Method and system for processing seroperitoneum image

Abstract

The invention relates to the technical field of image processing, and particularly discloses a method and system for processing a seroperitoneum image, and the method comprises the steps: S1, selecting a scanning region, planning a scanning path, obtaining an ultrasonic image set of the scanning region in real time through a color ultrasound instrument probe, and building an ultrasonic three-dimensional model; performing three-dimensional grid division to obtain a plurality of three-dimensional grids, and identifying a hydrops region and an organ region according to a gray value threshold range; s2, setting a segmentation plane to divide the ultrasonic three-dimensional model into an outer region and an inner region, and obtaining a cross section; vertical lines are obtained on the cross section, and every two adjacent vertical lines form a square column; acquiring a boundary line of the outermost periphery and all boundaries, selecting a boundary line closest to the boundary line, and calculating the thickness of the surface layer; selecting a square column with the minimum surface thickness, and obtaining the midpoint of the boundary line of the square column as a puncture point; and S3, sequentially selecting each point under the puncture point to obtain a puncture path.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method and system for processing peritoneal effusion images. Background Art

[0002] The treatment process of ascites includes the extraction of ascites. In clinical practice, a reasonable pressure value needs to be controlled when aspirating ascites. If the pressure value is too low, it may lead to incomplete aspiration and excessive residual fluid, affecting the treatment effect. If the pressure value is too high, it may cause damage to surrounding tissues, increasing the patient's pain and the risk of complications.

[0003] In the prior art, when aspirating peritoneal effusion clinically, the aspiration is often performed based on experience. Although experience is an important basis for doctors to diagnose and treat, it still has some limitations. For example, it is difficult to ensure that each aspiration can accurately locate the effusion area based on experience alone, especially when the amount of effusion is small or unevenly distributed. This may lead to incomplete aspiration, residual effusion, or accidental injury to surrounding tissues. For example, the specific circumstances of each patient are different, including the cause, nature, amount and physical condition of the effusion. It is difficult to fully take into account these individual differences based on experience alone, so as to develop a personalized treatment plan that best suits each patient. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for processing peritoneal effusion images to solve the above technical problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions: A method for processing an image of peritoneal effusion comprises the following steps: Step S1: selecting a scanning area on the patient's abdomen, planning a scanning path, acquiring an ultrasonic image of the scanning area in real time on the scanning path through a color ultrasound probe to obtain an ultrasonic image set; and establishing an ultrasonic three-dimensional model of the scanning area according to the ultrasonic image set; Performing three-dimensional grid division in the ultrasound three-dimensional model to obtain a plurality of three-dimensional grids; obtaining a gray value R of each three-dimensional grid, setting a gray value threshold range Rt, selecting all three-dimensional grids satisfying R∈Rt in the ultrasound three-dimensional model, recording them as low-grayscale three-dimensional grids, recording an area formed by all low-grayscale three-dimensional grids as an effusion area, and recording an area formed by all remaining three-dimensional grids as an organ area; Step S2: setting a segmentation plane, dividing the ultrasonic three-dimensional model into an outer area and an inner area, and obtaining a cross section of the outer area once every square side length, where the square side length is the side length of the three-dimensional square; obtaining a vertical line every square side length on the cross section, and every two adjacent vertical lines constitute a square column; In the square column, the outermost boundary line is obtained, and all boundary lines in the square column are obtained, and the boundary line is the boundary line between the effusion area and the organ area; the boundary line closest to the boundary line is selected, recorded as L, and the distance between the boundary line and L is obtained, recorded as the surface thickness; the square column with the smallest surface thickness value is selected, and the midpoint of the boundary line of the square column is obtained, recorded as the puncture point Q; Step S3: Make a reference plane S through the puncture point parallel to the segmentation plane, set the interval threshold It, and select a point P below the reference plane. The selection of point P satisfies , where |PQ| represents the distance between the puncture point Q and point P, and J represents the effusion area; according to the above steps, a new reference plane is made through point P, and point P1 is selected, and so on, until the latest point reaches the inner area; the puncture point and all the selected points are connected to obtain the puncture path, and the puncture path is displayed.

