CN117710432A

CN117710432A - Ultrasonic image processing method, ultrasonic image processing device, computer equipment and storage medium

Info

Publication number: CN117710432A
Application number: CN202311650043.XA
Authority: CN
Inventors: 李肯立; 彭锋锋; 段明星; 谭光华; 唐卓; 刘楚波; 肖国庆; 朱宁波; 李胜利
Original assignee: Hunan University
Current assignee: Hunan University
Priority date: 2023-12-05
Filing date: 2023-12-05
Publication date: 2024-03-15

Abstract

The present application relates to a method, an apparatus, a computer device, a storage medium and a computer program product for processing an ultrasound image. The method comprises the following steps: acquiring a plurality of ultrasonic section images, automatically grabbing each ultrasonic section image, screening out a standard section image from each ultrasonic section image according to the scoring result of each anatomical structure in the ultrasonic section images, and obtaining an accurate standard section image while improving the thickness measurement efficiency of a target structure without manually saving the ultrasonic section images by a doctor. According to the contour features of the target anatomical structure in the standard tangent plane image, the contour of the target anatomical structure is corroded to obtain a skeleton line of the target anatomical structure, each target point on the skeleton line is traversed, the thickness value of the target anatomical structure is determined according to the intersecting point between the normal line passing through the target point and the contour of the target anatomical structure, manual labeling of the thickness value is not needed, and manual processing time is saved.

Description

Ultrasonic image processing method, ultrasonic image processing device, computer equipment and storage medium

Technical Field

The present invention relates to the field of image processing technology, and in particular, to a method, an apparatus, a computer device, a storage medium, and a computer program product for processing an ultrasound image.

Background

In the current ultrasonic detection method, a doctor is required to move a probe to scan to form a video stream, then a relatively standard section image is manually found out from the video stream, and the section image contains various structures, so that the fetus is subjected to disease screening. Specifically, for a specific structure in a section image, the thickness of the structure needs to be obtained, and the thickness of the structure can be obtained to provide a reference for disease screening.

In the existing mode, whether the section image is a standard image or not and whether the thickness of the structure is the standard image or not depend on subjective judgment of doctors, and the problem of inaccurate thickness measurement of the specific structure exists.

Disclosure of Invention

In view of the foregoing, it is desirable to provide a method, an apparatus, a computer device, a computer-readable storage medium, and a computer program product for processing an ultrasound image that can improve the accuracy of measurement of a thickness value of a target structure.

In a first aspect, the present application provides a method for processing an ultrasound image. The method comprises the following steps:

acquiring a plurality of ultrasound section images, wherein each ultrasound section image comprises a plurality of anatomical structures; screening out a standard section image from each ultrasonic section image according to the scoring result of each anatomical structure in the ultrasonic section image; according to the contour features of the target anatomical structure in the standard section image, performing corrosion treatment on the contour of the target anatomical structure to obtain a skeleton line of the target anatomical structure; traversing each target point on the skeleton line and determining a thickness value of the target anatomy from an intersection point between a normal line passing through the target point and a contour of the target anatomy.

In one embodiment, the step of screening the standard section image from the ultrasonic section images according to the respective scoring result of each anatomical structure in the ultrasonic section images includes: classifying each anatomical structure of the ultrasound sectional images according to the importance of the anatomical structure aiming at each ultrasound sectional image to obtain a first type anatomical structure and a second type anatomical structure, wherein the importance of the first type anatomical structure is greater than that of the second type anatomical structure; if the type of the first type anatomical structure meets the verification requirement, scoring the first type anatomical structure and the second type anatomical structure to obtain the total score of the ultrasonic section image; and determining the ultrasonic section image with the total score being greater than or equal to a preset value as a standard section image.

In one embodiment, the step of scoring the first type of anatomical structure and the second type of anatomical structure to obtain a total score of the ultrasound sectional image comprises: weighting the scoring results of each anatomical structure in the first type of anatomical structure according to the weight coefficient matched with the type of the first type of anatomical structure to obtain a first scoring result of the first type of anatomical structure; assigning scores to the anatomical structures in the second class according to scoring criteria matched with the type of the anatomical structure in the second class to obtain a second scoring result of the anatomical structure in the second class; and combining the first scoring result and the second scoring result to obtain the total score of the ultrasonic section image.

In one embodiment, according to the contour features of the target anatomical structure in the standard section image, performing corrosion treatment on the contour of the target anatomical structure to obtain a skeleton line of the target anatomical structure, including: extracting the outline of the target anatomical structure according to the outline characteristics of the target anatomical structure in the standard section image to obtain an outline image of the target anatomical structure; performing binarization processing on the contour image to obtain a binarized contour image; and performing corrosion treatment on the profile image subjected to the binarization treatment according to preset structural elements to obtain the skeleton line of the target anatomical structure.

In one embodiment, the step of traversing each target point on the skeleton line and determining a thickness value of the target anatomy from an intersection between a normal passing through the target point and a contour of the target anatomy comprises: traversing each target point on the skeleton line, and determining each normal line segment according to an intersection point between a normal line passing through the target point and the outline of the target structure; screening each normal line segment according to the length sequencing result of the normal line segments to obtain each initial line segment; according to the pixel values at the end points of the initial line segments, adjusting each initial line segment to obtain each target line segment corresponding to each initial line segment; and determining the thickness value of the target structure according to the length sequencing result of each target line segment.

In one embodiment, the step of acquiring a plurality of ultrasound sectional images includes: acquiring an ultrasonic video stream; extracting the ultrasonic video stream to obtain a plurality of ultrasonic video frame images; and respectively intercepting each ultrasonic video frame image according to the image annotation type of the ultrasonic video frame image to obtain a plurality of ultrasonic section images.

