Cranial three-dimensional reconstruction method for low cranial pressure headache surgical navigation

Abstract

This invention relates to the field of three-dimensional reconstruction technology of the brain, specifically to a three-dimensional reconstruction method for brain surgery navigation in cases of low intracranial pressure headache. For any given angle, this invention obtains the local probability of interstitial gaps in each connected region based on the morphological features of each connected region in the MRI image and the morphological changes of adjacent connected regions. For MRI images at adjacent angles, it obtains the positional morphological similarity between connected regions based on their morphological features and positional distribution characteristics. Based on the positional morphological similarity between connected regions in MRI images at different adjacent angles, and the distribution of local probability of interstitial gaps in the connected regions, it obtains multiple sets of matching connected regions and the overall probability of interstitial gaps in each set of matching connected regions. Interstitial gap regions on MRI images at different angles are then selected for three-dimensional reconstruction. This invention obtains accurate three-dimensional reconstruction results by accurately marking tissue regions and gap regions between tissue regions.

Description

Technical Field

[0001] This invention relates to the field of three-dimensional brain reconstruction technology, and specifically to a three-dimensional brain reconstruction method for surgical navigation of low intracranial pressure headache. Background Technology

[0002] The navigation of low intracranial pressure headache surgery is assisted by neuronavigation technology. In the process of real-time three-dimensional reconstruction of the spatial position of surgical instruments and the patient's head based on cranial imaging data, due to the complexity and diversity of human brain tissue and the presence of grooves, it is necessary to accurately identify the gap regions between various tissues in the brain and accurately construct a three-dimensional model for neuronavigation technology.

[0003] In existing technologies, edge detection algorithms are used to identify the curves formed by edge curves and the boundary lines between adjacent regions as interstitial gap regions. However, this does not take into account the numerous grooves and crevices between brain tissues. Using edge detection algorithms alone may mistakenly identify groove regions as gap regions, leading to errors in the reconstruction of the 3D model and affecting the accuracy of neuronavigation technology. Summary of the Invention

[0004] To address the technical problem of mistaking groove areas for crevices, leading to errors in 3D model reconstruction and affecting navigation accuracy, this invention aims to provide a 3D cranial reconstruction method for navigation in low intracranial pressure headache surgery. The specific technical solution adopted is as follows:

[0005] This invention proposes a three-dimensional cranial reconstruction method for surgical navigation in cases of low intracranial pressure headache, the method comprising:

[0006] Acquire multi-angle MRI images of the patient's brain, wherein the images contain multiple connected components;

[0007] For any angle, based on the morphological characteristics of each connected region in the MRI image, the degree of narrowing and elongation of each connected region is obtained; based on the degree of narrowing and elongation of each connected region and the morphological change characteristics of adjacent connected regions, the local probability of interstitial gaps in each connected region is obtained.

[0008] For adjacent MRI images, the positional morphological similarity between connected domains is obtained based on the morphological features and positional distribution features of different connected domains; based on the positional morphological similarity between connected domains in different adjacent MRI images, and the local probability distribution of interstitial gaps in connected domains, multiple sets of matching connected domains and the overall probability of interstitial gaps in each set of matching connected domains are obtained.

[0009] Based on the overall probability of interstitial gaps in different groups of connected domains, interstitial gap regions on MRI images from different angles are selected for three-dimensional reconstruction.

[0010] Furthermore, the method for obtaining the degree of narrowness and elongation includes:

[0011] The maximum relative distance between two edge pixels in each connected component is selected as the longest connected component; perpendicular lines are drawn from the lines connecting the corresponding edge pixels, and the longest perpendicular line intercepted by the connected component is selected as the maximum width of the connected component.

[0012] The degree of narrowness and elongation is obtained by the ratio of the minimum circumcircle of each connected component to the number of pixels between connected components, as well as the difference between the longest length of the connected component and the maximum width of the connected component. Both the ratio of the number of pixels and the difference in the degree of narrowness and elongation are positively correlated with the degree of narrowness and elongation.

[0013] Furthermore, the method for obtaining the local probability of the interstitial gap includes:

[0014] The average number of pixels in all adjacent connected components of each connected component is used as the relative overall area.