[0006] As a further solution of the present invention: the scanning path planning process includes: The contact surface between the color ultrasound probe and the patient's abdomen is obtained, and the contact surface is recorded as a rectangle. The scanning area is grid-divided according to the rectangle to obtain a plurality of grids. Each grid is numbered in turn, and the center point of each grid is obtained. The center points are linked according to the number to obtain a scanning path.

[0007] As a further solution of the present invention: the process of establishing the ultrasonic three-dimensional model includes: An edge detection algorithm is used to perform edge detection on the ultrasonic image to identify the edge and contour of the ultrasonic image, and feature point detection is performed on the ultrasonic image to detect key points or feature points in the ultrasonic image; the ultrasonic image is registered and the registered ultrasonic images are stacked to form three-dimensional volume data, a three-dimensional view is generated according to the three-dimensional volume data using volume rendering technology, and isosurfaces are extracted from the three-dimensional volume data to generate a surface model of an organ or tissue, and finally an ultrasonic three-dimensional model is obtained.

[0008] As a further solution of the present invention: the process of setting the gray value threshold range includes: Acquire several ultrasonic image samples of peritoneal effusion, and manually mark the effusion area on the ultrasonic image samples; acquire the grayscale value of the effusion area, and acquire the maximum and minimum values ​​of all grayscale values, and obtain the grayscale value threshold range according to the maximum and minimum values.

[0009] As a further solution of the present invention: the process of setting the splitting surface includes: An initial spherical region is set, the radius of the initial spherical region is recorded as an initial radius r0, and a radius interval threshold rt is set, the initial radius is incremented according to the radius interval threshold, and starting from the initial radius, a region radius is obtained every radius interval threshold; When the radius is the radius of each area, we get several spherical areas {Area0, Area1, ..., Area n}, where Area0 is the initial spherical area, Area n is the spherical area corresponding to the radius of the nth area; obtain the effusion proportion Poa=V0 / V of each spherical area, where V0 is the sum of the volumes of all three-dimensional squares occupied by the intersection of the spherical area and the effusion area, and V is the total volume of the spherical area; select the effusion proportion that satisfies The spherical area is called the inner area, where Poa i Indicates Area i The corresponding effusion ratio, i∈[0,n] and i is a positive integer, Poa´ is a preset ratio threshold; in the ultrasonic three-dimensional model, the area other than the inner area is the outer area, and the interface between the outer area and the inner area is obtained and recorded as the segmentation surface.

[0010] As a further solution of the present invention: the process of determining the part below the reference plane includes: obtaining two sides of the reference plane, and recording the side where the inner area is located as the part below the reference plane.

[0011] As a further solution of the present invention: the process of obtaining the puncture path includes: When there are several puncture paths, the length of each puncture path is obtained, and the puncture path with the shortest length is selected as the puncture path finally displayed.

[0012] As a further solution of the present invention: a peritoneal effusion image processing system, comprising: Region division module: selects a scanning area on the patient's abdomen, plans a scanning path, obtains an ultrasonic image of the scanning area in real time on the scanning path through a color ultrasound probe, and obtains an ultrasonic image set; establishes an ultrasonic three-dimensional model of the scanning area based on the ultrasonic image set; Performing three-dimensional grid division in the ultrasound three-dimensional model to obtain a plurality of three-dimensional grids; obtaining a gray value R of each three-dimensional grid, setting a gray value threshold range Rt, selecting all three-dimensional grids satisfying R∈Rt in the ultrasound three-dimensional model, recording them as low-grayscale three-dimensional grids, recording an area formed by all low-grayscale three-dimensional grids as an effusion area, and recording an area formed by all remaining three-dimensional grids as an organ area; Puncture point determination module: setting a segmentation plane, dividing the ultrasonic three-dimensional model into an outer area and an inner area, and obtaining a cross section of the outer area once every square side length, where the square side length is the side length of the three-dimensional square; obtaining a vertical line every square side length on the cross section, and every two adjacent vertical lines constitute a square column; In the square column, the outermost boundary line is obtained, and all boundary lines in the square column are obtained, and the boundary line is the boundary line between the effusion area and the organ area; the boundary line closest to the boundary line is selected, recorded as L, and the distance between the boundary line and L is obtained, recorded as the surface thickness; the square column with the smallest surface thickness value is selected, and the midpoint of the boundary line of the square column is obtained, recorded as the puncture point Q; Puncture path determination module: Make a reference plane S parallel to the segmentation plane through the puncture point, set the interval threshold It, select a point P below the reference plane, and the selection of point P satisfies , where |PQ| represents the distance between the puncture point Q and point P, and J represents the effusion area; according to the above steps, a new reference plane is made through point P, and point P1 is selected, and so on, until the latest point reaches the inner area; the puncture point and all the selected points are connected to obtain the puncture path, and the puncture path is displayed.