In a second aspect, the present application further provides an ultrasound image processing apparatus. The device comprises:

the image acquisition module acquires a plurality of ultrasonic section images, wherein each ultrasonic section image comprises a plurality of anatomical structures; the image screening module is used for screening standard section images from the ultrasonic section images according to the scoring results of each anatomical structure in the ultrasonic section images; the processing module is used for corroding the outline of the target anatomical structure according to the outline characteristics of the target anatomical structure in the standard section image to obtain a skeleton line of the target anatomical structure; and the output module is used for traversing each target point on the skeleton line and determining the thickness value of the target anatomical structure according to the intersection point between the normal line passing through the target point and the outline of the target anatomical structure.

In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor which when executing the computer program performs the steps of:

In a fourth aspect, the present application also provides a computer-readable storage medium. The computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:

In a fifth aspect, the present application also provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements the steps of:

According to the processing method, the device, the computer equipment, the storage medium and the computer program product of the ultrasonic image, a plurality of ultrasonic section images are acquired, each ultrasonic section image comprises a plurality of anatomical structures, each ultrasonic section image is automatically grabbed, the standard section image is screened out from each ultrasonic section image according to the respective grading result of each anatomical structure in the ultrasonic section image, a doctor does not need to manually store the ultrasonic section image, and the accurate standard section image is obtained while the thickness measurement efficiency of the target structure is improved. According to the contour features of the target anatomical structure in the standard section image, the contour of the target anatomical structure is corroded to obtain a skeleton line of the target anatomical structure, each target point on the skeleton line is traversed, the thickness value of the target anatomical structure is determined according to the intersecting point between the normal line passing through the target point and the contour of the target anatomical structure, manual marking of the thickness value is not needed, and manual processing time is saved.

Drawings

FIG. 1 is an application environment diagram of a method of processing an ultrasound image in one embodiment;

FIG. 2 is a flow chart of a method of processing an ultrasound image in one embodiment;

FIG. 3 is a flow chart of a method of generating a standard slice image in one embodiment;

FIG. 4 is a flow chart of a method of measuring growth parameters in one embodiment;

FIG. 5 is a schematic illustration of an anatomical structure in one embodiment;

FIG. 6 is a block diagram of an ultrasound image processing apparatus in one embodiment;

fig. 7 is an internal structural diagram of a computer device in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.

The method for processing the ultrasonic image, provided by the embodiment of the application, can be applied to an application environment shown in fig. 1. Wherein the terminal 102 communicates with the ultrasound probe 104.

The terminal 102 acquires a plurality of ultrasonic section images through the ultrasonic probe 104, wherein each ultrasonic section image comprises a plurality of anatomical structures, the terminal 102 screens out standard section images from each ultrasonic section image according to the scoring result of each anatomical structure in the ultrasonic section images, the terminal 102 performs corrosion treatment on the outline of the target anatomical structure according to the outline characteristics of the target anatomical structure in the standard section images to obtain a skeleton line of the target anatomical structure, the terminal 102 traverses each target point on the skeleton line, and determines the thickness value of the target anatomical structure according to the intersection point between the normal line passing through the target point and the outline of the target anatomical structure.

The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices, and portable wearable devices, where the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart vehicle devices, and the like. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. For example, a terminal to which the ultrasonic detection apparatus corresponds. Specifically, the terminal 102 may store a plurality of ultrasound sectional images in advance, and the terminal 102 may acquire a plurality of ultrasound sectional images through the ultrasound probe 104 in real time. The ultrasonic probe 104 is used to acquire an ultrasonic image of an object to be detected, such as a human body.

In one embodiment, as shown in fig. 2, there is provided a method for processing an ultrasound image, which is described by taking a terminal in fig. 1 as an example, including:

s202, acquiring a plurality of ultrasonic section images, wherein each ultrasonic section image comprises a plurality of anatomical structures.

The ultrasonic image is a display device reflecting the difference of medium acoustic parameters, and can be used for reference by doctors when the doctors detect the patients by using the ultrasonic probes. The ultrasonic section image can be an image obtained by performing a series of processing on the original ultrasonic image, and reflects the original state of the ultrasonic detection target.

The method for acquiring the ultrasonic section image comprises real-time acquisition and pre-acquisition, and specifically, the real-time acquisition of the ultrasonic section image can be in response to the operation of an ultrasonic probe by a doctor. The pre-acquired ultrasound sectional images may be acquired from a database according to the type of ultrasound sectional image. For example, the types of ultrasound sectional images include: head ultrasound sectional images, chest ultrasound sectional images, abdomen ultrasound sectional images, buttock ultrasound sectional images, and the like.

The anatomical structure may be, among other things, the anatomical result of a biological human or other animal body, e.g., an organ, tissue, etc.

The type and number of structures required to form the examination results are not the same for different examination items, and therefore it is desirable to display as many anatomical structures as possible in one ultrasound sectional image. Specifically, before a plurality of ultrasonic section images are acquired, an ultrasonic video stream is acquired, the ultrasonic video stream is extracted to obtain a plurality of ultrasonic video frame images, and the plurality of ultrasonic video frame images are respectively intercepted to obtain a plurality of ultrasonic section images.

S204, screening out standard section images from the ultrasonic section images according to the scoring results of each anatomical structure in the ultrasonic section images.

Wherein the scoring of the anatomical structure is used to describe whether the anatomical structure is clear, whether there is a deletion, whether the shape is normal, etc. Specifically, after the ultrasonic section image is obtained, the target detection model may be used to identify and extract the ultrasonic section image, so as to obtain a plurality of anatomical structures.

