A method for estimating the principal symmetry plane of kidney three-dimensional images
By extracting the spindle and center of mass in the kidney three-dimensional image processing, combining the kidney anatomical features, and using direct non-iteration geometric calculation method, the problem of relying on artificial technology for main symmetry plane estimation in kidney three-dimensional images is solved, and fast and accurate automated estimation is achieved. Suitable for humans and other species with similar anatomical structures.
Patent Information
- Application Number
- CN202411076393.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-07
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-08-07
AI Technical Summary
In the image processing of kidney three-dimensional images, the main symmetry plane estimation is overly dependent on the level of operational manual technology, resulting in high uncertainty and lack of fast and accurate automation methods.
By obtaining the three-dimensional segmentation image of the kidney, extracting the main axis and center of mass, calculate the normal vector and the cross-multiple of the center of mass as the normal vector of the main symmetry plane, combined with the anatomical features of the kidney, such as the kidney phylum and the convex hull centroid, direct non-iterative geometric calculation is used to calculate the normal vector. The method automatically estimates the main symmetry surface.
The rapid and accurate estimation of the main symmetry plane of the kidney is achieved, reducing the need for computing resources, suitable for humans and other species with similar anatomical structures, and improving estimation efficiency and accuracy.
Smart Images

Figure CN118608596B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the fields of medicine and radiology, and more specifically, to a fast, non-iterative method for estimating a kidney principal symmetry plane. Background Art
[0002] Currently, in the image processing of three-dimensional kidney images, the main symmetry plane is usually estimated manually, which is overly dependent on the operator's technical level and has high uncertainty. Summary of the invention
[0003] The purpose of the present invention is to propose a method for estimating the principal symmetry plane of a three-dimensional kidney image in order to solve the problem of excessive reliance on the technical level of manual operation and high uncertainty in the principal symmetry plane estimation in the image processing of three-dimensional images.
[0004] The technical solution of the present invention is:
[0005] A method for estimating a principal symmetry plane of a kidney three-dimensional image, the method comprising the following steps:
[0006] S1, obtaining a three-dimensional segmented image of the kidney;
[0007] S2, obtain the principal axis on the three-dimensional segmented image, defined as vector v1;
[0008] S3, obtaining the center of mass of the kidney on the three-dimensional segmented image, obtaining the center of mass of the renal hilum or the convex hull, and defining the direction of the center of mass of the kidney toward the center of mass of the renal hilum or the convex hull as vector v2;
[0009] S4. Perform a cross product of the vector v1 and the vector v2 to obtain a normal vector n of the principal symmetry plane of the three-dimensional kidney image; define the principal symmetry plane MSP of the three-dimensional kidney image based on the normal vector n and the kidney's center of mass.
[0010] Furthermore, the main axis direction of the kidney in S2 points from the end close to the head to the end close to the foot.
[0011] Furthermore, in S2, obtaining the principal axis on the three-dimensional segmented image is specifically as follows:
[0012] The three-dimensional object moment method is used to extract the three-dimensional features of the three-dimensional segmented image, and the main axis is obtained based on the three-dimensional features and defined as vector v1.
[0013] Furthermore, in S2, obtaining the principal axis on the three-dimensional segmented image is specifically as follows:
[0014] Extract the centroid C of the lower slice in the kidney CT / MRI image inf , the lower slice is close to the foot;
[0015] Extract the centroid C of the upper slice in the kidney CT / MRI image sup, the upper slice is close to the head;
[0016] The direction from the centroid of the upper slice to the centroid of the lower slice is taken as the principal axis and defined as vector v1.
[0017] Furthermore, obtaining the kidney mass center on the three-dimensional segmented image in S3 is specifically as follows: extracting the surface of the kidney three-dimensional segmented image to obtain the kidney mass center C kidney .
[0018] Furthermore, the renal hilum is located on the semi-surface of the kidney close to the spine and is the shortest distance to the center of mass of the kidney.