[0013] Beneficial effects of the present invention: The present invention can more accurately locate the position and range of peritoneal effusion by acquiring ultrasonic images in real time and establishing ultrasonic three-dimensional models; this method provides more spatial information than traditional two-dimensional ultrasonic images, which helps doctors to judge the condition more accurately; by setting the segmentation surface, calculating the surface thickness and determining the puncture point Q, an optimal puncture path can be planned; this not only reduces the harm to the patient, but also improves the safety and success rate of the operation; through accurate puncture path planning, damage to surrounding important organs during the puncture process can be avoided, thereby reducing the risk of complications; this is of great significance for improving the patient's recovery speed and quality of life; by quickly and accurately determining the puncture point and puncture path, doctors can extract or inject peritoneal effusion more quickly; this can not only shorten the operation time, but also improve the overall treatment efficiency; the puncture path obtained by connecting the puncture point and all selected points can be intuitively displayed to doctors and patients; this visualization effect helps doctors better understand the treatment plan and explain the treatment process and expected results to patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The present invention will be further described below in conjunction with the accompanying drawings.

[0015] Figure 1 It is a flow chart of a method for processing peritoneal effusion images according to the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] See also Figure 1 As shown, the present invention is a method for processing peritoneal effusion images, comprising the following steps: Step S1: selecting a scanning area on the patient's abdomen, planning a scanning path, acquiring an ultrasonic image of the scanning area in real time on the scanning path through a color ultrasound probe to obtain an ultrasonic image set; and establishing an ultrasonic three-dimensional model of the scanning area according to the ultrasonic image set; Performing three-dimensional grid division in the ultrasound three-dimensional model to obtain a plurality of three-dimensional grids; obtaining a gray value R of each three-dimensional grid, setting a gray value threshold range Rt, selecting all three-dimensional grids satisfying R∈Rt in the ultrasound three-dimensional model, recording them as low-grayscale three-dimensional grids, recording an area formed by all low-grayscale three-dimensional grids as an effusion area, and recording an area formed by all remaining three-dimensional grids as an organ area; It can be understood that the color ultrasound probe is moved over the scanning area of ​​the patient's abdomen to capture ultrasound images of the area in real time; these images are collected and form an ultrasound image set; using the collected ultrasound image set, a three-dimensional ultrasound model is constructed through specific algorithms and technologies; this model represents the tissues and structures in the scanning area; the ultrasound image set is a series of two-dimensional ultrasound images, reflecting the internal structural information of the human body; As a preferred embodiment of the present invention, the scanning path planning process includes: The contact surface between the color ultrasound probe and the patient's abdomen is obtained, and the contact surface is recorded as a rectangle, and the scanning area is grid-divided according to the rectangle to obtain a plurality of grids; each grid is numbered in turn, and the center point of each grid is obtained, and each center point is linked according to the number to obtain a scanning path; It can be understood that a contact surface is determined by contacting the color ultrasound probe with the patient's abdomen; the contact surface is defined as a rectangular area; the divided grids are numbered, and the center point of each grid is determined; the numbering helps to track and manage each grid, and the center point is a key reference point for planning the scanning path; the center points are connected in sequence according to the numbering sequence of the grids; thus, by connecting these center points in sequence, a complete scanning path is formed, and the scanning path is generally S-shaped; As a preferred embodiment of the present invention, the process of establishing the ultrasonic three-dimensional model includes: Perform edge detection on the ultrasonic image using an edge detection algorithm to identify the edge and contour of the ultrasonic image, and perform feature point detection on the ultrasonic image to detect key points or feature points in the ultrasonic image; perform registration on the ultrasonic image, and stack the registered ultrasonic images to form three-dimensional volume data, generate a three-dimensional view based on the three-dimensional volume data using a volume rendering technology, and extract an isosurface from the three-dimensional volume data to generate a surface model of an organ or tissue, and finally obtain an ultrasonic three-dimensional model; It should be noted that when constructing an ultrasonic three-dimensional model through the ultrasonic image set, the image is first preprocessed, and the preprocessing process includes denoising, enhancement and segmentation; denoising: since the ultrasonic image may contain noise, denoising processing is first required to improve the image quality; enhancement: applying image enhancement technology, such as histogram equalization or filtering, to improve the contrast and details of the image; segmentation: if necessary, the image is segmented to separate the area of ​​interest (such as organs, tumors, etc.); then feature extraction is performed on the ultrasonic image, and the feature extraction process includes edge detection and feature point detection, and the edge detection is based on the Canny edge detector; after the feature extraction is completed, the ultrasonic image is The ultrasonic image is registered, and the image registration process includes rigid registration and non-rigid registration. If the relative position between the images does not change much, the rigid registration method can be used to align the images. For the case of large deformation, non-rigid registration technology, such as a method based on a free deformation model, needs to be used. After the image registration is completed, the ultrasonic image is reconstructed in three dimensions, including volume rendering and surface reconstruction. The volume rendering is to stack the registered two-dimensional images to form three-dimensional volume data. Then, the volume rendering technology (such as ray casting) is applied to generate a three-dimensional view. The surface reconstruction is to extract isosurfaces from the volume data to generate a surface model of an organ or tissue. Commonly used algorithms include the marching cubes algorithm. After the three-dimensional reconstruction, an initial model is obtained, and the initial model is optimized. The process includes smoothing and refinement. The smoothing is to smooth the reconstructed three-dimensional model to reduce noise and artifacts, and the refinement is to refine the model as needed to improve its resolution and accuracy. As a preferred embodiment of the present invention, the process of setting the gray value threshold range includes: Acquire a number of ultrasonic