[0015] The local probability of interstitial gaps for each connected region is obtained based on the narrowness and elongation of each connected region, and the first difference between the relative overall area and the number of pixels in each connected region. Both the narrowness and elongation and the first difference are positively correlated with the local probability of interstitial gaps.

[0016] Furthermore, the method for obtaining the positional morphological similarity includes:

[0017] For MRI images at adjacent angles, the ratio of the number of intersecting pixels between two connected regions to the relative distance between their centroids is obtained as the first similarity coefficient;

[0018] Obtain the first relative distance between the centroid of two connected components and the center point of the image; obtain the second relative distance between any corner of the image and the center point.

[0019] The positional morphological similarity between corresponding connected regions is obtained based on the angular difference between adjacent MRI images, the distance ratio between the first relative distance and the second relative distance between two connected regions, and the first similarity coefficient. The angular difference, the distance ratio, and the first similarity coefficient are all positively correlated with the positional morphological similarity.

[0020] Furthermore, the method for obtaining the positional morphological similarity includes:

[0021] Obtain the sinusoidal value of the angular difference between adjacent MRI images, and weight the sinusoidal value with the ratio between the first relative distance and the second relative distance as the adjustment coefficient;

[0022] The sum of the positive integer 1 and the adjustment coefficient is obtained as the adjustment weight; the product of the adjustment weight and the first similarity coefficient is obtained as the positional morphological similarity between connected components.

[0023] Furthermore, the method for obtaining the matched connected components includes:

[0024] For MRI images at adjacent angles, the connected regions with the highest positional and morphological similarity values ​​are selected, and the corresponding connected regions are used as matching connected regions to obtain multiple sets of matching connected regions for all adjacent MRI images.

[0025] Furthermore, the method for obtaining the overall probability of the interstitial gaps includes:

[0026] Based on the differences in positional and morphological similarity between MRI images at different adjacent angles for each set of matched connected domains, as well as the maximum and minimum values ​​of the local probability of interstitial gaps, the overall probability of interstitial gaps for each set of matched connected domains is obtained. The maximum and minimum values, as well as the differences in positional and morphological similarity, are all positively correlated with the overall probability of interstitial gaps.

[0027] Furthermore, the method for obtaining the overall probability of the interstitial gaps includes:

[0028] The cumulative difference in positional and morphological similarity between MRI images at different adjacent angles for each set of matched connected regions is obtained. The product of the cumulative difference, the maximum and minimum values ​​of the local probability of interstitial gaps in each set of matched connected regions is obtained and normalized to serve as the overall probability of interstitial gaps in each set of matched connected regions.

[0029] Furthermore, the method for obtaining the interstitial gap region includes:

[0030] If the overall probability of interstitial gaps in each set of matched connected regions is greater than or equal to a preset gap threshold, the corresponding matched connected regions on MRI images at different angles are taken as interstitial gap regions.

[0031] Furthermore, the preset gap threshold value is 0.75.

[0032] The present invention has the following beneficial effects:

[0033] This invention considers that interstitial gaps are typically narrow and elongated. For any angle, based on the morphological characteristics of each connected region in the MRI image, the degree of narrowness and elongation of each connected region is obtained. Based on the degree of narrowness and elongation of each connected region and the morphological changes of adjacent connected regions, the local probability of each connected region being an interstitial gap is obtained. By combining the morphology of the connected region with that of adjacent connected regions, the probability of the connected region being an interstitial gap is evaluated. For MRI images at adjacent angles, based on the morphological characteristics and positional distribution of different connected regions, the positional morphological similarity between connected regions is obtained, which helps in the analysis of subsequent connected region matching. Considering that the probability of a gap on a single image is unreliable, based on the positional morphological similarity between connected regions in MRI images at different adjacent angles, and the distribution of the local probability of interstitial gaps in the connected regions, multiple sets of matching connected regions and the overall probability of interstitial gaps in each set of matching connected regions are obtained, which helps in evaluating the overall probability of gaps between matching connected regions. Interstitial gap regions on MRI images at different angles are selected for three-dimensional reconstruction. This invention obtains accurate three-dimensional reconstruction results by accurately marking tissue regions and gap regions between tissue regions. Attached Figure Description

