Cerebral cortex thickness measurement method and system based on level set image segmentation algorithm

Abstract

The invention provides a cerebral cortex thickness measurement method and system based on a level set image segmentation algorithm, and belongs to the technical field of neural image analysis, and the method comprises the steps: obtaining a brain magnetic resonance imaging image, and carrying out the skull stripping processing; performing three-dimensional visualization on the stripped image through volume rendering, and acquiring three-dimensional coordinates of points of interest and a region of interest on the surface of the cortex by using a preset interaction means; using a level set method to carry out brain grey matter and brain white matter segmentation on the region of interest to obtain a brain grey matter segmentation layer and a brain white matter segmentation layer, and storing three-dimensional coordinate data of points in the segmentation layers; calculating the minimum Euclidean distance according to the three-dimensional coordinate data of the points in the segmentation layer and the three-dimensional coordinates of the points of interest to obtain the cortex thickness of the points of interest; the method solves the problems that the existing cerebral cortex thickness measurement lacks a visual means, is not supported by a mathematical model, cannot be accurate to the thickness of an interest point, and cannot display and retain comprehensive information.

Description

Technical Field

[0001] This invention belongs to the field of neuroimaging analysis technology, and in particular relates to a method and system for measuring the thickness of the cerebral cortex based on a level set image segmentation algorithm. Background Technology

[0002] Cerebral cortical thickness measurement is a key topic in neuroimaging analysis. Its main purpose is to quantitatively assess the spatial distribution and changing trends of brain gray matter thickness. This technology is of great significance for studying normal brain development, aging processes, and various neuropsychiatric diseases (such as Alzheimer's disease, schizophrenia, autism, and multiple sclerosis). Cortical thickness is considered an important indicator reflecting neuronal density, hierarchical structure, and brain region health status. Therefore, it is often used as a potential biomarker for early disease screening or longitudinal tracking.

[0003] Cortical thickness measurements are typically based on structural magnetic resonance imaging (MRI), especially high-resolution T1-weighted images, because these images provide good contrast between gray and white matter. The basic principle of the measurement is to extract the white matter-gray matter boundary (i.e., the white matter surface) and the gray matter-cerebrospinal fluid boundary (i.e., the brain surface or pial surface), and then calculate the thickness at each point between the two, usually defined as the shortest distance or some average distance between the two surfaces.

[0004] In existing technologies, the surface of the cerebral cortex is automatically segmented through geometric neural networks, but the surface of the cerebral cortex cannot be directly reconstructed, and there is a lack of visualization methods and measurement of the thickness of the cerebral cortex. It does not output accurate thickness information supported by mathematical models. In existing technologies, cortical thickness is measured based on magnetic resonance imaging (MR) scan information. By establishing triangular networks of the outer surface, inner surface, and intermediate surface, the cortical thickness is calculated and mapped. However, comprehensive information is not displayed and preserved, and the accuracy of cortical thickness is insufficient, failing to pinpoint the thickness at the point of interest. Summary of the Invention

[0005] To address the aforementioned shortcomings in existing technologies, this invention provides a method and system for measuring cortical thickness based on a level set image segmentation algorithm. This method solves the problems of lack of visualization methods, lack of mathematical model support, inability to accurately measure thickness down to the point of interest, and inability to comprehensively display and retain information in existing cortical thickness measurements.

[0006] To achieve the above objectives, the technical solution adopted by this invention is as follows: On one hand, it provides a method for measuring cortical thickness based on a level set image segmentation algorithm, comprising the following steps: S1. Obtain brain magnetic resonance imaging images and perform skull dissection processing on the brain magnetic resonance imaging images to obtain dissected images. S2. Through volume rendering, the stripped image is visualized in three dimensions to obtain a three-dimensional brain image. Using preset interactive methods, the three-dimensional coordinates of points of interest and regions of interest on the cortical surface are obtained. S3. Using the level set method, the region of interest is segmented into gray matter and white matter to obtain gray matter segmentation layers and white matter segmentation layers. The three-dimensional coordinate data of the points in the gray matter segmentation layers and white matter segmentation layers are then stored. S4. Based on the three-dimensional coordinate data of points within the gray matter segmentation layer and white matter segmentation layer of the brain, combined with the three-dimensional coordinates of the point of interest, the cortical thickness at the point of interest is obtained by calculating the minimum Euclidean distance.