Wherein the plurality of anatomical structures may be divided into a major structure and a minor structure according to the extent to which the anatomical structures affect the outcome of the examination item. For the important structures, if at least one important anatomical structure type does not meet the verification requirement, for example, lacks a certain important structure, an abnormal result is directly returned, and if each important anatomical structure type meets the verification requirement, for example, does not lack any important structure, the important structure and the secondary structure are further scored.

Wherein, each anatomy can be scored according to the verification rule matched by the anatomical structure type. It should be noted that, since the anatomical structures are greatly different from each other and the influence on the same anatomical structure caused in different ultrasound examination items is also different, it is necessary to customize the verification rule of each anatomical structure under different examination items. For example, according to the history verification rule matched with the anatomical structure under the current examination item in the verification history library, the verification rule of the anatomical structure is obtained in response to the confirmation operation of the doctor on the history verification rule, and then the corresponding anatomical structure is scored according to the verification rule of the anatomical structure, so as to obtain the scoring result of the anatomical structure.

The inspection items include: it may be to screen for specific structures, such as the posterior cervical skin transparent layer (NT). The examination item comprises a specific structure, other anatomical structures associated with the specific structure.

The standard ultrasound sectional image may be a higher scoring ultrasound sectional image, for example, an ultrasound sectional image having a score exceeding a preset value. Further processing may be performed based on standard ultrasound sectional images.

S206, performing corrosion treatment on the outline of the target anatomical structure according to the outline characteristics of the target anatomical structure in the standard tangent plane image to obtain a skeleton line of the target anatomical structure.

Wherein the target anatomy may be an anatomy preset for the examination item, the features of the target anatomy being directly related to the examination result of the examination item. The target anatomy is of greater importance than the anatomy of the anatomy other than the target anatomy. Taking the examination item as an example of a posterior cervical skin transparent layer item, the target anatomy may be a posterior cervical skin transparent layer structure.

Among them, the erosion process is a morphological process that finely reduces objects in an image by removing pixels from the image boundary. For the item of the transparent layer of the skin behind the neck, the maximum thickness of the transparent layer of the skin behind the neck is key information, so that the outline of the target anatomical structure can be corroded to obtain the skeleton line of the target anatomical structure, then the normal line segments are searched according to the skeleton line, and the maximum thickness of the transparent layer of the skin behind the neck is determined from the normal line segments.

Specifically, a preset structural element can be adopted to perform corrosion treatment on the outline of the target anatomical structure, so as to obtain the skeleton line of the target anatomical structure.

S208, traversing each target point on the skeleton line, and determining the thickness value of the target anatomical structure according to the intersection point between the normal line passing through the target point and the outline of the target anatomical structure.

The skeleton line may be a curve in the target structure, each normal line of the skeleton line is obtained according to an intersection point between a normal line passing through the target point and a contour of the target anatomical structure, and a longest length value of each normal line segment is determined as a thickness value of the target anatomical structure.

Wherein the thickness value of the target anatomy can be sent to a terminal held by a doctor, providing a reference for the doctor. Taking the posterior cervical skin transparency layer as an example, the posterior cervical skin transparency layer may be used to diagnose whether the fetus is a down syndrome or not to provide a reference.

In the processing method of the ultrasonic image, a plurality of ultrasonic section images are acquired, wherein each ultrasonic section image comprises a plurality of anatomical structures, each ultrasonic section image is automatically grabbed, and according to the scoring result of each anatomical structure in each ultrasonic section image, standard section images are screened out from each ultrasonic section image, a doctor does not need to manually store the ultrasonic section images, and the accurate standard section images are obtained while the thickness measurement efficiency of a target structure is improved. According to the contour features of the target anatomical structure in the standard section image, the contour of the target anatomical structure is corroded to obtain a skeleton line of the target anatomical structure, each target point on the skeleton line is traversed, the thickness value of the target anatomical structure is determined according to the intersecting point between the normal line passing through the target point and the contour of the target anatomical structure, manual marking of the thickness value is not needed, and manual processing time is saved.

In one embodiment, the step of screening the standard section image from the ultrasonic section images according to the respective scoring result of each anatomical structure in the ultrasonic section images, as shown in the flow chart of the generating method of the standard section image in fig. 3, includes:

s302, classifying each anatomical structure of the ultrasound sectional images according to the importance of the anatomical structure aiming at each ultrasound sectional image to obtain a first type anatomical structure and a second type anatomical structure, wherein the importance of the first type anatomical structure is greater than that of the second type anatomical structure.

Wherein, for each ultrasound section image, it is necessary to confirm whether the ultrasound section image is a standard section image. Specifically, the ultrasound sectional image may be scored, and the ultrasound sectional image having the score greater than or equal to the preset value is determined as the standard sectional image. And determining the ultrasonic section image with the score smaller than the preset value as a non-standard section image, and judging the next ultrasonic section image.

The importance of the anatomical structure may be the influence degree of the anatomical structure on the ultrasound section image, and the greater the importance of the anatomical structure, the greater the influence degree of the anatomical structure on the ultrasound section image, and vice versa.

Wherein the first type of anatomy may be a major anatomy and the second type of anatomy may be a minor anatomy.

And S304, if the type of the first type of anatomical structure meets the verification requirement, scoring the first type of anatomical structure and the second type of anatomical structure to obtain the total score of the ultrasonic section image.