[0034] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 A flowchart illustrating a three-dimensional cranial reconstruction method for surgical navigation in cases of low intracranial pressure headache, provided as an embodiment of the present invention;

[0036] Figure 2 This is a flowchart illustrating a method for obtaining positional morphological similarity according to an embodiment of the present invention. Detailed Implementation

[0037] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a three-dimensional cranial reconstruction method for surgical navigation in cases of low intracranial pressure headache according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0039] The following describes in detail, with reference to the accompanying drawings, a specific scheme of the three-dimensional reconstruction method for cranial brain used in surgical navigation for low intracranial pressure headache provided by the present invention.

[0040] Please see Figure 1 The diagram illustrates a flowchart of a three-dimensional cranial reconstruction method for surgical navigation in cases of low intracranial pressure headache, provided by an embodiment of the present invention. The specific method includes:

[0041] Step S1: Obtain multi-angle MRI images of the patient's brain, which contain multiple connected regions.

[0042] In an embodiment of the present invention, considering the complex and diverse grooves and ridges within brain tissue, it is necessary to identify groove and ridge regions to avoid misidentifying them as crevices. First, before surgery for low intracranial pressure headache, an MRI scanner is used to perform MRI imaging on the patient's brain region, capturing multi-angle MRI images of the patient's brain region around the center point of the brain before surgery. The images contain multiple connected domains.

[0043] It should be noted that, in the embodiments of the present invention, the image is analyzed in a unified coordinate system, and an edge detection algorithm is used to detect the image to obtain the edge curves that constitute the connected domain. The specific means are well known to those skilled in the art and are not limited or described here.

[0044] Step S2: For any angle, based on the morphological characteristics of each connected region in the MRI image, obtain the degree of narrowing and elongation of each connected region; based on the degree of narrowing and elongation of each connected region and the morphological change characteristics of adjacent connected regions, obtain the local probability of interstitial gaps in each connected region.

[0045] The surface of tissues in the human brain has many grooves and they are highly interlocked. This may lead to misidentification of groove areas as gap areas, resulting in reconstruction errors. Gaps between tissues are usually narrow and elongated. Therefore, by analyzing the morphological characteristics of connected regions, the narrowness and elongation of connected regions can be quantified.

[0046] Preferably, in one embodiment of the present invention, the method for obtaining the degree of narrowness and elongation includes:

[0047] The maximum relative distance between two edge pixels in each connected component is selected as the longest connected component; perpendicular lines are drawn from the lines connecting the corresponding edge pixels, and the longest perpendicular line intercepted by the connected component is selected as the maximum width of the connected component.

[0048] The degree of narrowness and elongation is obtained by the ratio of the minimum circumcircle of each connected component to the number of pixels between connected components, as well as the difference between the longest length of the connected component and the maximum width of the connected component. Both the ratio of the number of pixels and the difference in the degree of narrowness and elongation are positively correlated with the degree of narrowness and elongation.

[0049] It should be noted that in some embodiments of the present invention, the relative distance between pixels is calculated by Euclidean distance or Manhattan distance. The specific means are well known to those skilled in the art and will not be described in detail here.

[0050] It should be noted that the number of pixels reflects the size of the area. The more pixels there are, the larger the area. The ratio of the number of pixels between the smallest circumcircle and the connected region reflects the compactness of the connected region. The larger the ratio, the larger the area of ​​the smallest circumcircle is relative to the connected region, and the less full the shape of the connected region is, and the narrower and more elongated it is.

[0051] The difference between the longest and widest connected component reflects its extensibility. A longer longest connected component is more likely to be long, while a smaller maximum width is more likely to be narrow. Therefore, a larger difference indicates a more concentrated distribution of pixels along the principal axis and a narrower distribution perpendicular to the principal axis, resulting in a more elongated and narrow connected component. Thus, both the ratio of the number of pixels and the difference in length are positively correlated with the degree of narrowness and elongation.