image samples of peritoneal effusion, and manually mark the effusion area on the ultrasonic image samples; acquire the gray value of the effusion area, and acquire the maximum and minimum values ​​of all gray values, and obtain the gray value threshold range according to the maximum and minimum values; It is understandable that a number of ultrasound image samples of peritoneal effusion are collected, and these samples will be used to analyze and determine the grayscale value characteristics of the effusion area; these ultrasound image samples are manually annotated, that is, the location of the effusion area is clearly marked on the image; this step is to ensure the accuracy of subsequent analysis, because manual annotation can provide an accurate reference standard; grayscale values ​​are extracted from the annotated effusion area; grayscale value is an indicator of image brightness, usually used to describe the color depth of pixels in the image; here, it is used to distinguish different tissue types; in the process of calculating the maximum and minimum grayscale values ​​of all effusion areas, these two values ​​represent the range of grayscale values ​​of the effusion area; Step S2: setting a segmentation plane, dividing the ultrasonic three-dimensional model into an outer area and an inner area, and obtaining a cross section of the outer area once every square side length, where the square side length is the side length of the three-dimensional square; obtaining a vertical line every square side length on the cross section, and every two adjacent vertical lines constitute a square column; In the square column, the outermost boundary line is obtained, and all boundary lines in the square column are obtained, and the boundary line is the boundary line between the effusion area and the organ area; the boundary line closest to the boundary line is selected, recorded as L, and the distance between the boundary line and L is obtained, recorded as the surface thickness; the square column with the smallest surface thickness value is selected, and the midpoint of the boundary line of the square column is obtained, recorded as the puncture point Q; It can be understood that a segmentation plane is set in the ultrasound three-dimensional model to divide the model into an outer area and an inner area; this segmentation plane is usually a virtual plane used to distinguish different parts of the model; a cross section of the outer area is obtained every other square side length (the side length of the three-dimensional square); these cross sections are slice views of the model at different depth levels for further analysis; on each cross section, a vertical line is obtained every other square side length; every two adjacent vertical lines form a grid column; these grid columns are small areas in the model for more detailed analysis; the boundary line is an important boundary between the effusion area and the organ area; the distance between the boundary line and L is measured, and this distance is called the surface thickness, which reflects the vertical distance from the effusion area to the organ surface; among all the grid columns, find the grid column with the smallest surface thickness; find the midpoint on the boundary line of this grid column, which is recorded as the puncture point Q; this puncture point is the best position for extracting peritoneal effusion because it is located closest to the skin in the effusion area and has the smallest surface thickness; As a preferred embodiment of the present invention, the process of setting the splitting plane includes: An initial spherical region is set, the radius of the initial spherical region is recorded as an initial radius r0, and a radius interval threshold rt is set, the initial radius is incremented according to the radius interval threshold, and starting from the initial radius, a region radius is obtained every radius interval threshold; When the radius is the radius of each area, we get several spherical areas {Area0, Area1, ..., Area n}, where Area0 is the initial spherical area, Area n is the spherical area corresponding to the radius of the nth area; obtain the effusion proportion Poa=V0 / V of each spherical area, where V0 is the sum of the volumes of all three-dimensional squares occupied by the intersection of the spherical area and the effusion area, and V is the total volume of the spherical area; select the effusion proportion that satisfies The spherical area is called the inner area, where Poa i Indicates Area i The corresponding effusion ratio, i∈[0,n] and i is a positive integer, Poa´ is a preset ratio threshold; in the ultrasound three-dimensional model, the area other than the inner area is the outer area, and the interface between the outer area and the inner area is obtained and recorded as the segmentation surface; It can be understood that the setting range of the percentage threshold is [50%, 90%]; Step S3: Make a reference plane S through the puncture point parallel to the segmentation plane, set the interval threshold It, and select a point P below the reference plane. The selection of point P satisfies , where |PQ| represents the distance between the puncture point Q and point P, and J represents the effusion area; according to the above steps, a new reference plane is made through point P, and point P1 is selected, and so on, until the latest point reaches the inner area; the puncture point and all the selected points are connected to obtain the puncture path, and the puncture path is displayed; It should be noted that the reference plane itself is not included below the reference plane; a new reference plane is made through point P, and a new point P1 is selected; this process is repeated, and a new point is selected below the new reference plane each time, until the latest point reaches the inner area; the puncture point Q is connected with all the selected points (including P, P1, etc.) to obtain a puncture path; this path is a curve starting from the puncture point Q, passing through a series of selected points, and finally reaching the inner area; finally, the puncture path is displayed so that the doctor or technician can clearly see the entire path from the puncture point to the target area; As a preferred embodiment of the present invention, the process of determining the area below the reference plane includes: obtaining two sides of the reference plane, and recording the side where the inner area is located as the area below the reference plane; It can be understood that, of the two sides of the reference plane, the side where the inner area is located is recorded as the lower side of the reference plane; this process is to clarify the direction of the puncture path and the target area, and ensure that the puncture operation can accurately reach the effusion area; As a preferred embodiment of the present invention, the process of obtaining the puncture path includes: When there are several puncture paths, the length of each puncture path is obtained, and the puncture path with the shortest length is selected as the puncture path finally displayed; Understandably, a shorter puncture path means that the puncture needle needs to penetrate less tissue; this can significantly reduce damage to healthy tissue and reduce the risk of bleeding, infection and other complications; a shorter path is usually more direct, reducing the possibility of deviation during puncture, thereby improving the accuracy of puncture; this is very important to ensure that the puncture needle accurately reaches the target area (such as an area of ​​effusion).

