A method and device for obtaining a height map of the posterior surface of an eyeball based on 3D-MRI

Abstract

The present invention provides a method and device for obtaining an eyeball posterior surface height map based on 3D-MRI, characterized in that the method comprises: preprocessing an MRI image using an algorithm to obtain a vitreous body boundary; obtaining a matrix of the vitreous body 3D boundary, and calculating the geometric center VC of the vitreous body and the geometric center CC of the cornea; segmenting the eyeball image, calculating the heights of all points on the eyeball posterior surface, and obtaining a surface height map; subtracting a standard reference sphere with a radius of 12.5 mm from the surface height map to calculate a residual height map of the eyeball; and obtaining eyeball morphological parameters by analyzing the residual surface height map. The present invention has the beneficial effects of using a breadth-first algorithm to search for the vitreous body boundary, so that the determination of the vitreous body boundary is more precise and accurate; applying the Rodrigues formula to the eyeball image segmentation ensures the accuracy of the equatorial plane.

Description

Technical Field

[0001] The present invention belongs to the field of eye parameter positioning, and in particular relates to a method and a device for obtaining a posterior surface height map of an eyeball based on 3D-MRI. Background Art

[0002] Posterior staphyloma is a common complication of pathological myopia, manifested by protrusion of the posterior part of the eyeball. Retinal changes, such as retinoschisis and retinal detachment, are often observed in patients and lead to irreversible visual impairment and decreased quality of life. With the global prevalence of high myopia increasing year by year, it is crucial to detect posterior staphyloma in a timely manner and assess its severity. Early detection and classification of posterior staphyloma mainly rely on ophthalmoscopy or imaging results, such as optical coherence tomography or B-ultrasound. It is also characterized by thinning of the choroid toward the edge of the staphyloma, inward protrusion of the sclera, and scleral reversal in the posterior marginal area. These findings are usually localized and do not reveal the overall shape of the eyeball.

[0003] Three-dimensional MRI can fully display the morphology of the eye. Recently, some scholars have begun to classify posterior staphylomas based on quantitative analysis of 3D-MRI images. Some fit the largest part of the MRI image to an ellipse and calculate the ratio between the major and minor axes to quantify the asphericity; others use MRI images and Fourier transforms to analyze the shape of the eyeball. However, these studies only used certain arbitrarily selected parts. As more studies are completed, it is shown that the shape of staphylomas may be more variable than previously thought.

[0004] Traditional methods of describing posterior staphylomas rely on subjective judgment and cannot fully capture the complexity of the eyeball's morphology. Although recent studies have used 3D-MRI to quantitatively analyze posterior staphylomas, there is still a lack of comprehensive understanding of the overall shape of the eyeball. Therefore, a method is needed to fully reveal the overall shape of the eyeball. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a method for obtaining a height map of the posterior surface of the eyeball based on 3D-MRI, which is particularly suitable for the accurate positioning of parameters of posterior ocular staphyloma.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: in the first aspect, a method for obtaining a height map of the posterior surface of an eyeball based on 3D-MRI, characterized in that the method includes: preprocessing the MRI image with an algorithm to obtain the vitreous boundary; obtaining the matrix of the vitreous 3D boundary, and calculating the geometric center VC of the vitreous and the geometric center CC of the cornea; segmenting the eyeball image, calculating the height of all points on the posterior surface of the eyeball, and obtaining a surface height map; subtracting a standard reference sphere with a radius of 12.5 mm from the surface height map to calculate the residual height map of the eyeball; and obtaining the eyeball morphological parameters by analyzing the residual surface height map.

[0007] Furthermore, the algorithm for preprocessing the MRI image refers to searching for the vitreous boundary using a breadth-first algorithm.

[0008] Furthermore, the breadth-first algorithm starts searching from any point in the vitreous body and moves from the center to the periphery. If the contrast difference between the next point and the previous point is less than a threshold, the next point is contained in the vitreous matrix; otherwise, the search stops.

[0009] Furthermore, the 3D boundary extraction method of the vitreous body refers to stacking the slices in sequence to obtain a matrix of the 3D boundary of the vitreous body through determining the MRI field of view and the inter-slice distance.