Wherein the verification requirement may be whether the number of anatomical structures of the first type reaches a preset value, for example, for a posterior cervical skin transparency layer examination item, the ultrasound sectional image includes a plurality of anatomical structures, and the important anatomical structures include: posterior cervical skin transparent layer, nasal tip, anterior nasal skin, maxilla, metaencephalic, mandible, etc. If the important anatomical structures all appear in one ultrasound sectional image, the type of the first type of anatomical structure can be considered to meet the verification requirement. If all important anatomical structures do not appear in one ultrasonic section image, the ultrasonic section image is considered to be abnormal, and an abnormal result is returned. Also taking the examination item as an example of a posterior cervical skin transparent layer, the second type of anatomical structure includes: nasal bone, caltrop, lateral ventricle, etc.

Wherein scoring the first type of anatomy and the second type of anatomy comprises: scoring the first anatomical structure by using a scoring standard customized by the first anatomical structure to obtain a first scoring result, scoring the second anatomical structure by using a scoring standard customized by the second anatomical structure to obtain a second scoring result, and integrating the first scoring result and the first scoring result to obtain the total score of the ultrasonic section image.

S306, determining the ultrasonic section image with the total score being greater than or equal to a preset value as a standard section image.

Wherein the preset value may be 80 minutes.

Specifically, both the first type of scoring result and the second type of scoring result may be a scoring score for a score of deduction, e.g., 5 points if there is one anatomical structure in the first type of anatomical structure or in the second type of anatomical structure that does not meet the scoring criteria. And obtaining the total score of the ultrasonic section according to the initial score and the deduction value of each anatomical structure.

The initial score may be set according to practical situations, for example, the initial score may be set to 95 scores, and if not more than two anatomical structures do not meet the scoring standard, the total score is obtained to be greater than or equal to 80 scores, and the ultrasound section image is determined to be the standard section image.

In this embodiment, for each ultrasound sectional image, each anatomical structure of the ultrasound sectional image is classified according to the importance of the anatomical structure, if the type of the first anatomical structure meets the verification requirement, the first anatomical structure and the second anatomical structure are scored to obtain the total score of the ultrasound sectional image, the ultrasound sectional image with the total score being greater than or equal to the preset value is determined as the standard sectional image, the accuracy of the total score is improved, an accurate standard sectional image is obtained, and a basis is provided for the subsequent determination of the thickness value of the target anatomical structure.

In one embodiment, the step of scoring the first and second types of anatomical structures to obtain a total score for the ultrasound sectional image comprises: weighting the scoring results of each anatomical structure in the first anatomical structure according to the weight coefficient matched with the type of the first anatomical structure to obtain a first scoring result of the first anatomical structure, scoring each anatomical structure in the second anatomical structure according to the scoring standard matched with the type of the second anatomical structure to obtain a second scoring result of the second anatomical structure, and integrating the first scoring result and the second scoring result to obtain the total score of the ultrasound section image.

Wherein the weighting coefficients are used to weight the respective scoring results for each anatomical structure in the first class of anatomical structures. Specifically, the weight coefficient corresponding to each anatomical structure can be determined according to the importance of each anatomical structure in the first type of anatomical structure, and it can be understood that the greater the importance of each anatomical structure, the greater the weight coefficient corresponding to each anatomical structure.

The first scoring result of the first type of anatomy may be an average of scores of corresponding anatomical structures in the first type of anatomy, e.g., three anatomical structures are included in the first type of anatomy, the scores of the three anatomical structures being 90, 80, and 100 points, respectively. The first scoring result for the first type of anatomy is 90 points.

And the scoring standard matched with the type of the second type of anatomical structure is used for scoring each anatomical structure in the second type of anatomical structure to obtain the score of the anatomical structure.

For example, the scoring criteria is "nasal bone clarity is below a threshold," the anatomical structure is a nasal bone, specifically, the anatomical structure is scored according to the scoring criteria, and if nasal bone clarity is below the threshold in the image, the anatomical structure is scored as negative five points.

For another example, if the scoring criteria is "the accuracy of the rhombus is below the threshold, the anatomical structure is scored, and specifically, if the accuracy of the rhombus in the image is below the threshold, the anatomical structure is scored as negative five points.

Wherein the score of the second scoring result of the second class of anatomical structures may be a score of 0 or less. And if the second type of anatomical structure accords with the scoring standard matched with the type of the second type of anatomical structure, the score representing the second scoring result is 0. If a second type of anatomical structure does not meet the scoring criteria, the score representing the second scoring result is less than 0, which may be-5 points, and no multi-buckle 5 points are present.

Specifically, the first scoring result and the second scoring result may be added to obtain a total score of the ultrasound section image

In this embodiment, according to the scoring criteria matched by the type of the second anatomical structure, each anatomical structure in the second anatomical structure is scored to obtain a second scoring result of the second anatomical structure, and the first scoring result and the second scoring result are combined to obtain the total score of the ultrasound section image. And obtaining the total score of the accurate ultrasonic section image, and providing a basis for the subsequent determination of the thickness value of the target anatomical structure.

In one embodiment, the step of corroding the contour of the target anatomy according to the contour features of the target anatomy in the standard tangent plane image to obtain the skeleton line of the target anatomy includes: and extracting the outline of the target anatomical structure according to the outline characteristics of the target anatomical structure in the standard section image to obtain an outline image of the target anatomical structure. And carrying out binarization processing on the contour image to obtain a contour image after the binarization processing. And (3) performing corrosion treatment on the profile image subjected to binarization treatment according to preset structural elements to obtain the skeleton line of the target anatomical structure.

Wherein the contour feature of the target anatomy may be a morphological feature of the target anatomy, in particular the contour feature may comprise: area, perimeter, contour shape, etc. of the target anatomy.