[0018] A peritoneal effusion image processing system, characterized by comprising: Region division module: selects a scanning area on the patient's abdomen, plans a scanning path, obtains an ultrasonic image of the scanning area in real time on the scanning path through a color ultrasound probe, and obtains an ultrasonic image set; establishes an ultrasonic three-dimensional model of the scanning area based on the ultrasonic image set; Performing three-dimensional grid division in the ultrasound three-dimensional model to obtain a plurality of three-dimensional grids; obtaining a gray value R of each three-dimensional grid, setting a gray value threshold range Rt, selecting all three-dimensional grids satisfying R∈Rt in the ultrasound three-dimensional model, recording them as low-grayscale three-dimensional grids, recording an area formed by all low-grayscale three-dimensional grids as an effusion area, and recording an area formed by all remaining three-dimensional grids as an organ area; Puncture point determination module: setting a segmentation plane, dividing the ultrasonic three-dimensional model into an outer area and an inner area, and obtaining a cross section of the outer area once every square side length, where the square side length is the side length of the three-dimensional square; obtaining a vertical line every square side length on the cross section, and every two adjacent vertical lines constitute a square column; In the square column, the outermost boundary line is obtained, and all boundary lines in the square column are obtained, and the boundary line is the boundary line between the effusion area and the organ area; the boundary line closest to the boundary line is selected, recorded as L, and the distance between the boundary line and L is obtained, recorded as the surface thickness; the square column with the smallest surface thickness value is selected, and the midpoint of the boundary line of the square column is obtained, recorded as the puncture point Q; Puncture path determination module: Make a reference plane S parallel to the segmentation plane through the puncture point, set the interval threshold It, select a point P below the reference plane, and the selection of point P satisfies , where |PQ| represents the distance between the puncture point Q and point P, and J represents the effusion area; according to the above steps, a new reference plane is made through point P, and point P1 is selected, and so on, until the latest point reaches the inner area; the puncture point and all the selected points are connected to obtain the puncture path, and the puncture path is displayed.