[0010] Furthermore, the eyeball image segmentation method refers to using Rodriguez's formula to determine a rotation matrix that aligns the vector VC_to_CC with the vertical downward direction, and using the matrix to rotate the entire eyeball.

[0011] Furthermore, the analysis is only performed within a radius of 11 mm from the center of the remaining surface height map.

[0012] In a second aspect, a device for obtaining a posterior surface height map of an eyeball based on 3D-MRI comprises:

[0013] A data collection module, for receiving MRI data and performing image preprocessing;

[0014] 3D reconstruction module, used to perform 3D reconstruction and automatically segment the vitreous body and other structures of the eyeball;

[0015] The height map generation module is used to perform coordinate transformation and alignment on the 3D model, detect the equatorial plane, and generate a height map of the posterior surface of the eyeball.

[0016] According to a third aspect, an electronic device includes:

[0017] processor, memory, and bus;

[0018] The bus is used to connect the processor and the memory;

[0019] The memory is used to store operation instructions;

[0020] The processor is used to execute a method for obtaining a posterior surface height map of an eyeball based on 3D-MRI as described in any one of claims 1 to 6 by calling the operation instruction.

[0021] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium is used to store computer instructions, and when the computer-readable storage medium is run on a computer, the computer can execute a method for obtaining a posterior surface height map of an eyeball based on 3D-MRI as described in any one of claims 1 to 6.

[0022] Compared with the prior art, this method uses a breadth-first algorithm to search for vitreous boundaries, including the selection of search starting points, the determination of contrast thresholds, and search stop conditions. It emphasizes how to determine whether a point is contained in the vitreous matrix based on contrast, making the determination of vitreous boundaries more precise and accurate.

[0023] In addition, the Rodrigues formula is applied to the eyeball image segmentation to ensure that the VC_to_CC vector is aligned with the vertical downward direction; the rotation matrix is ​​used to segment the entire eyeball into two parts, front and back, to ensure the accuracy of the equatorial plane.

[0024] In addition, the ability to describe the morphology of the posterior surface of the eyeball has been greatly improved by using Zernike polynomial analysis. Due to the orthogonality and interpretability of the Zernike coefficients, each Zernike coefficient can represent a specific morphological component of the eyeball. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 Schematic diagram of a method flow chart of an embodiment of the present invention

[0026] Figure 2 Schematic diagram of stacking 2D MRI images of the eyeball according to an embodiment of the present invention

[0027] Figure 3 Schematic diagram of the height of the posterior surface of the eyeball according to an embodiment of the present invention

[0028] Figure 4 This is a topographic map of the posterior surface of the eyeball when viewed from above according to an embodiment of the present invention.

[0029] Figure 5 The eyeball residual height topographic map according to an embodiment of the present invention is

[0030] Figure 6 Schematic diagram of the geometric shape of Zernike coefficients according to an embodiment of the present invention

[0031] Figure 7This is a diagram explaining the meaning of Zernike coefficients according to an embodiment of the present invention.

[0032] Figure 8 This is a coefficient distribution diagram of the height after Zernike decomposition of an embodiment of the present invention DETAILED DESCRIPTION

[0033] In the following description, those skilled in the art will recognize that the embodiments of the invention described below can be implemented in various ways (e.g., processes, devices, systems, equipment, or methods) on a tangible computer-readable medium. Throughout this discussion, components may be described as separate functional units (which may include subunits), but those skilled in the art will recognize that various components or parts thereof may be divided into separate components, or may be integrated together (including integration into a single system or component). It should be noted that the functions or operations discussed herein may be implemented as components. Components may be implemented in software, hardware, or a combination thereof.

[0034] The embodiments of the present invention are described below with reference to the accompanying drawings.

[0035] like Figure 1 As shown, the embodiment of the present invention discloses a method for obtaining a posterior surface height map of an eyeball based on 3D-MRI, comprising:

[0036] like Figure 2 As shown, the algorithm is used to pre-process the MRI image to obtain the vitreous boundary;

[0037] The pydicom library is used to read and parse MRI DICOM files, because this library can directly extract image data and other metadata from DICOM files, traverse all slices of the MRI scan sequence, convert each image into a numpy array, and store the image metadata; the algorithm starts from a certain point in the vitreous body and explores the neighboring points outward, and records all points marked as vitreous matrix in a three-dimensional matrix, which represents the internal structure of the vitreous body.

