Method for aneurysm wall thickness calculation based on high-resolution blood vessel wall magnetic resonance imaging

Abstract

The present application belongs to the field of medical image processing, and particularly relates to an aneurysm wall thickness calculation method based on high-resolution blood vessel wall magnetic resonance imaging. The method comprises the following steps: marking the aneurysm wall position on the magnetic resonance image, and three-dimensionally modeling the aneurysm in the aneurysm wall region; obtaining the inner and outer boundaries of the aneurysm wall by using an edge extraction method, and performing ellipsoid fitting on the aneurysm; calculating the maximum value of the difference between the inner diameter r and the outer diameter r' of the aneurysm wall as the aneurysm wall thickness value at the position; constructing a histogram of the entire aneurysm wall thickness distribution; and adopting a color modulation method to present the thickness value in the form of different colors on the three-dimensional image of the aneurysm, so as to facilitate observation. The present application can reduce the errors generated in the subjective judgment of the aneurysm wall thickness and manual calculation, solve the problem that the thickness can only be manually measured on a two-dimensional plane in the past, can statistically analyze the distribution of the aneurysm wall thickness, and visualize the change of the aneurysm wall thickness with the position, thereby helping the user to better observe the spatial characteristics of the aneurysm wall.

Description

Technical Field

[0001] This invention belongs to the field of medical image processing, specifically relating to a method for calculating aneurysm wall thickness based on high-resolution vascular wall magnetic resonance imaging. Background Technology

[0002] An aneurysm is an abnormal protrusion or bulge that develops on the surface of an artery during its development. When it appears inside the skull, it is called a cerebral aneurysm. Most cerebral aneurysms do not cause direct symptoms, but rupture can pose serious risks to the patient. [1] The thickness of an aneurysm wall is highly correlated with its risk of rupture. Therefore, calculating the thickness of the aneurysm wall through imaging methods can help doctors understand the progression of the aneurysm and has important clinical significance.

[0003] High-resolution vascular wall magnetic resonance imaging (HR-VW MRI), as an advanced imaging technique, can perform thin-slice, high-resolution imaging. Combined with black-blood techniques, it can provide better observation of the vessel wall. [2][3] However, during manual observation of aneurysm walls via imaging, doctors need to adjust the three-dimensional display and select an appropriate viewing angle. They can only measure the distance between line segments in a two-dimensional plane and cannot determine the thickness of the aneurysm at various locations in three dimensions. Therefore, a method for calculating the thickness of the aneurysm wall at various locations based on high-resolution vascular wall magnetic resonance imaging has significant practical value. [4][5] . Summary of the Invention

[0004] The present invention aims to propose a method for calculating aneurysm wall thickness based on high-resolution vascular wall magnetic resonance imaging, so as to reduce the error when subjectively judging the aneurysm wall thickness, and to visualize the aneurysm wall thickness in three dimensions, so as to facilitate the observation of the aneurysm wall thickness at various locations.

[0005] Due to the irregularity of the aneurysm, the thickness distribution of the aneurysm wall is uneven and lacks a general pattern, making calculation difficult. Directly calculating the thickness using a three-dimensional Cartesian coordinate system cannot fully utilize the shape characteristics of the aneurysm, and the correspondence between the inner and outer edges is difficult to determine, causing significant challenges to the calculation. Therefore, this invention proposes a new scheme for calculating the aneurysm wall thickness, with the specific steps as follows:

[0006] (1) Mark the location of the aneurysm wall on the two-dimensional imaging section of the magnetic resonance image, and add a new layer between the layers of the image by the nearest neighbor interpolation method to make the spatial resolution of the three dimensions of the image the same; then use the expanded two-dimensional images of each layer and the marked aneurysm wall area to perform three-dimensional modeling of the aneurysm.

[0007] (2) Obtain the inner and outer boundaries of the aneurysm wall based on the markings of the two-dimensional cross-section, and then perform ellipsoid fitting on the aneurysm body; establish a three-dimensional spherical coordinate system with the center of the ellipsoid as the origin, and transform the spatial position (x, y, z) of each point into the position (r, θ, φ) in spherical coordinates through spherical coordinate transformation; quantize the values ​​of angles θ and φ into k-order (θ1, θ2, …, θ) respectively. k ) and (φ1, φ2, …. , φ k The spherical space is divided into k*k concentric sector intervals. For aneurysms of general size, the value of k can be between 8 and 16.

[0008] (3) For each set of angle parameters (θ) i , φ j For the corresponding sector cone space, calculate the maximum value of the difference between the inner diameter r and the outer diameter r' of the tumor wall contained in each interval, and use it as the tumor wall thickness value at that location;

[0009] (4) Perform thickness statistics and construct a histogram of the thickness distribution of the entire tumor wall;

[0010] (5) Using color modulation, the thickness value is presented in different colors on the three-dimensional image of the aneurysm, which is convenient for observation.

[0011] Compared with the prior art, the present invention has the following advantages:

[0012] (1) The method for calculating the thickness of aneurysm wall proposed in this invention can reduce the errors caused by subjective judgment of aneurysm wall thickness and manual calculation, and solve the problem that the thickness can only be manually measured on a two-dimensional plane in the past.