[0007] The beneficial effects of this invention are as follows: This invention provides a visual way to accurately measure the cortical thickness of target points by setting up a phased measurement of cerebral cortex thickness and using volume rendering for three-dimensional visualization. It also uses a technique to accurately calculate cortical thickness by taking points on the cortical surface and uses the level set method as the segmentation method to improve the accuracy of white and gray matter segmentation of the brain, improve interpretability, has a mathematical basis, and is suitable for medical use.

[0008] Further, S2 includes the following steps: S201. Using graphics technology, through volume rendering, the stripped image is visualized in three dimensions to obtain a three-dimensional brain image. S202. Using preset interactive methods, by obtaining the screen mouse click position and the camera direction in the computer graphics space, effective point detection is performed on the three-dimensional brain image to obtain the coordinates of interest points on the cortical surface. S203. Using the coordinates of the point of interest as the center and the preset length as the side length, delineate the cubic spatial region to obtain the region of interest.

[0009] The beneficial effects of the above-mentioned further solutions are as follows: the present invention performs three-dimensional visualization through volume rendering and realizes the sampling of cortical surface through human-computer interaction, providing a data basis for the measurement of cerebral cortex thickness and improving the accuracy of cerebral cortex thickness measurement.

[0010] Furthermore, step S3 includes the following steps: S301. Using a level set segmentation algorithm based on multiplicative and additive bias, by introducing multiplicative and additive bias fields and considering Gaussian noise, a brain magnetic resonance imaging image is modeled to obtain a brain magnetic resonance imaging model. Based on the fact that the real image is approximately constant in a preset region, an approximate constant is obtained. S302. Based on the points in the region of interest, select a local neighborhood, and based on the gradual variability of the multiplicative bias field, obtain an approximate representation of the multiplicative bias field. Based on the approximate constant, obtain an approximate representation of the image intensity within the local neighborhood. S303. Based on the approximate representation of image intensity, define cluster centers, rewrite the preset clustering criterion function using the cluster centers, and introduce a non-negative window function to obtain the rewritten clustering criterion function. S304. Based on the rewritten clustering criterion function, the energy function is obtained by defining energy and combining it with the regularization term; S305. Minimize the energy function, take partial derivatives of the multiplicative bias field, the additive bias field, and the approximate constant, and update them alternately to obtain the brain gray matter segmentation layer and the brain white matter segmentation layer, and store the three-dimensional coordinate data of the points in the brain gray matter segmentation layer and the brain white matter segmentation layer.

[0011] Furthermore, the expression for the approximate representation of image intensity within the local neighborhood is as follows: ; in, Point The image intensity approximation is as follows: Indicates the preset area One point inside, Point Multiplicative bias field, Represents a constant. Represents a local area One point inside, Point Zero-mean Gaussian noise at that location.

[0012] Furthermore, the rewritten expression for the clustering criterion function is as follows: ; in, This represents the rewritten clustering criterion function. Indicates the total number of preset areas. This represents a non-negative window function.

[0013] Furthermore, the expressions for partial derivatives of the multiplicative bias field, the additive bias field, and the approximate constant are as follows: ; ; ; in, Represents the partial derivative of the approximate constant. This represents the kernel function that applies weighting within a local neighborhood. This represents the partial derivative of the multiplicative bias field. Indicates the area membership function, This indicates the first abbreviation formula, specifically: , This indicates the second abbreviation formula, specifically: , This represents the partial derivative of the additive bias field.

[0014] The beneficial effects of the above-mentioned further solutions are as follows: This invention improves the accuracy of brain magnetic resonance imaging image modeling by using a level set segmentation algorithm based on multiplicative and additive biases, and enhances interpretability by using a mathematical model as support.

[0015] Furthermore, S4 specifically refers to: Based on the three-dimensional coordinate data of points within the gray matter and white matter segmentation layers of the brain, the point set of the segmentation layer is obtained. Combined with the three-dimensional coordinate data of the point of interest, the cortical thickness at the point of interest is calculated by calculating the minimum Euclidean distance from the point of interest to all points in the point set of the segmentation layer.

[0016] Furthermore, the expression for calculating the minimum Euclidean distance from the point of interest to all points in the point set of the segmentation layer is as follows: ; in, This indicates the thickness of the cortex at the selected point of interest A. This represents the x-axis coordinate of point A of interest. This represents the y-coordinate of point of interest A. This represents the z-axis coordinate of point A of interest. This represents the x-coordinate of the j-th point in the segmentation layer. This represents the y-coordinate of the j-th point in the segmentation layer. This represents the z-axis coordinate of the j-th point in the segmentation layer. This indicates the j-th element, and , This indicates that the segmentation layer has a total of M points.