[0038] Obtain the matrix of the vitreous 3D boundary and calculate the geometric center VC of the vitreous and the geometric center CC of the cornea;

[0039] Based on the field of view and inter-layer spacing of MRI, the actual physical size of each pixel in different directions is determined, the vitreous boundary is extracted on each layer, and then these boundary points are stacked to form a three-dimensional matrix representing the boundary of the entire vitreous body, and the vitreous center VC and corneal center CC are found. These two points are used for subsequent alignment and segmentation operations.

[0040] like Figure 3-4 As shown, the eyeball image is segmented, the heights of all points on the posterior surface of the eyeball are calculated, and a surface height map is obtained;

[0041] Using the Rodriguez rotation formula, the eyeball is rotated around the axis until the VC_to_CC vector is aligned with the vertical direction. This can simplify subsequent geometric calculations and determine the plane of the eyeball's maximum cross-sectional area. This is used as the equatorial plane to divide the eyeball into the front and back. The distance of each point on the equatorial plane along the VC_to_CC direction is measured. These distances represent the height of the surface. Use the mpl_toolkits.mplot3d.Axes3D.plot_surface function to plot these height values ​​into a three-dimensional surface height map to intuitively display the topography of the posterior surface of the eyeball.

[0042] like Figure 5 As shown, the surface height map is subtracted from a standard reference sphere with a radius of 12.5 mm to calculate the residual height map of the eyeball;

[0043] A sphere model with a radius of 12.5 mm was created and used as a reference. The deviation between the surface of the eyeball and the surface of the sphere was calculated to generate the residual height map.

[0044] like Figure 6-8 As shown, the eyeball morphology parameters are obtained by analyzing the remaining surface height map.

[0045] In a limited central area, the surface height map is decomposed into a series of orthogonal functions using Zernike polynomials. Each function represents a specific morphological feature, and each Zernike coefficient represents a specific geometric deviation, such as spherical aberration, coma, etc. These deviations can be intuitively displayed graphically, showing the numerical distribution of the 1st to 27th Zernike coefficients, which quantify the degree of deviation of the eye surface from the ideal sphere.

[0046] The ability to describe the morphology of the posterior surface of the eyeball has been greatly improved by using Zernike polynomial analysis. Due to the orthogonality and interpretability of the Zernike coefficients, each Zernike coefficient can represent a specific morphological component of the eyeball.

[0047] The algorithm for preprocessing MRI images refers to searching the vitreous boundary using a breadth-first algorithm.

[0048] The search starts from any point in the vitreous body and moves from the center to the periphery. If the contrast difference between the next point and the previous point is less than the threshold, the next point is contained in the vitreous matrix; otherwise, the search stops.

[0049] The breadth-first search algorithm starts from a point in the vitreous body and explores neighboring points outward. For each point, check whether its contrast is lower than the set threshold. If it is lower than the threshold, it is considered to belong to the vitreous matrix. As the search proceeds, all points marked as vitreous matrix are recorded in a three-dimensional matrix, representing the internal structure of the vitreous body.

[0050] The search for the vitreous boundary uses a breadth-first algorithm, including the selection of the search starting point, the determination of the contrast threshold, and the search stop condition. It emphasizes how to judge whether a point is contained in the vitreous matrix based on the contrast, making the determination of the vitreous boundary more precise and accurate.

[0051] The 3D boundary extraction method of the vitreous body refers to determining the MRI field of view and inter-layer spacing, and stacking the slices in sequence to obtain a matrix of the 3D boundary of the vitreous body.

[0052] Based on the MRI field of view and inter-layer spacing, the actual physical size of each pixel in different directions is determined, the vitreous boundary is extracted on each layer, and then these boundary points are stacked to form a three-dimensional matrix representing the boundary of the entire vitreous.

[0053] The method for eyeball image segmentation is to use Rodriguez's formula to determine a rotation matrix that aligns the vector VC_to_CC with the vertical downward direction, and use the matrix to rotate the entire eyeball.