[0019] Furthermore, the renal hilum is obtained in S3, and the direction of the kidney mass center to the renal hilum is defined as vector v2:
[0020] S31, extract the half surface of the kidney close to the spine and including the renal hilum;
[0021] S32, calculating the distances from all points on the half surface to the center of mass of the kidney;
[0022] S33, selecting the point corresponding to the minimum distance as the renal hilum H;
[0023] S34. Define the direction of the kidney's center of mass toward the renal hilum as vector v2.
[0024] Furthermore, when the three-dimensional segmented image of the kidney is a CT / MRI image, the half surface of the kidney close to the spine and including the renal hilum is extracted based on DICOM information.
[0025] Furthermore, between S33 and S34, the following steps are also included:
[0026] Taking the renal hilum H obtained in S33 as the sphere center, an arbitrary sphere is taken around the sphere center;
[0027] The mean coordinate within this sphere is calculated and the hilum H is updated.
[0028] Furthermore, the convex hull centroid is obtained in S3, and the direction from the kidney centroid to the convex hull centroid is defined as vector v2, including:
[0029] Fill the concave area of the kidney 3D segmentation image into a convex hull and obtain the convex hull centroid C ch ;
[0030] Define the direction from the center of mass of the kidney to the center of mass of the convex hull as vector v2.
[0031] A computer device comprising: one or more processors;
[0032] The processor is used to store one or more programs;
[0033] When the one or more programs are executed by the one or more processors, the method for estimating the principal symmetry plane of the three-dimensional image of the kidney is implemented.
[0034] A computer-readable storage medium stores a computer program, which, when executed, implements the method for estimating the principal symmetry plane of a three-dimensional kidney image.
[0035] Beneficial effects of the present invention:
[0036] The method of the present invention utilizes typical anatomical features of the kidney and adopts direct, non-iterative geometric calculations, which can produce accurate results in a short time and requires very few computing resources. This method can automatically and efficiently estimate the main symmetry plane MSP of the human kidney.
[0037] The method of the present invention is mainly based on the unaffected human renal capsule anatomy and is specifically used to estimate the main symmetry planes of the human kidney, but is also applicable to other species with similar renal anatomy.
[0038] Other features and advantages of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] The above and other objects, features and advantages of the present invention will become more apparent through a more detailed description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, wherein like reference numerals generally represent like components throughout the exemplary embodiments of the present invention.
[0040] Figure 1 A schematic diagram showing the present invention defines the main symmetry plane MSP based on the normal vector and the kidney's centroid on a kidney anatomical image.
[0041] Figure 2 A schematic diagram showing the method of defining a vector v2 from the kidney mass center toward the renal hilum in the present invention.
[0042] Figure 3 A schematic diagram showing the method of defining a vector v2 in the direction of the kidney mass center toward the convex hull mass center in the present invention. DETAILED DESCRIPTION
[0043] The preferred embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein.
[0044] The present invention proposes a method for estimating the principal symmetry plane of a three-dimensional kidney image, which is based on the unaffected human kidney capsule anatomy and is specifically used to estimate the principal symmetry plane of the human kidney, and is also applicable to other species with similar kidney anatomy. The technique uses direct, non-iterative geometric calculations to produce accurate results in a short time, and can automatically and efficiently estimate the principal symmetry plane MSP of the human kidney.
[0045] Embodiment 1:
[0046] like Figure 1 As shown, the present invention provides a method for estimating the principal symmetry plane of a kidney three-dimensional image, the method comprising the following steps:
[0047] S1, obtaining a three-dimensional segmented image of the kidney;
[0048] S2, obtaining a principal axis on the three-dimensional segmented image, defined as a vector v1; the principal axis direction of the kidney points from an end close to the head to an end close to the foot;
[0049] Among them, the following two methods can be used to obtain the main axis on the three-dimensional segmentation image:
[0050] (a) The three-dimensional object moment method is used to extract the three-dimensional features of the three-dimensional segmented image, and the main axis is obtained based on the three-dimensional features, which is defined as vector v1.