[0017] The beneficial effects of the above-mentioned further solutions are as follows: the present invention solves the distance from a point to a set by the minimum Euclidean distance, thereby achieving accurate measurement of the thickness of the cerebral cortex.

[0018] On the other hand, a system for measuring cortical thickness based on a level set image segmentation algorithm is provided, including: The data acquisition module is used to acquire brain magnetic resonance imaging images; The brain tissue extraction algorithm module is used to perform cranial bone stripping processing on brain magnetic resonance imaging images to obtain stripped images; The 3D visualization module is used to visualize the stripped image in 3D using graphics technology, thereby obtaining a 3D brain image; The cortical surface point acquisition module is used to perform effective point detection on a 3D brain image by using preset interactive methods, acquiring the screen mouse click position and the camera orientation in the computer graphics space, and obtaining the 3D coordinates of interest points on the cortical surface. The region of interest extraction module is used to obtain regions of interest. The brain gray and white matter segmentation algorithm module is used to segment the brain gray and white matter of the region of interest using the level set method, to obtain the brain gray matter segmentation layer and the brain white matter segmentation layer, and to store the three-dimensional coordinate data of the points in the brain gray matter segmentation layer and the brain white matter segmentation layer. The cerebral cortex thickness measurement module is used to obtain the cortical thickness at the point of interest by calculating the minimum Euclidean distance based on the three-dimensional coordinate data of points within the gray matter segmentation layer and the white matter segmentation layer of the brain, combined with the three-dimensional coordinates of the point of interest.

[0019] The beneficial effects of the above-mentioned further solutions are as follows: the present invention performs three-dimensional visualization through volume rendering and improves the accuracy of cortical thickness calculation by sampling points on the cortical surface and extracting regions of interest. Attached Figure Description

[0020] Figure 1 This is a flowchart of the method of the present invention.

[0021] Figure 2 This is a system flowchart in this embodiment. Detailed Implementation

[0022] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0023] Before describing this embodiment, the following terms will be explained: MRI: Magnetic Resonance Imaging.

[0024] Example 1 like Figure 1 As shown, this invention provides a method for measuring cortical thickness based on a level set image segmentation algorithm, the implementation of which is as follows: S1. Obtain brain magnetic resonance imaging (MRI) images and perform cranial dissection processing on the brain MRI images to obtain dissected images.

[0025] In this embodiment, brain MRI medical image data is acquired, and the brain MRI images are subjected to skull dissection processing to remove non-brain parts, resulting in dissected images.

[0026] S2. Through volume rendering, the stripped image is visualized in three dimensions to obtain a three-dimensional brain image. Using preset interactive methods, the three-dimensional coordinates of points of interest and regions of interest on the cortical surface are obtained. The specific steps are as follows: S201. Using graphics technology, through volume rendering, the stripped image is visualized in three dimensions to obtain a three-dimensional brain image. S202. Using preset interactive methods, by obtaining the screen mouse click position and the camera direction in the computer graphics space, effective point detection is performed on the three-dimensional brain image to obtain the coordinates of interest points on the cortical surface. S203. Using the coordinates of the point of interest as the center and the preset length as the side length, delineate the cubic spatial region to obtain the region of interest.

[0027] In this embodiment, the dissected image of the skull is visualized in three dimensions using volume rendering through graphics techniques. After three-dimensional visualization, users can intuitively obtain the coordinates of points of interest and local regions of interest on the cortical surface through preset interactive methods.

[0028] In this embodiment, the specific steps for interest point extraction are as follows: MRI volume data is visualized in three dimensions using volume rendering in computer graphics technology; effective points are detected using the screen mouse click position and the camera direction in the computer graphics space, i.e., the first non-noise point with a voxel value is detected; the information of the interest point, including its three-dimensional coordinates and voxel value, is returned to obtain the coordinates of the interest point on the cortical surface.

[0029] In this embodiment, the acquisition of the local region of interest specifically refers to a cubic spatial region centered on the coordinates of the point of interest and defined by a specific side length. By extracting data from this region in the original medical image, accurate measurement can be achieved and computational speed can be improved.

[0030] S3. Using the level set method, segment the region of interest into gray and white matter to obtain gray matter and white matter segmentation layers. Then, store the three-dimensional coordinate data of the points within the gray and white matter segmentation layers. The specific steps are as follows: S301. Using a level set segmentation algorithm based on multiplicative and additive biases, a brain magnetic resonance imaging (MRI) image is modeled by introducing multiplicative and additive bias fields and considering Gaussian noise. The brain MRI model is obtained, and the approximate constant is obtained based on the fact that the real image is approximately constant in a preset region.