[0052] In one embodiment of the present invention, the product between the degree difference and the quantity ratio is obtained as the narrowness or elongation of each connected region; therefore, based on the above basic mathematical operations, a correlation is constructed between the degree difference, the quantity ratio and the narrowness or elongation, that is, the larger the degree difference, the larger the quantity ratio, and the greater the narrowness or elongation.

[0053] Interstitial gaps are typically narrow and elongated. The likelihood of an interstitial gap is quantified by analyzing the narrowness and elongation of connected regions. Considering that interstitial gap regions are usually located between tissue regions, and the area of ​​a tissue region is larger than that of a gap region, the morphological characteristics of adjacent connected regions are analyzed. The larger the area of ​​an adjacent connected region is relative to each connected region, the more likely it is to be a gap region. Therefore, based on the narrowness and elongation of each connected region and the morphological variation characteristics of adjacent connected regions, the local likelihood of an interstitial gap in each connected region is obtained.

[0054] Preferably, in one embodiment of the present invention, the method for obtaining the local probability of interstitial gaps includes:

[0055] The average number of pixels in all adjacent connected components of each connected component is used as the relative overall area.

[0056] The local probability of interstitial gaps for each connected region is obtained based on the narrowness and elongation of each connected region, and the first difference between the relative overall area and the number of pixels in each connected region. Both the narrowness and elongation and the first difference are positively correlated with the local probability of interstitial gaps.

[0057] It should be noted that, in the embodiments of the present invention, if the centroid connecting connected regions does not pass through other connected regions, the corresponding connected regions are considered to be adjacent connected regions.

[0058] It should be noted that the narrowness and elongation of each connected region helps assess the likelihood that the connected region is an interstitial gap. The narrower and longer the region, the more likely it is to be an interstitial gap region. The overall area level of all adjacent connected regions is quantified by the mean and compared with the area of ​​each connected region. That is, the larger the first difference between the relative overall area and the number of pixels in each connected region, the larger the relative overall area, the smaller the number of pixels in each connected region, and the more likely the connected region is to be an interstitial gap region if it is located in tissue larger than itself. Therefore, the narrowness and elongation and the first difference are both positively correlated with the local probability of an interstitial gap.

[0059] In one embodiment of the present invention, the narrowing and elongation of each connected region is normalized. That is, the region with the largest narrowing and elongation among all connected regions is selected, and the ratio of the narrowing and elongation of each connected region to the maximum narrowing and elongation is calculated. The greater the narrowing and elongation of each connected region, the greater the normalization result. The product of the normalization result and the first difference between the relative overall area and the number of pixels in each connected region is calculated as the local probability of interstitial gaps in each connected region. Therefore, based on the above basic mathematical operations, a correlation is constructed between the narrowing and elongation of each connected region, the first difference between the relative overall area and the number of pixels in each connected region, and the local probability of interstitial gaps. That is, the greater the narrowing and elongation of each connected region and the greater the first difference, the greater the local probability of interstitial gaps.

[0060] It should be noted that, in the embodiments of the present invention, normalization can be performed by linear normalization or a normalization function. The specific means are well known to those skilled in the art and will not be described in detail here.

[0061] Step S3: For adjacent MRI images, obtain the positional morphological similarity between connected regions based on the morphological features and positional distribution features of different connected regions; based on the positional morphological similarity between connected regions in MRI images of different adjacent angles, and the local probability distribution of interstitial gaps in connected regions, obtain multiple sets of matching connected regions and the overall probability of interstitial gaps in each set of matching connected regions.

[0062] The human brain region is relatively small, and the morphological changes of the same region at different angles vary greatly due to the large angle changes. The linear displacement of the region near the center of the image is small. For MRI images at adjacent angles, the positional morphological similarity between connected regions is obtained based on the morphological characteristics and positional distribution characteristics of different connected regions.

[0063] Preferably, in one embodiment of the present invention, the method for obtaining positional morphological similarity is described in [reference needed]. Figure 2 It shows a flowchart of a method for obtaining positional morphological similarity, including:

[0064] Step S201: For adjacent angle MRI images, obtain the ratio of the number of intersecting pixels between two connected regions to the relative distance between their centroids, and use it as the first similarity coefficient.