Specifically, a pixel matrix of the target anatomical structure is constructed according to the area feature, the perimeter feature and the shape feature, and then the pixel matrix of the target anatomical structure is converted into a contour image of the target anatomical structure.

The binarization processing means that the gray value of the pixel point on the contour image is set to be 0 or 255, and the contour image after the binarization processing is convenient for corrosion processing.

The erosion processing is a morphological processing mode, and the erosion processing finely reduces objects in the image by removing pixels from the image boundary, specifically, the erosion processing is performed on the contour image after the binarization processing by adopting preset structural elements, and the processing process is similar to a convolution kernel, and the image is subjected to convolution operation.

The preset structural element may be a continuous pixel point in the image.

Specifically, traversing the edge of the contour image after binarization processing by adopting a preset structural element to obtain the skeleton line of the target anatomical structure.

In this embodiment, according to the contour features of the target anatomical structure in the standard section image, the contour of the target anatomical structure is extracted to obtain a contour image of the target anatomical structure, the contour image is subjected to binarization processing to obtain a binarized contour image, and according to the preset structural elements, the binarized contour image is subjected to corrosion processing to obtain a skeleton line of the target anatomical structure, so that an accurate skeleton line is obtained, and a basis is provided for the subsequent determination of the thickness value of the target anatomical structure.

In one embodiment, the step of traversing each target point on the skeleton line and determining a thickness value of the target anatomy from an intersection point between a normal passing through the target point and a contour of the target anatomy comprises: traversing each target point on the skeleton line, and determining each normal line segment according to the intersection point between the normal line passing through the target point and the outline of the target structure. And screening each normal line segment according to the length sequencing result of the normal line segments to obtain each initial line segment. And adjusting each initial line segment according to the pixel value at the end point of the initial line segment to obtain each target line segment corresponding to each initial line segment. And determining the thickness value of the target structure according to the length sequencing result of each target line segment.

The skeleton line can be regarded as a curve, a tangent line passing through the target point is determined according to the target point on the skeleton line and the slope of the target point on the skeleton line, then a normal line perpendicular to the tangent line passing through the target point is obtained according to the tangent line, and each normal line segment is determined through an intersection point between the normal line of the target point and the contour of the target structure.

Wherein the length corresponding to the normal line segment characterizes the thickness value of the target anatomy. And screening the normal line segments to obtain the longest normal line segment, and determining the longest normal line segment length as the thickness value of the target structure.

Specifically, taking the target anatomical structure as the transparent layer structure of the skin behind the neck as an example, the length of the line segment can be redetermined to improve the accuracy of the thickness value of the target anatomical structure due to the fact that the edge of the transparent layer structure of the skin behind the neck may be blurred by the image processing means.

Wherein, each normal line segment can be screened to obtain each initial line segment, specifically, a preset number of initial line segments can be screened to obtain 10 initial line segments, for example.

After each initial line segment is obtained, each initial line segment is adjusted according to the pixel value at the end point of the initial line segment, and a target line segment corresponding to each initial line segment is obtained, for example, 10 target line segments are obtained. The difference between the target line segment and the initial line segment is that the lengths may be different, for example, a certain number (may be 10) of pixel bands are selected in the extending direction and the returning direction of the initial line segment according to the pixel value and the end point at the end point of the initial line segment, the real end point is recalculated, and the initial line segment is obtained according to the real end point.

It should be noted that, the real endpoint may be a pixel point with the largest gray value transformation, for example, if for a certain initial line segment, the gray distribution is: for example, "000788", the gray value corresponding to the third pixel is 0, and the pixel is the point with the maximum gray value conversion, and can be determined as the real endpoint.

Specifically, the two ends of the initial line segment are adjusted according to the method, and the target line segment corresponding to each initial line segment is obtained. The target line segment may be larger than the initial line segment, may be smaller than the initial line segment, or may be equal to the initial line segment, depending on the gray distribution of the initial line segment in the extending direction. And determining the thickness value of the target structure according to the length sequencing result of each target line segment.

In this embodiment, each target point on the skeleton line is traversed, each normal line segment is determined according to the intersection point between the normal line passing through the target point and the contour of the target structure, each normal line segment is screened according to the length sequencing result of the normal line segments, each initial line segment is obtained, the normal line segments with larger numbers are screened, a preset number of initial line segments are obtained, the calculated amount is reduced, each initial line segment is adjusted according to the pixel value at the end point of the initial line segment, each target line segment corresponding to each initial line segment is obtained, the length of each initial line segment is redetermined, and the accuracy of the thickness value of the target structure generated later is improved.

In one embodiment, the step of acquiring a plurality of ultrasound slice images comprises: and acquiring an ultrasonic video stream. And extracting the ultrasonic video stream to obtain a plurality of ultrasonic video frame images. And respectively intercepting each ultrasonic video frame image according to the image annotation type of the ultrasonic video frame image to obtain a plurality of ultrasonic section images.

The ultrasonic video stream may be a video stream directly acquired by an ultrasonic probe. Or may be a pre-stored video stream. The ultrasound video stream includes a plurality of ultrasound video frame images.

The image of the ultrasonic video frame can be a specific image, the image can comprise a plurality of structures, and in order to avoid mutual influence of the structures in the image recognition process, the image of the ultrasonic video frame needs to be intercepted, so that an ultrasonic section image is obtained.

Specifically, each ultrasonic video frame image can be intercepted according to the image annotation type of the ultrasonic video frame image. The image annotation type may be an annotation type of each structure in the image of the ultrasonic video frame, for example, for the fetal ultrasonic image, the head structure part of the image, the chest structure part of the image, the abdomen structure part of the image, the buttock structure part of the image, and the like may be truncated.