[0019] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for processing peritoneal effusion images, characterized in that: The following steps are involved: Step S1: selecting a scanning area on the patient's abdomen, planning a scanning path, acquiring an ultrasonic image of the scanning area in real time on the scanning path through a color ultrasound probe to obtain an ultrasonic image set; and establishing an ultrasonic three-dimensional model of the scanning area according to the ultrasonic image set; Performing three-dimensional grid division in the ultrasound three-dimensional model to obtain a plurality of three-dimensional grids; obtaining a gray value R of each three-dimensional grid, setting a gray value threshold range Rt, selecting all three-dimensional grids satisfying R∈Rt in the ultrasound three-dimensional model, recording them as low-grayscale three-dimensional grids, recording an area formed by all low-grayscale three-dimensional grids as an effusion area, and recording an area formed by all remaining three-dimensional grids as an organ area; Step S2: setting a segmentation plane, dividing the ultrasonic three-dimensional model into an outer area and an inner area, and obtaining a cross section of the outer area once every square side length, where the square side length is the side length of the three-dimensional square; obtaining a vertical line every square side length on the cross section, and every two adjacent vertical lines constitute a square column; In the square column, the outermost boundary line is obtained, and all boundary lines in the square column are obtained, and the boundary line is the boundary line between the effusion area and the organ area; Select the boundary line closest to the boundary line, record it as L, and obtain the distance between the boundary line and L, record it as the surface thickness; select the square column with the smallest value of the surface thickness, obtain the midpoint of the boundary line of the square column, record it as the puncture point Q; Step S3: Make a reference plane S through the puncture point parallel to the segmentation plane, set the interval threshold It, and select a point P below the reference plane. The selection of point P satisfies , where |PQ| represents the distance between the puncture point Q and point P, and J represents the effusion area; according to the above steps, a new reference plane is made through point P, and point P1 is selected, and so on, until the latest point reaches the inner area; the puncture point and all the selected points are connected to obtain the puncture path, and the puncture path is displayed.

2. A method for processing peritoneal effusion images according to claim 1, characterized in that: In step S1, the scanning path planning process includes: The contact surface between the color ultrasound probe and the patient's abdomen is obtained, and the contact surface is recorded as a rectangle. The scanning area is grid-divided according to the rectangle to obtain a plurality of grids. Each grid is numbered in turn, and the center point of each grid is obtained. The center points are linked according to the number to obtain a scanning path.

3. A method for processing peritoneal effusion images according to claim 1, characterized in that: In step S1, the process of establishing the ultrasonic three-dimensional model includes: An edge detection algorithm is used to perform edge detection on the ultrasonic image to identify the edge and contour of the ultrasonic image, and feature point detection is performed on the ultrasonic image to detect key points or feature points in the ultrasonic image; the ultrasonic image is registered and the registered ultrasonic images are stacked to form three-dimensional volume data, a three-dimensional view is generated according to the three-dimensional volume data using volume rendering technology, and isosurfaces are extracted from the three-dimensional volume data to generate a surface model of an organ or tissue, and finally an ultrasonic three-dimensional model is obtained.