[0054] Find the vitreous center (VC) and corneal center (CC), these two points are used for subsequent alignment and segmentation operations, use the Rodriguez rotation formula, rotate the eyeball around the axis until the VC_to_CC vector is aligned with the vertical direction, use the Rodriguez rotation formula, rotate the eyeball around the axis until the VC_to_CC vector is aligned with the vertical direction.

[0055] The analysis was performed only within a 11 mm radius from the center of the remaining surface height map.

[0056] The Zernike analysis was performed only in the central area with a radius of 11 mm to reduce the influence of edge effects.

[0057] A device for obtaining a height map of the posterior surface of an eyeball based on 3D-MRI, comprising: a data collection module for receiving MRI data and performing image preprocessing; a 3D reconstruction module for performing 3D reconstruction and automatically segmenting the vitreous body and other structures of the eyeball; and a height map generation module for performing coordinate transformation and alignment of the 3D model and detecting the equatorial plane, and generating a height map of the posterior surface of the eyeball.

[0058] An electronic device comprises: a processor, a memory and a bus; the bus is used to connect the processor and the memory; the memory is used to store operation instructions; the processor is used to call the operation instructions.

[0059] A computer-readable storage medium is used to store computer instructions.

[0060] The embodiments of the present invention are described in detail above, but the contents described are only preferred embodiments 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.

[0061] The scheme of the present invention is described below in conjunction with the accompanying drawings, wherein the experimental methods without specific operating steps are all carried out in accordance with the corresponding product instructions, and the instruments, reagents, and consumables used in the examples can all be purchased from commercial companies unless otherwise specified.

Claims

1. A method for obtaining a posterior surface height map of an eyeball based on 3D-MRI, characterized in that: The method includes: The algorithm is used to preprocess the MRI images to obtain the vitreous boundary; Obtain the matrix of the vitreous 3D boundary and calculate the geometric center VC of the vitreous and the geometric center CC of the cornea; Segment the eyeball image, calculate the height of all points on the posterior surface of the eyeball, and obtain a surface height map; The residual height map of the eyeball is calculated by subtracting a standard reference sphere with a radius of 12.5 mm from the surface height map; By analyzing the remaining surface height map, the eyeball morphology parameters are obtained.

2. The method according to claim 1, characterized in that: The algorithm for preprocessing MRI images refers to searching the vitreous boundary using a breadth-first algorithm.

3. The breadth-first algorithm according to claim 2, characterized in that: The search starts from any point in the vitreous body and moves from the center to the periphery. If the contrast difference between the next point and the previous point is less than the threshold, the next point is contained in the vitreous matrix; otherwise, the search stops.

4. The method according to claim 1, characterized in that: The 3D boundary extraction method of the vitreous body refers to determining the MRI field of view and inter-slice distance, and stacking the slices in sequence to obtain a matrix of the 3D boundary of the vitreous body.

5. The method according to claim 1, characterized in that The eyeball image segmentation method refers to using Rodriguez's formula to determine a rotation matrix that aligns the vector VC_to_CC with the vertical downward direction, and using the matrix to rotate the entire eyeball.

6. The method according to claim 1, characterized in that The analysis was performed only within a radius of 11 mm from the center of the remaining surface height map.

7. A device for obtaining a height map of the posterior surface of an eyeball based on 3D-MRI, characterized in that: include: A data collection module, for receiving MRI data and performing image preprocessing; 3D reconstruction module, used to perform 3D reconstruction and automatically segment the vitreous body and other structures of the eyeball; The height map generation module is used to perform coordinate transformation and alignment on the 3D model, detect the equatorial plane, and generate a height map of the posterior surface of the eyeball.

8. An electronic device, characterized in that: It includes: processor, memory, and bus; The bus is used to connect the processor and the memory; The memory is used to store operation instructions; The processor is used to execute a method for obtaining a posterior surface height map of an eyeball based on 3D-MRI as described in any one of claims 1 to 6 by calling the operation instruction.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store computer instructions, and when it is run on a computer, it enables the computer to execute the method for obtaining a posterior surface height map of the eyeball based on 3D-MRI as described in any one of claims 1 to 6.

Images