It can be understood that according to different examination items, different structures can be intercepted in the ultrasonic video frame image to obtain an ultrasonic section image. The ultrasound sectional image includes a primary structural portion and a secondary structural portion. For example, an ultrasound sectional image corresponding to the posterior cervical skin transparent layer includes a head structure portion and a chest structure portion, where the primary structure is the head and the chest is the secondary structure.

In this embodiment, an ultrasound video stream is acquired. And extracting the ultrasonic video stream to obtain a plurality of ultrasonic video frame images. According to the image annotation type of the ultrasonic video frame images, each ultrasonic video frame image is intercepted respectively to obtain a plurality of ultrasonic section images, a plurality of ultrasonic section images are obtained, and the accuracy of the thickness value of the target structure generated later is improved.

In one embodiment, as shown in fig. 4, there is provided a method for measuring a growth parameter, comprising:

s402, acquiring an ultrasonic video stream.

S404, extracting the ultrasonic video stream to obtain a plurality of ultrasonic video frame images.

S406, according to the image annotation type of the ultrasonic video frame images, each ultrasonic video frame image is intercepted respectively, and a plurality of ultrasonic section images are obtained.

The ultrasonic video frame image can be a head-hip length CRL section, the ultrasonic section image can be a NT section, and labels respectively corresponding to the head, the chest, the abdomen and the buttocks are added in the head-hip length CRL section. Specifically, the head and hip length CRL section is cut, and the label of the head can be selected as an ultrasonic section image, and the label of the head and the chest can be also selected as an ultrasonic section image.

Wherein each ultrasound sectional image includes a plurality of anatomical structures.

Specifically, the target recognition can be performed on the ultrasonic video frame image, so that the number, type, position and other characteristics of the anatomical structures in the ultrasonic section image are obtained. Wherein, target recognition includes: performing target detection training through a YOLOv8 neural network to obtain the structural positioning capability of the NT section, and specifically, performing target detection on the super section image by adopting the YOLOv8 neural network to obtain an anatomical structure and a target detection frame of the anatomical structure.

After the target recognition is performed on the ultrasonic video frame image, the anatomical structure can be extracted from the ultrasonic section image, so that the outline of the anatomical structure can be obtained. Specifically, a target detection frame of the anatomical structure and the anatomical structure is identified based on a Unet semantic segmentation model of the MobileNet, so that the outline of the anatomical structure is obtained.

S408, classifying each anatomical structure of the ultrasound sectional images according to the importance of the anatomical structure aiming at each ultrasound sectional image to obtain a first type anatomical structure and a second type anatomical structure.

Wherein the first type of anatomy is of greater importance than the second type of anatomy.

Wherein, as shown in the schematic view of the anatomical structure of fig. 5, the first type of structural structure may comprise: : the posterior cervical skin transparency layer, nasal tip, anterior nasal skin, maxilla, metaencephalic, mandible, etc., the first type of structural structure may include: nasal bone, rhombic brain, lateral ventricle.

S410, if the type of the first type of anatomical structure meets the verification requirement, weighting the scoring result of each anatomical structure in the first type of anatomical structure according to the weight coefficient matched with the type of the first type of anatomical structure to obtain a first scoring result of the first type of anatomical structure.

And S412, assigning scores to the anatomical structures in the second type of anatomical structures according to the scoring criteria matched with the type of the second type of anatomical structures, so as to obtain a second scoring result of the second type of anatomical structures.

And S414, combining the first scoring result and the second scoring result to obtain the total score of the ultrasonic section image.

S416, determining the ultrasonic section image with the total score being greater than or equal to a preset value as a standard section image.

S418, extracting the outline of the target anatomical structure according to the outline features of the target anatomical structure in the standard section image to obtain an outline image of the target anatomical structure.

S420, performing binarization processing on the contour image to obtain a binarized contour image.

And S422, performing corrosion treatment on the profile image subjected to binarization treatment according to the preset structural elements to obtain the skeleton line of the target anatomical structure.

S424, traversing each target point on the skeleton line, and determining each normal line segment according to the intersection point between the normal line passing through the target point and the outline of the target structure.

S426, screening each normal line segment according to the length sequencing result of the normal line segments to obtain each initial line segment.

And S428, adjusting each initial line segment according to the pixel value at the end point of the initial line segment to obtain each target line segment corresponding to each initial line segment.

S430, determining the thickness value of the target structure according to the length sequencing result of each target line segment.

Wherein the growth parameter may be a thickness value of the target structure.

In this embodiment, a plurality of ultrasound section images are acquired, where each ultrasound section image includes a plurality of anatomical structures, each ultrasound section image is automatically captured, and according to the scoring result of each anatomical structure in the ultrasound section images, a standard section image is screened out from each ultrasound section image, so that a doctor does not need to manually save the ultrasound section images, and an accurate standard section image is obtained while the thickness measurement efficiency of a target structure is improved. According to the contour features of the target anatomical structure in the standard section image, the contour of the target anatomical structure is corroded to obtain a skeleton line of the target anatomical structure, each target point on the skeleton line is traversed, the thickness value of the target anatomical structure is determined according to the intersecting point between the normal line passing through the target point and the contour of the target anatomical structure, manual marking of the thickness value is not needed, and manual processing time is saved. According to the method, the fetal NT section structure detection and the key structure segmentation are carried out by the basis deep learning, and the automatic grabbing of the ultrasonic NT section and the maximum thickness measurement of the NT structure can be realized. The method adopts the YOLOv8 model as a target detector, so that the position of the NT structure can be rapidly and accurately detected; the Unet model based on MobileNet is adopted as the NT structure divider, so that the outline of the NT structure can be obtained rapidly and accurately; the skeleton line of the NT structure can be effectively calculated by adopting the NT structure maximum measurement technology based on the skeleton line algorithm, and the maximum thickness is calculated by utilizing the skeleton line; based on actual conditions, the labels of the CRL sections are added, and the NT sections and the CRL sections are accurately distinguished. And a fine tuning algorithm is adopted to improve the accuracy of NT maximum thickness measurement.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the application also provides an ultrasonic image processing device for realizing the above related ultrasonic image processing method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation in the embodiments of the device for processing one or more ultrasound images provided below may be referred to the limitation of the method for processing ultrasound images hereinabove, and will not be described herein.