[0013] (2) The present invention can statistically analyze the distribution of tumor wall thickness and visualize the change of tumor wall thickness with position, thereby helping users to better observe the spatial characteristics of tumor wall. Attached Figure Description

[0014] Figure 1 : Schematic diagram of tumor wall thickness.

[0015] Figure 2 : Flowchart of the method for calculating aneurysm wall thickness.

[0016] Figure 3 : Tumor wall location annotation and 3D visualization.

[0017] Figure 4 Histogram of tumor wall thickness distribution.

[0018] Figure 5 3D visualization of tumor wall thickness. Implementation

[0019] The present invention will be further described below with reference to specific embodiments and accompanying drawings.

[0020] Figure 2 This paper demonstrates the flowchart of a method for calculating aneurysm wall thickness based on high-resolution vascular wall magnetic resonance imaging (MRI). Using high-resolution vascular wall MRI data of an aneurysm as a test case, the calculation method described in this invention is as follows:

[0021] 1. Layer-by-layer annotation of the tumor wall was performed on the two-dimensional cross-section of the magnetic resonance image. The three-dimensional visualization of the annotated image is shown in the appendix. Figure 3 By adding new layers between layers of the image using the nearest neighbor interpolation method, the spatial resolution of the image in all three dimensions is made the same, at 0.3 mm / pixel.

[0022] 2. Extract the outer edge of the labeled tumor wall image to obtain the location of the tumor wall's outer shell. Perform ellipsoid fitting based on the points on the tumor wall's outer shell to obtain the coordinates of the ellipsoid's center.

[0023] 3. Establish a spherical coordinate system with the origin at the center of the ellipsoid. Transform the spatial position (x, y, z) of each point into its position (r, θ, φ) in spherical coordinates using spherical coordinate transformation. Quantize the values ​​of angles θ and φ into 10th-order values ​​(θ1, θ2, …, θ…). 10 ) and (φ1, φ2, …. , φ 10 Divide the spherical space into 100 concentric sector-shaped intervals. Calculate the maximum difference between the inner diameter r and outer diameter r' of the tumor wall contained in each interval, and use this as the tumor wall thickness value at that location.

[0024] 4. Perform thickness statistics to obtain a histogram of the entire tumor wall thickness distribution. (See attached image) Figure 4 As shown;

[0025] 5. A color modulation method is used to present the thickness value in different colors on the three-dimensional image of the aneurysm, facilitating observation. (See attached image) Figure 5 As shown.

[0026] [1] Wang Weinan, Li Ziru, Yao Yuan, et al. Research progress on endovascular treatment of aneurysmal subarachnoid hemorrhage [J]. International Journal of Medicine and Health, 2021, 27(21): 3305-3307.

[0027] [2] Li ML, Xu WH, Lan S, et al. Atherosclerosis of middle cerebral artery: evaluation with high-resolution MR imaging at 3T [J]. Atherosclerosis, 2009, 204(2):447-452.

[0028] [3] Edjlali M, Gentric JC, Regent-Rodriguez C, et al. Does aneurysmal wall enhancement on vessel wall MRI help to distinguish stable from unstable intracranial aneurysms? [J]. Stroke, 2014, 45(12):3704.

[0029] [4] Rajabzadeh-Oghaz H, Varble N, Shallwani H, et al. Computer-assisted 3D morphology evaluation of intracranial aneurysms[J]. WorldNeurosurgery, 2018, 119.

[0030] [5] Geng Jiewen, Ji Zhe, Shen Jie, et al. Consistency analysis of morphological parameters of intracranial aneurysms measured by physicians with different experience[J]. Chinese Journal of Cerebrovascular Diseases, 2019, 16(12): 633-636.

Claims

1. A method for calculating arterial wall thickness based on high-resolution magnetic resonance imaging of blood vessel walls, characterized in that, The specific steps are as follows: (1) Mark the location of the aneurysm wall on the two-dimensional imaging section of the magnetic resonance image, and add a new layer between the layers of the image by the nearest neighbor interpolation method to make the spatial resolution of the three dimensions of the image the same; then use the expanded two-dimensional images of each layer and the marked aneurysm wall area to perform three-dimensional modeling of the aneurysm. (2) Obtain the inner and outer boundaries of the aneurysm wall based on the markings of the two-dimensional cross-section, and then perform ellipsoid fitting on the aneurysm body; establish a three-dimensional spherical coordinate system with the center of the ellipsoid as the origin, and transform the spatial position (x, y, z) of each point into the position (r, θ, φ) in spherical coordinates through spherical coordinate transformation; quantize the values ​​of angles θ and φ into k-order (θ1, θ2, …, θ k ) and (φ1, φ2, …. , φ k The spherical space is divided into k*k concentric sector intervals. For aneurysms of general size, the value of k is between 8 and 16. (3) For each set of angle parameters (θ) i , φ j For the corresponding sector cone space, calculate the maximum value of the difference between the inner diameter r and the outer diameter r' of the tumor wall contained in each interval, and use it as the tumor wall thickness value at that location; (4) Perform thickness statistics and construct a histogram of the thickness distribution of the entire tumor wall; (5) Using color modulation, the thickness value is presented in different colors on the three-dimensional image of the aneurysm, which is convenient for observation.

Images