[0031] In this embodiment, the brain MRI image is first modeled, and the expression is as follows: ; in, Represents the observed image, Represents a real image. Indicates a multiplicative bias field. Indicates an additive bias field. This represents zero-mean Gaussian noise; In a very small area The inner approximation is a constant. That is, for have , Values ​​representing the real image, Indicates the preset area A little bit inside.

[0032] S302. Based on the points in the region of interest, select a local neighborhood, and based on the gradual variability of the multiplicative bias field, obtain an approximate representation of the multiplicative bias field. Based on the approximate constant, obtain an approximate representation of the image intensity within the local neighborhood. S303. Based on the approximate representation of image intensity, define cluster centers, rewrite the preset clustering criterion function using the cluster centers, and introduce a non-negative window function to obtain the rewritten clustering criterion function.

[0033] In this embodiment, based on the region of interest Points in In the local neighborhood Inside, Represents a local area radius, Represents local neighborhood any point in ; Due to the multiplicative bias field The gradual changeability can be approximated as for... have: Therefore, we have: , Point Multiplicative bias field, Point Multiplicative bias field; Therefore, within the local neighborhood, image intensity It can be approximated as: ; in, Point Zero-mean Gaussian noise at the location; The intensity values ​​within this local neighborhood cluster around a local cluster center, which is defined as follows: .

[0034] In this embodiment, a criterion function for measuring the clustering effect is defined, and its expression is as follows: ; in, This represents the criterion function for measuring the clustering effect. Represents a local neighborhood. Indicates the area The membership function satisfies when ; Based on cluster centers It is a local cluster center, replace it with The rewritten clustering criterion function is: ; in, This represents the rewritten clustering criterion function, which introduces a non-negative window function. , making when hour .

[0035] S304. Based on the rewritten clustering criterion function, the energy function is obtained by defining energy and combining it with regularization terms.

[0036] In this embodiment, energy is defined based on the rewritten clustering criterion function, and the specific expression is as follows: ; in, Represents the energy function; Combining this with the regularization term, the expression for the energy function is obtained as follows: ; in, This represents the energy function after incorporating the regularization term. Describes the level set function. This represents the true image of a region that is approximately constant. and Both represent regularization terms. The parameter that controls the degree of smoothness. A function that controls the degree of smoothness. The parameter represents the property of maintaining the sign distance of the level set function. This represents a function that maintains the sign distance property of the level set function.

[0037] S305. Minimize the energy function, take partial derivatives of the multiplicative bias field, the additive bias field, and the approximate constant, and update them alternately to obtain the brain gray matter segmentation layer and the brain white matter segmentation layer, and store the three-dimensional coordinate data of the points in the brain gray matter segmentation layer and the brain white matter segmentation layer.

[0038] In this embodiment, an energy minimization strategy is adopted, which calculates the partial derivatives of each variable and updates them alternately in sequence. The specific expression is as follows: ; ; in, Represents the partial derivative of the approximate constant. This represents the kernel function that applies weighting within a local neighborhood. This represents the partial derivative of the multiplicative bias field. This indicates the first abbreviation formula, specifically: , This indicates the second abbreviation formula, specifically: ; ; in, Represents the partial derivative of the additive bias field; Iterative calculation of the above parameters yields the final segmented image and bias field. Due to the extremely small region... The inner approximation is a constant. The image segmentation is specifically as follows: The bias field is specifically a multiplicative bias field. and additive bias field .

[0039] S4. Based on the three-dimensional coordinate data of points within the gray matter and white matter segmentation layers of the brain, combined with the three-dimensional coordinates of the points of interest, the cortical thickness at the points of interest is obtained by calculating the minimum Euclidean distance, specifically: Based on the three-dimensional coordinate data of points within the gray matter and white matter segmentation layers of the brain, the point set of the segmentation layer is obtained. Combined with the three-dimensional coordinate data of the point of interest, the cortical thickness at the point of interest is calculated by calculating the minimum Euclidean distance from the point of interest to all points in the point set of the segmentation layer.