4. The method for processing peritoneal effusion images according to claim 1, characterized in that: In step S1, the gray value threshold range setting process includes: Acquire several ultrasonic image samples of peritoneal effusion, and manually mark the effusion area on the ultrasonic image samples; acquire the grayscale value of the effusion area, and acquire the maximum and minimum values ​​of all grayscale values, and obtain the grayscale value threshold range according to the maximum and minimum values.

5. The method for processing peritoneal effusion images according to claim 1, characterized in that: In step S2, the process of setting the splitting plane includes: An initial spherical region is set, the radius of the initial spherical region is recorded as an initial radius r0, and a radius interval threshold rt is set, the initial radius is incremented according to the radius interval threshold, and starting from the initial radius, a region radius is obtained every radius interval threshold; When the radius is the radius of each area, we get several spherical areas {Area0, Area1, ..., Area n }, where Area0 is the initial spherical area, Area n is the spherical area corresponding to the radius of the nth area; obtain the effusion proportion Poa=V0 / V of each spherical area, where V0 is the sum of the volumes of all three-dimensional squares occupied by the intersection of the spherical area and the effusion area, and V is the total volume of the spherical area; select the effusion proportion that satisfies The spherical area is called the inner area, where Poa i Indicates Area i The corresponding effusion ratio, i∈[0,n] and i is a positive integer, Poa´ is a preset ratio threshold; in the ultrasonic three-dimensional model, the area other than the inner area is the outer area, and the interface between the outer area and the inner area is obtained and recorded as the segmentation surface.

6. The method for processing peritoneal effusion images according to claim 1, characterized in that: In step S3, the process of determining the area below the reference plane includes: acquiring two sides of the reference plane, and recording the side where the inner area is located as the area below the reference plane.

7. The method for processing peritoneal effusion images according to claim 1, characterized in that: In step S3, the process of obtaining the puncture path includes: When there are several puncture paths, the length of each puncture path is obtained, and the puncture path with the shortest length is selected as the puncture path finally displayed.

8. A peritoneal effusion image processing system, characterized in that: include: Region division module: selects a scanning area on the patient's abdomen, plans a scanning path, obtains an ultrasonic image of the scanning area in real time on the scanning path through a color ultrasound probe, and obtains an ultrasonic image set; establishes an ultrasonic three-dimensional model of the scanning area based on the ultrasonic image set; Performing three-dimensional grid division in the ultrasound three-dimensional model to obtain a plurality of three-dimensional grids; obtaining a gray value R of each three-dimensional grid, setting a gray value threshold range Rt, selecting all three-dimensional grids satisfying R∈Rt in the ultrasound three-dimensional model, recording them as low-grayscale three-dimensional grids, recording an area formed by all low-grayscale three-dimensional grids as an effusion area, and recording an area formed by all remaining three-dimensional grids as an organ area; Puncture point determination module: setting a segmentation plane, dividing the ultrasonic three-dimensional model into an outer area and an inner area, and obtaining a cross section of the outer area once every square side length, where the square side length is the side length of the three-dimensional square; obtaining a vertical line every square side length on the cross section, and every two adjacent vertical lines constitute a square column; In the square column, the outermost boundary line is obtained, and all boundary lines in the square column are obtained, and the boundary line is the boundary line between the effusion area and the organ area; Select the boundary line closest to the boundary line, record it as L, and obtain the distance between the boundary line and L, record it as the surface thickness; select the square column with the smallest value of the surface thickness, obtain the midpoint of the boundary line of the square column, record it as the puncture point Q; Puncture path determination module: Make a reference plane S parallel to the segmentation plane through the puncture point, set the interval threshold It, select a point P below the reference plane, and the selection of point P satisfies , where |PQ| represents the distance between the puncture point Q and point P, and J represents the effusion area; according to the above steps, a new reference plane is made through point P, and point P1 is selected, and so on, until the latest point reaches the inner area; the puncture point and all the selected points are connected to obtain the puncture path, and the puncture path is displayed.

Images