[0051] (b) Extract the centroid C of the lower slice in the kidney CT / MRI image inf , the lower slice is close to the foot; extract the centroid C of the upper slice in the kidney CT / MRI image sup , the upper slice is close to the head; the direction from the centroid of the upper slice to the centroid of the lower slice is taken as the main axis, which is defined as vector v1:
[0052] v1 = C sup -C inf / ||C sup -C inf ||
[0053] || || means normalizing the vector and transforming a single sample feature vector into a feature vector with unit length.
[0054] S3. Obtain the center of mass of the kidney on the three-dimensional segmented image, obtain the center of mass of the kidney hilum or convex hull, and define the direction of the center of mass of the kidney toward the center of mass of the kidney hilum or convex hull as vector v2, including:
[0055] (a) Extracting the surface of the kidney 3D segmentation image;
[0056] (b) Obtaining the kidney center of mass C based on the surface of the kidney 3D segmentation image kidney ;
[0057] (c) Obtain the center of mass of the kidney hilum or convex hull, and move the center of mass of the kidney to the center of mass of the kidney hilum H or convex hull C ch The direction is defined as vector v2;
[0058] S4, cross-product the vector v1 and the vector v2 to obtain the normal vector n of the main symmetry plane of the kidney three-dimensional image; define the main symmetry plane MSP of the kidney three-dimensional image based on the normal vector n and the kidney mass center, as follows: Figure 1 shown.
[0059] Specifically, the surface (capsule) of the human kidney is non-convex and exhibits a single depression, the hilum, which we represent as a single point H lying on the principal symmetry plane.
[0060] In S3, when the renal hilum H is obtained, the kidney centroid C kidney When the direction toward the renal hilum H is defined as vector v2, such as Figure 2 shown (note: blood vessels are not considered part of the kidney segmentation), including:
[0061] S31, extracting the semi-surface of the kidney close to the spine and including the renal hilum; when the three-dimensional segmented image of the kidney is a CT / MRI image, extracting the semi-surface of the kidney close to the spine and including the renal hilum based on DICOM information;
[0062] S32, calculate all points on the half surface to reach the kidney mass center C kidney The distance d;
[0063] S33, selecting the point corresponding to the minimum distance min(d) as the renal hilum H;
[0064] S34, define the direction of the kidney's mass center toward the renal hilum as vector v2;
[0065] v2 = C kidney -H / ||C kidney - H||
[0066] To increase robustness, the steps between S33 and S34 also include: taking the renal hilum H obtained in S33 as the sphere center and taking an arbitrary sphere around the sphere center; calculating the average coordinates within the sphere and updating the renal hilum H; this helps to not rely on a single point and reduce errors.
[0067] Specifically, when obtaining the convex hull centroid C in S3 ch , the kidney centroid C kidney To the centroid C of the convex hull ch When the direction is defined as vector v2, such as Figure 3 shown (note: blood vessels are not considered part of the kidney segmentation), including:
[0068] Fill the concave area of the kidney 3D segmentation image into a convex hull and obtain the convex hull centroid C ch ;
[0069] Define the direction from the center of mass of the kidney to the center of mass of the convex hull as vector v2;
[0070] v2 = C kidney - C ch / ||C kidney - C ch ||.
[0071] Embodiment 2:
[0072] A computer device comprising: one or more processors;
[0073] The processor is used to store one or more programs;
[0074] When the one or more programs are executed by the one or more processors, the method for estimating the principal symmetry plane of the three-dimensional image of the kidney is implemented.
[0075] Embodiment 3:
[0076] A computer-readable storage medium stores a computer program, which, when executed, implements the method for estimating the principal symmetry plane of a three-dimensional kidney image.