[0040] In this embodiment, the point set of the segmentation layer is obtained based on the three-dimensional coordinate data of points within the gray matter segmentation layer and the white matter segmentation layer of the brain. The 3D coordinate data of the point of interest are ; The cortical thickness can be defined as the minimum Euclidean distance from point A to all points in set B, and the calculation expression is as follows: ; in, This indicates the thickness of the cortex at the selected point of interest A. This represents the x-axis coordinate of point A of interest. This represents the y-coordinate of point of interest A. This represents the z-axis coordinate of point A of interest. This represents the x-coordinate of the j-th point in the segmentation layer. This represents the y-coordinate of the j-th point in the segmentation layer. This represents the z-axis coordinate of the j-th point in the segmentation layer. This indicates the j-th element, and , This indicates that the segmentation layer has a total of M points; Based on the calculated minimum Euclidean distance, the cortical thickness at the point of interest is obtained, thus completing the measurement of cerebral cortex thickness.

[0041] Example 2 like Figure 2 As shown, the present invention provides a cerebral cortex thickness measurement system based on a level set image segmentation algorithm, which is applied to the cerebral cortex thickness measurement method based on a level set image segmentation algorithm as described in Example 1, including: a data acquisition module for acquiring brain magnetic resonance imaging images; The brain tissue extraction algorithm module is used to perform cranial bone stripping processing on brain magnetic resonance imaging images to obtain stripped images; The 3D visualization module is used to visualize the stripped image in 3D using graphics technology, thereby obtaining a 3D brain image; The cortical surface point acquisition module is used to perform effective point detection on a 3D brain image by using preset interactive methods, acquiring the screen mouse click position and the camera orientation in the computer graphics space, and obtaining the 3D coordinates of interest points on the cortical surface. The region of interest extraction module is used to obtain regions of interest. The brain gray and white matter segmentation algorithm module is used to segment the brain gray and white matter of the region of interest using the level set method, to obtain the brain gray matter segmentation layer and the brain white matter segmentation layer, and to store the three-dimensional coordinate data of the points in the brain gray matter segmentation layer and the brain white matter segmentation layer. The cerebral cortex thickness measurement module is used to obtain the cortical thickness at the point of interest by calculating the minimum Euclidean distance based on the three-dimensional coordinate data of points within the gray matter segmentation layer and the white matter segmentation layer of the brain, combined with the three-dimensional coordinates of the point of interest.

Claims

1. A method for measuring cortical thickness based on a level set image segmentation algorithm, characterized in that, Includes the following steps: S1. Obtain brain magnetic resonance imaging images and perform skull dissection processing on the brain magnetic resonance imaging images to obtain dissected images. S2. Through volume rendering, the stripped image is visualized in three dimensions to obtain a three-dimensional brain image. Using preset interactive methods, the three-dimensional coordinates of points of interest and regions of interest on the cortical surface are obtained. S3. Using the level set method, the brain gray matter and brain white matter are segmented in the region of interest to obtain the brain gray matter segmentation layer and the brain white matter segmentation layer, and the three-dimensional coordinate data of the points in the brain gray matter segmentation layer and the brain white matter segmentation layer are stored. S4. Based on the three-dimensional coordinate data of points within the gray matter segmentation layer and white matter segmentation layer of the brain, combined with the three-dimensional coordinates of the point of interest, the cortical thickness at the point of interest is obtained by calculating the minimum Euclidean distance.

2. The method for measuring cortical thickness based on level set image segmentation algorithm according to claim 1, characterized in that, S2 includes the following steps: S201. Using graphics technology, through volume rendering, the stripped image is visualized in three dimensions to obtain a three-dimensional brain image. S202. Using preset interactive methods, by obtaining the screen mouse click position and the camera direction in the computer graphics space, effective point detection is performed on the three-dimensional brain image to obtain the coordinates of interest points on the cortical surface. S203. Using the coordinates of the point of interest as the center and the preset length as the side length, delineate the cubic spatial region to obtain the region of interest.

3. The method for measuring cortical thickness based on level set image segmentation algorithm according to claim 1, characterized in that, S3 includes the following steps: S301. Using a level set segmentation algorithm based on multiplicative and additive bias, by introducing multiplicative and additive bias fields and considering Gaussian noise, a brain magnetic resonance imaging image is modeled to obtain a brain magnetic resonance imaging model. Based on the fact that the real image is approximately constant in a preset region, an approximate constant is obtained. S302. Based on the points in the region of interest, select a local neighborhood, and based on the gradual variability of the multiplicative bias field, obtain an approximate representation of the multiplicative bias field. Based on the approximate constant, obtain an approximate representation of the image intensity within the local neighborhood. S303. Based on the approximate representation of image intensity, define cluster centers, rewrite the preset clustering criterion function using the cluster centers, and introduce a non-negative window function to obtain the rewritten clustering criterion function. S304. Based on the rewritten clustering criterion function, the energy function is obtained by defining energy and combining it with the regularization term; S305. Minimize the energy function, take partial derivatives of the multiplicative bias field, the additive bias field, and the approximate constant, and update them alternately to obtain the brain gray matter segmentation layer and the brain white matter segmentation layer, and store the three-dimensional coordinate data of the points in the brain gray matter segmentation layer and the brain white matter segmentation layer.

