blood vessels in the contrast ultrasound image.

[0106] For explanations of the terms, please refer to the related descriptions in the aforementioned embodiment of the blood vessel segmentation method, which will not be elaborated here.

[0107] It should be noted that the specific implementation process can be found in Figures 1 to 3 The specific description of the illustrated embodiment will not be repeated here.

[0108] The present application also provides a computer-readable storage medium, which can store a plurality of instructions, and the instructions are suitable for being loaded and executed by a processor as described above. Figures 1 to 3 The method steps of the embodiment shown in the figure can be found in the specific execution process. Figures 1 to 3 The specific description of the illustrated embodiment will not be repeated here.

[0109] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0110] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any fusion thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk), etc.

[0111] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

[0112] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0113] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A blood vessel segmentation method, characterized in that: The method comprises: Acquire a brightness ultrasound image in a brightness mode and a contrast ultrasound image in an ultrasound contrast mode, wherein both the brightness ultrasound image and the contrast ultrasound image are ultrasound images containing a target tissue of an object, and a position of the target tissue in the brightness ultrasound image is the same as a position of the target tissue in the contrast ultrasound image; Segmenting the contour of the target tissue in the brightness ultrasound image to determine the position of the contour in the brightness ultrasound image; Determining the position of the target tissue in the contrast ultrasound image according to the position of the contour in the brightness ultrasound image; The blood vessels in the target tissue in the contrast-enhanced ultrasound image are segmented to determine the positions of the blood vessels in the contrast-enhanced ultrasound image.

2. The method according to claim 1, characterized in that The number of the brightness ultrasound images and the number of the contrast ultrasound images are both at least two; the position of the target tissue in the brightness ultrasound image is the same as the position of the target tissue in the contrast ultrasound image, which means that the position of the target tissue is the same in the brightness ultrasound image and the contrast ultrasound image with the same acquisition timestamp; there is a one-to-one correspondence between the at least two brightness ultrasound images, the at least two contrast ultrasound images and the at least two acquisition timestamps; After segmenting the blood vessels in the target tissue in the contrast-enhanced ultrasound image to determine the positions of the blood vessels in the contrast-enhanced ultrasound image, the method further includes: respectively determining image brightness values ​​of at least two of the contrast ultrasound images, wherein the image brightness values ​​are used to characterize the brightness of the contrast ultrasound images, and at least two acquisition time stamps of at least two of the contrast ultrasound images are continuous; Determining a critical time period for ultrasound contrast imaging based on the image brightness value and the acquisition timestamp of each of the contrast ultrasound images, wherein the image brightness value of the contrast ultrasound image acquired in the critical time period is greater than a brightness threshold; The positions of at least two of the blood vessels in at least two target ultrasound images are merged to obtain the position of the target blood vessel; the target ultrasound image refers to a contrast ultrasound image whose acquisition timestamp is within the key time period.

3. The method according to claim 2, characterized in that The step of respectively determining the image brightness values ​​of at least two of the contrast ultrasound images comprises: Converting at least two of the contrast ultrasound images into grayscale images respectively; Calculating the average brightness value of the pixel points of the target tissue in each of the grayscale images; The average brightness value is used as the image brightness value of the corresponding contrast ultrasound image.

4. The method according to any one of claims 1 to 3, characterized in that: The step of segmenting the contour of the target tissue in the brightness ultrasound image to determine the position of the contour in the brightness ultrasound image includes: Inputting the brightness ultrasound image into a contour segmentation model to obtain the position of the contour in the brightness ultrasound image; The contour segmentation model is obtained by supervised training with first training data and first supervised data, wherein the first training data is a first ultrasound image in brightness mode, and the first supervised data is a reference contour of a reference tissue marked in the first ultrasound image.

5. The method according to any one of claims 1 to 3, characterized in that: The segmenting of the blood vessels in the target tissue in the contrast-enhanced ultrasound image to determine the position of the blood vessels in the contrast-enhanced ultrasound image includes: Setting the target tissue in the contrast ultrasound image as a region of interest; Inputting the contrast ultrasound image provided with the region of interest into a blood vessel segmentation model to obtain the position of the blood vessel in the contrast ultrasound image; The blood vessel segmentation model is obtained by supervised training with second training data and second supervisory data, the second training data is a second ultrasound image in which a reference tissue is marked as a reference region of interest, the second supervisory data is a reference blood vessel marked in the second ultrasound image, and the second ultrasound image is an ultrasound image in ultrasound contrast mode.

6. The method according to any one of claims 1 to 3, characterized in that: The step of acquiring a brightness ultrasound image in a brightness mode and a contrast ultrasound image in an ultrasound contrast mode comprises: Acquire an initial brightness image and an initial contrast image with the same acquisition timestamp, wherein the initial brightness image is an ultrasound image in a brightness mode, and the initial contrast image is an ultrasound image in an ultrasound contrast mode, and both the initial brightness image and the initial contrast image are ultrasound images obtained by scanning the target tissue of the object; Preprocessing the initial brightness image to obtain the brightness ultrasound image; Preprocessing the initial contrast image to obtain the contrast ultrasound image; The brightness ultrasound image and the contrast ultrasound image obtained after preprocessing contain the same information and have the same size.

7. A blood vessel segmentation device, characterized in that: The device comprises: an image acquisition module, used to acquire a brightness ultrasound image in a brightness mode and a contrast ultrasound image in an ultrasound contrast mode, wherein both the brightness ultrasound image and the contrast ultrasound image are ultrasound images containing a target tissue of an object, and a position of the target tissue in the brightness ultrasound image is the same as a position of the target tissue in the contrast ultrasound image; A contour segmentation module, used to segment the contour of the target tissue in the brightness ultrasound image to determine the position of the contour in the brightness ultrasound image; a position determination module, configured to determine the position of the target tissue in the contrast ultrasound image according to the position of the contour in the brightness ultrasound image; The blood vessel segmentation module is used to segment the blood vessels in the target tissue in the contrast ultrasound image to determine the position of the blood vessels in the contrast ultrasound image.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory stores program instructions; when the program instructions are executed by the processor, the processor executes the method as claimed in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program; when the computer program is run on one or more processors, the method according to any one of claims 1 to 6 is executed.

10. A computer program product, characterized in that The computer program product comprises a computer program or instructions; when the computer program or instructions are run on a computer, the computer is caused to execute the method according to any one of claims 1 to 6.

Images