[0065] It should be noted that the images are distributed in a unified coordinate system. Intersecting pixels between connected components represent pixels with the same coordinates. The number of intersecting pixels reflects the morphological similarity of the connected components. The more intersecting pixels, the more similar the connected components are in shape, and the larger the first similarity coefficient. The relative distance between centroids reflects the degree of spatial separation and the relative positional relationship between connected components. The larger the relative distance between centroids, the farther apart the connected components are in relative position, the worse the morphological correlation, and the smaller the first similarity coefficient.

[0066] It should be noted that in some embodiments of the present invention, the centroid in the image is calculated by the irregular shape centroid calculation method, and the relative distance between the centroids is calculated by Euclidean distance or Manhattan distance. The specific means are well known to those skilled in the art and will not be described in detail here.

[0067] Step S202: Obtain the first relative distance between the centroid of two connected components and the center point of the image, and obtain the second relative distance between any corner of the image and the center point.

[0068] The relative distance between the centroid and the image center reflects the positional offset of the connected domain relative to the image center. The smaller the relative distance, the closer the connected domain is to the central region of the image. The relative distance between any corner of the image and the center point reflects the scale property of the image itself, which helps to normalize other spatial measurements within the image and eliminate the influence of changes in image size and scale.

[0069] Step S203: Based on the angular difference between adjacent MRI images, the distance ratio between the first relative distance and the second relative distance between two connected regions, and the first similarity coefficient, the positional morphological similarity between corresponding connected regions is obtained. The angular difference, the distance ratio, and the first similarity coefficient are all positively correlated with the positional morphological similarity.

[0070] It should be noted that, in the embodiments of the present invention, the difference is the absolute value of the calculated difference.

[0071] It should be noted that the ratio of the first relative distance to the second relative distance reflects the distance of the connected domain from the image center. The larger the ratio, the larger the first relative distance, the farther the centroid of the connected domain is from the image center, and the greater the influence of angular rotation, the more the influence of angular changes needs to be considered. The angular difference between adjacent MRI images reflects the degree of deformation between the images. The larger the angular difference, the greater the change in appearance, shape, and position of the same anatomical structure in adjacent images, and the greater the influence of angular changes. The judgment of similarity can be relatively lenient, and the larger the weight of the contrast value, the greater the morphological similarity. The larger the first similarity coefficient, the greater the positional and morphological similarity of the connected domain. Therefore, the angular difference between adjacent MRI images, the ratio of the first relative distance to the second relative distance, and the first similarity coefficient are all positively correlated with positional and morphological similarity.

[0072] Preferably, in one embodiment of the present invention, the method for obtaining positional morphological similarity includes: obtaining the sine value of the angular difference between adjacent MRI images; weighting the distance ratio between the first relative distance and the second relative distance as an adjustment coefficient; obtaining the sum of the positive integer 1 and the adjustment coefficient as an adjustment weight; and obtaining the product between the adjustment weight and the first similarity coefficient as the positional morphological similarity between connected components.

[0073] Therefore, based on the above basic mathematical operations, the correlation between the angular difference between adjacent MRI images, the distance ratio between the first relative distance and the second relative distance, and the first similarity coefficient and the positional morphological similarity is constructed. That is, the greater the angular difference between adjacent MRI images, the greater the distance ratio between the first relative distance and the second relative distance, the greater the first similarity coefficient, and the greater the positional morphological similarity.

[0074] The complex internal structure of the human brain region leads to a high probability of significant differences in the appearance of the same tissue region in MRI images from different angles. Directly relying on the morphological similarity of tissues in different images is insufficient to accurately match the same tissue region. Therefore, considering that at certain angles, the grooves and ridges of the tissue region are dense and small, which may be mistaken for gaps, the probability that the same tissue region belongs to the target region varies greatly in different images. Moreover, compared to the tissue region being a single, continuous area, the gaps between tissues are often smaller and vary greatly. Based on the positional and morphological similarity between connected domains in MRI images from different adjacent angles, and the local probability distribution of gaps between tissues in the connected domains, multiple sets of matching connected domains and the overall probability of gaps between tissues in each set of matching connected domains are obtained.