In one embodiment, as shown in fig. 6, there is provided an ultrasound image processing apparatus including: an image acquisition module 602, an image screening module 604, a processing module 606, and an output module 608, wherein:

an image acquisition module 602 that acquires a plurality of ultrasound slice images, wherein each ultrasound slice image includes a plurality of anatomical structures;

the image screening module 604 is configured to screen a standard section image from the ultrasound section images according to the respective scoring result of each anatomical structure in the ultrasound section images;

the processing module 606 is configured to perform corrosion processing on the contour of the target anatomical structure according to the contour feature of the target anatomical structure in the standard tangent plane image, so as to obtain a skeleton line of the target anatomical structure;

an output module 608 for traversing each target point on the skeleton line and determining a thickness value of the target anatomy from an intersection point between a normal line passing through the target point and a contour of the target anatomy.

In one embodiment, the image screening module 604 is further configured to classify, for each ultrasound sectional image, each anatomical structure of the ultrasound sectional image according to the importance of the anatomical structure, to obtain a first type anatomical structure and a second type anatomical structure, where the importance of the first type anatomical structure is greater than the importance of the second type anatomical structure; if the type of the first type of anatomical structure meets the verification requirement, scoring the first type of anatomical structure and the second type of anatomical structure to obtain the total score of the ultrasonic section image; and determining the ultrasonic section image with the total score being greater than or equal to a preset value as a standard section image.

In one embodiment, the image screening module 604 is further configured to weight the scoring result of each anatomical structure in the first type of anatomical structure according to the weight coefficient matched by the type of the first type of anatomical structure, so as to obtain a first scoring result of the first type of anatomical structure; assigning scores to the anatomical structures in the second class according to scoring criteria matched with the types of the anatomical structures in the second class to obtain a second scoring result of the anatomical structures in the second class; and combining the first scoring result and the second scoring result to obtain the total score of the ultrasonic section image.

In one embodiment, the processing module 606 is further configured to extract a contour of the target anatomical structure according to the contour features of the target anatomical structure in the standard tangent plane image, to obtain a contour image of the target anatomical structure; performing binarization processing on the contour image to obtain a binarized contour image; and (3) performing corrosion treatment on the profile image subjected to binarization treatment according to preset structural elements to obtain the skeleton line of the target anatomical structure.

In one embodiment, the output module 608 is further configured to traverse each target point on the skeleton line, and determine each normal line segment according to an intersection point between a normal line passing through the target point and a contour of the target structure; screening each normal line segment according to the length sequencing result of the normal line segments to obtain each initial line segment; according to the pixel value at the end point of the initial line segment, adjusting each initial line segment to obtain each target line segment corresponding to each initial line segment; and determining the thickness value of the target structure according to the length sequencing result of each target line segment.

In one embodiment, the image acquisition module 602 is further configured to acquire an ultrasound video stream; extracting an ultrasonic video stream to obtain a plurality of ultrasonic video frame images; and respectively intercepting each ultrasonic video frame image according to the image annotation type of the ultrasonic video frame image to obtain a plurality of ultrasonic section images.

The respective modules in the above-described processing apparatus for an ultrasound image may be implemented in whole or in part by software, hardware, and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing ultrasound slice image data. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of processing an ultrasound image.

It will be appreciated by those skilled in the art that the structure shown in fig. 7 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:

acquiring a plurality of ultrasound sectional images, wherein each ultrasound sectional image comprises a plurality of anatomical structures; screening out a standard section image from each ultrasonic section image according to the scoring result of each anatomical structure in the ultrasonic section image; according to the contour features of the target anatomical structure in the standard section image, performing corrosion treatment on the contour of the target anatomical structure to obtain a skeleton line of the target anatomical structure; each target point on the skeleton line is traversed and a thickness value of the target anatomy is determined from an intersection point between a normal passing through the target point and a contour of the target anatomy.

In one embodiment, the processor when executing the computer program further performs the steps of:

classifying each anatomical structure of the ultrasound sectional images according to the importance of the anatomical structure aiming at each ultrasound sectional image to obtain a first type anatomical structure and a second type anatomical structure, wherein the importance of the first type anatomical structure is greater than that of the second type anatomical structure; if the type of the first type of anatomical structure meets the verification requirement, scoring the first type of anatomical structure and the second type of anatomical structure to obtain the total score of the ultrasonic section image; and determining the ultrasonic section image with the total score being greater than or equal to a preset value as a standard section image.

weighting the scoring results of each anatomical structure in the first type of anatomical structure according to the weight coefficient matched with the type of the first type of anatomical structure to obtain a first scoring result of the first type of anatomical structure; assigning scores to the anatomical structures in the second class according to scoring criteria matched with the types of the anatomical structures in the second class to obtain a second scoring result of the anatomical structures in the second class; and combining the first scoring result and the second scoring result to obtain the total score of the ultrasonic section image.

extracting the outline of the target anatomical structure according to the outline features of the target anatomical structure in the standard section image to obtain an outline image of the target anatomical structure; performing binarization processing on the contour image to obtain a binarized contour image; and (3) performing corrosion treatment on the profile image subjected to binarization treatment according to preset structural elements to obtain the skeleton line of the target anatomical structure.

traversing each target point on the skeleton line, and determining each normal line segment according to the intersection point between the normal line passing through the target point and the outline of the target structure; screening each normal line segment according to the length sequencing result of the normal line segments to obtain each initial line segment; according to the pixel value at the end point of the initial line segment, adjusting each initial line segment to obtain each target line segment corresponding to each initial line segment; and determining the thickness value of the target structure according to the length sequencing result of each target line segment.