Claims
1. A method for estimating the principal symmetry plane of a three-dimensional kidney image, characterized in that: The method comprises the following steps: S1, obtaining a three-dimensional segmented image of the kidney; S2, obtain the principal axis on the three-dimensional segmented image, defined as vector v1; S3, obtaining the center of mass of the kidney on the three-dimensional segmented image, obtaining the center of mass of the renal hilum or the convex hull, and defining the direction of the center of mass of the kidney toward the center of mass of the renal hilum or the convex hull as vector v2; S4, performing a cross product of the vector v1 and the vector v2 to obtain a normal vector n of the principal symmetry plane of the kidney three-dimensional image; defining the principal symmetry plane MSP of the kidney three-dimensional image based on the normal vector n and the kidney mass center; In S2, the main axis is obtained on the three-dimensional segmented image as follows: Extract the centroid C of the lower slice in the kidney CT / MRI image inf , the lower slice is close to the foot; Extract the centroid C of the upper slice in the kidney CT / MRI image sup , the upper slice is close to the head; The direction from the centroid of the upper slice to the centroid of the lower slice is taken as the principal axis and defined as vector v1.
2. The method for estimating the principal symmetry plane of a kidney three-dimensional image according to claim 1, characterized in that In S2, the main axis direction of the kidney is from the end close to the head to the end close to the foot.
3. The method for estimating the principal symmetry plane of a kidney three-dimensional image according to claim 1, characterized in that In S2, the main axis is obtained on the three-dimensional segmented image as follows: The three-dimensional object moment method is used to extract the three-dimensional features of the three-dimensional segmented image, and the main axis is obtained based on the three-dimensional features and defined as vector v1.
4. The method for estimating the principal symmetry plane of a kidney three-dimensional image according to claim 1, characterized in that In S3, the kidney mass center is obtained on the three-dimensional segmented image by extracting the surface of the kidney three-dimensional segmented image and obtaining the kidney mass center C kidney .
5. The method for estimating the principal symmetry plane of a kidney three-dimensional image according to claim 1, characterized in that The renal hilum is located on the semi-surface of the kidney close to the spine and is shortest to the center of mass of the kidney.
6. The method for estimating the principal symmetry plane of a kidney three-dimensional image according to claim 1, characterized in that In S3, the renal hilum is obtained, and the direction of the kidney mass center toward the renal hilum is defined as vector v2: S31, extract the half surface of the kidney close to the spine and including the renal hilum; S32, calculating the distances from all points on the half surface to the center of mass of the kidney; S33, selecting the point corresponding to the minimum distance as the renal hilum H; S34. Define the direction of the kidney's center of mass toward the renal hilum as vector v2.
7. The method for estimating the principal symmetry plane of a kidney three-dimensional image according to claim 6, characterized in that: When the three-dimensional segmented image of the kidney is a CT / MRI image, the half surface of the kidney close to the spine and including the renal hilum is extracted based on DICOM information.
8. The method for estimating the principal symmetry plane of a kidney three-dimensional image according to claim 6, characterized in that: The S33 and S34 also include: Taking the renal hilum H obtained in S33 as the sphere center, an arbitrary sphere is taken around the sphere center; The mean coordinate within this sphere is calculated and the hilum H is updated.
9. The method for estimating the principal symmetry plane of a kidney three-dimensional image according to claim 1, characterized in that In S3, the convex hull centroid is obtained, and the direction from the kidney centroid to the convex hull centroid is defined as vector v2, including: Fill the concave area of the kidney 3D segmentation image into a convex hull and obtain the centroid Cch of the convex hull; Define the direction from the center of mass of the kidney to the center of mass of the convex hull as vector v2.
10. A computer device, characterized in that: include: one or more processors; The processor is used to store one or more programs; When the one or more programs are executed by the one or more processors, the method for estimating the principal symmetry plane of a kidney three-dimensional image as claimed in any one of claims 1 to 9 is implemented.
11. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed, the method for estimating the principal symmetry plane of a kidney three-dimensional image as claimed in any one of claims 1 to 9 is implemented.