4. The method for measuring cortical thickness based on level set image segmentation algorithm according to claim 3, characterized in that, The expression for the brain magnetic resonance imaging model is as follows: in, Represents the observed image, Represents a real image. Indicates a multiplicative bias field. Indicates an additive bias field. This represents zero-mean Gaussian noise.

5. The method for measuring cortical thickness based on level set image segmentation algorithm according to claim 4, characterized in that, The expression for the approximate representation of the image intensity within the local neighborhood is as follows: in, Point The image intensity approximation is as follows: Indicates the preset area One point inside, Point Multiplicative bias field, Represents a constant. Represents a local area One point inside, Point Zero-mean Gaussian noise at that location.

6. The method for measuring cortical thickness based on level set image segmentation algorithm according to claim 5, characterized in that, The rewritten expression for the clustering criterion function is as follows: in, This represents the rewritten clustering criterion function. Indicates the total number of preset areas. This represents a non-negative window function.

7. The method for measuring cortical thickness based on level set image segmentation algorithm according to claim 6, characterized in that, The expressions for the partial derivatives of the multiplicative bias field, the additive bias field, and the approximate constant are shown below: in, Represents the partial derivative of the approximate constant. This represents the kernel function that applies weighting within a local neighborhood. This represents the partial derivative of the multiplicative bias field. Indicates the region membership function, This indicates the first abbreviation formula, specifically: , This indicates the second abbreviation formula, specifically: , This represents the partial derivative of the additive bias field.

8. The method for measuring cortical thickness based on level set image segmentation algorithm according to claim 1, characterized in that, Specifically, S4 is: Based on the three-dimensional coordinate data of points within the gray matter and white matter segmentation layers of the brain, the point set of the segmentation layer is obtained. Combined with the three-dimensional coordinate data of the point of interest, the cortical thickness at the point of interest is calculated by calculating the minimum Euclidean distance from the point of interest to all points in the point set of the segmentation layer.

9. The method for measuring cortical thickness based on level set image segmentation algorithm according to claim 8, characterized in that, The expression for calculating the minimum Euclidean distance from the point of interest to all points in the point set of the segmentation layer is as follows: in, This indicates the thickness of the cortex at the selected point of interest A. This represents the x-axis coordinate of point A of interest. This represents the y-coordinate of point of interest A. This represents the z-axis coordinate of point A of interest. This represents the x-coordinate of the j-th point in the segmentation layer. This represents the y-coordinate of the j-th point in the segmentation layer. This represents the z-axis coordinate of the j-th point in the segmentation layer. This indicates the j-th element, and , This indicates that the segmentation layer has a total of M points.

10. A system for measuring cortical thickness based on a level set image segmentation algorithm, applied to a method for measuring cortical thickness based on a level set image segmentation algorithm as described in claims 1-9, characterized in that, include: The data acquisition module is used to acquire brain magnetic resonance imaging images; The brain tissue extraction algorithm module is used to perform cranial bone stripping processing on brain magnetic resonance imaging images to obtain stripped images; The 3D visualization module is used to visualize the stripped image in 3D using graphics technology, thereby obtaining a 3D brain image; The cortical surface point acquisition module is used to perform effective point detection on a 3D brain image by using preset interactive methods, acquiring the screen mouse click position and the camera orientation in the computer graphics space, and obtaining the 3D coordinates of interest points on the cortical surface. The region of interest extraction module is used to obtain regions of interest. The brain gray and white matter segmentation algorithm module is used to segment the brain gray and white matter of the region of interest using the level set method, to obtain the brain gray matter segmentation layer and the brain white matter segmentation layer, and to store the three-dimensional coordinate data of the points in the brain gray matter segmentation layer and the brain white matter segmentation layer. The cerebral cortex thickness measurement module is used to obtain the cortical thickness at the point of interest by calculating the minimum Euclidean distance based on the three-dimensional coordinate data of points within the gray matter segmentation layer and the white matter segmentation layer of the brain, combined with the three-dimensional coordinates of the point of interest.

Images