Acquiring an ultrasonic video stream; extracting an ultrasonic video stream to obtain a plurality of ultrasonic video frame images; and respectively intercepting each ultrasonic video frame image according to the image annotation type of the ultrasonic video frame image to obtain a plurality of ultrasonic section images.

In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, the computer program when executed by the processor further performs the steps of:

In one embodiment, a computer program product is provided comprising a computer program which, when executed by a processor, performs the steps of:

It should be noted that, the user information (including, but not limited to, user equipment information, user personal information, etc.) and the data (including, but not limited to, data for analysis, stored data, presented data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party, and the collection, use, and processing of the related data are required to meet the related regulations.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the various embodiments provided herein may include at least one of relational databases and non-relational databases. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc., without being limited thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The above examples only represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the present application. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application shall be subject to the appended claims.

Claims

1. A method of processing an ultrasound image, the method comprising:

acquiring a plurality of ultrasound section images, wherein each ultrasound section image comprises a plurality of anatomical structures;

screening out a standard section image from each ultrasonic section image according to the scoring result of each anatomical structure in the ultrasonic section image;

according to the contour features of the target anatomical structure in the standard section image, performing corrosion treatment on the contour of the target anatomical structure to obtain a skeleton line of the target anatomical structure;

Traversing each target point on the skeleton line and determining a thickness value of the target anatomy from an intersection point between a normal line passing through the target point and a contour of the target anatomy.

2. The method of claim 1, wherein the step of screening out a standard section image from each of the ultrasound section images according to a respective scoring result for each of the anatomical structures in the ultrasound section images comprises:

classifying each anatomical structure of the ultrasound sectional images according to the importance of the anatomical structure aiming at each ultrasound sectional image to obtain a first type anatomical structure and a second type anatomical structure, wherein the importance of the first type anatomical structure is greater than that of the second type anatomical structure;

if the type of the first type anatomical structure meets the verification requirement, scoring the first type anatomical structure and the second type anatomical structure to obtain the total score of the ultrasonic section image;

and determining the ultrasonic section image with the total score being greater than or equal to a preset value as a standard section image.

3. The method of claim 2, wherein scoring the first type of anatomy and the second type of anatomy to obtain a total score for the ultrasound sectional image comprises:

Weighting the scoring results of each anatomical structure in the first type of anatomical structure according to the weight coefficient matched with the type of the first type of anatomical structure to obtain a first scoring result of the first type of anatomical structure;

assigning scores to the anatomical structures in the second class according to scoring criteria matched with the type of the anatomical structure in the second class to obtain a second scoring result of the anatomical structure in the second class;

and combining the first scoring result and the second scoring result to obtain the total score of the ultrasonic section image.

4. The method according to claim 1, wherein the step of corroding the contour of the target anatomy according to the contour features of the target anatomy in the standard tangent plane image to obtain the skeleton line of the target anatomy comprises the steps of:

extracting the outline of the target anatomical structure according to the outline characteristics of the target anatomical structure in the standard section image to obtain an outline image of the target anatomical structure;

performing binarization processing on the contour image to obtain a binarized contour image;

and performing corrosion treatment on the profile image subjected to the binarization treatment according to preset structural elements to obtain the skeleton line of the target anatomical structure.

5. The method according to claim 1, wherein the step of traversing each target point on the skeleton line and determining a thickness value of the target anatomy from an intersection between a normal passing through the target point and a contour of the target anatomy comprises:

traversing each target point on the skeleton line, and determining each normal line segment according to an intersection point between a normal line passing through the target point and the outline of the target structure;

screening each normal line segment according to the length sequencing result of the normal line segments to obtain each initial line segment;

according to the pixel values at the end points of the initial line segments, adjusting each initial line segment to obtain each target line segment corresponding to each initial line segment;

and determining the thickness value of the target structure according to the length sequencing result of each target line segment.

6. The method of claim 1, wherein the step of acquiring a plurality of ultrasound slice images comprises:

acquiring an ultrasonic video stream;

extracting the ultrasonic video stream to obtain a plurality of ultrasonic video frame images;

and respectively intercepting each ultrasonic video frame image according to the image annotation type of the ultrasonic video frame image to obtain a plurality of ultrasonic section images.

7. An ultrasound image processing apparatus, the apparatus comprising:

the image acquisition module acquires a plurality of ultrasonic section images, wherein each ultrasonic section image comprises a plurality of anatomical structures;

the image screening module is used for screening standard section images from the ultrasonic section images according to the scoring results of each anatomical structure in the ultrasonic section images;

the processing module is used for corroding the outline of the target anatomical structure according to the outline characteristics of the target anatomical structure in the standard section image to obtain a skeleton line of the target anatomical structure;

and the output module is used for traversing each target point on the skeleton line and determining the thickness value of the target anatomical structure according to the intersection point between the normal line passing through the target point and the outline of the target anatomical structure.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 6 when the computer program is executed.

9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.