[0075] Preferably, in one embodiment of the present invention, the method for obtaining the matched connected components includes:

[0076] For MRI images at adjacent angles, the connected regions with the highest positional and morphological similarity values ​​are selected, and the corresponding connected regions are used as matching connected regions to obtain multiple sets of matching connected regions for all adjacent MRI images.

[0077] To give an example, if the positional and morphological similarity between connected component A in image 1 and connected component B in image 2 is the largest, then connected component A in image 1 and connected component B in image 2 match. If there is a match between connected component B in image 2 and connected component C in image 3, then ABC is a set of matching connected components, representing the same region within the brain.

[0078] Preferably, in one embodiment of the present invention, the method for obtaining the overall probability of interstitial gaps includes:

[0079] Based on the differences in positional and morphological similarity between MRI images at different adjacent angles for each set of matched connected domains, as well as the maximum and minimum values ​​of the local probability of interstitial gaps, the overall probability of interstitial gaps for each set of matched connected domains is obtained. The maximum and minimum values, as well as the differences in positional and morphological similarity, are all positively correlated with the overall probability of interstitial gaps.

[0080] It should be noted that the difference in positional and morphological similarity between each set of matched connected regions in MRI images at different adjacent angles reflects the magnitude of change of the matched connected regions in different images. The greater the difference, the more inconsistent the positional and morphological similarity, the greater the magnitude of change, and the greater the overall probability of interstitial gaps. The maximum and minimum values ​​of the local probability of interstitial gaps reflect the closeness of the local probability of interstitial gaps in the matched connected regions on different images. The larger the maximum and minimum values, the greater the probability that the connected region is a gap. Therefore, the maximum and minimum values, as well as the differences in positional and morphological similarity, are all positively correlated with the overall probability of interstitial gaps.

[0081] In one embodiment of the present invention, the cumulative difference in positional and morphological similarity between MRI images at different adjacent angles for each set of matched connected regions is obtained; the product of the cumulative difference, the maximum and minimum values ​​of the local probability of interstitial gaps in each set of matched connected regions is obtained and normalized to serve as the overall probability of interstitial gaps in each set of matched connected regions; therefore, based on the above basic mathematical operations, a correlation is constructed between the maximum and minimum values, the difference in positional and morphological similarity, and the overall probability of interstitial gaps, i.e., the larger the maximum and minimum values, the greater the difference in positional and morphological similarity, and the greater the overall probability of interstitial gaps.

[0082] Step S4: Based on the overall probability of interstitial gaps in different groups of connected domains, filter out interstitial gap regions on MRI images from different angles and perform three-dimensional reconstruction.

[0083] Preferably, the overall probability of interstitial gaps reflects the confidence level of real, continuous gaps rather than random noise or artifacts. The higher the overall probability of interstitial gaps, the more realistic the gaps are, thus avoiding misjudgment. In one embodiment of the present invention, the method for obtaining the interstitial gap region includes:

[0084] If the overall probability of interstitial gaps in each set of matched connected regions is greater than or equal to a preset gap threshold, the corresponding matched connected regions on MRI images at different angles are taken as interstitial gap regions.

[0085] It should be noted that, in one embodiment of the present invention, the preset gap threshold is set to 0.75; in other embodiments of the present invention, the preset gap threshold can be set according to specific circumstances, and will not be limited or elaborated here.

[0086] Based on this, the interstitial gap regions on images from different angles were selected and marked, and a voxel-based 3D reconstruction method was used to reconstruct the images in 3D, effectively improving the accuracy of neural navigation.

[0087] In summary, this invention, for any angle, obtains the local probability of interstitial gaps in each connected region based on the morphological features of each connected region in the MRI image and the morphological changes of adjacent connected regions; for MRI images at adjacent angles, it obtains the positional morphological similarity between connected regions based on the morphological features and positional distribution characteristics of different connected regions; based on the positional morphological similarity between connected regions in MRI images at different adjacent angles, and the distribution of local probability of interstitial gaps in connected regions, it obtains multiple sets of matching connected regions and the overall probability of interstitial gaps in each set of matching connected regions; and it filters out interstitial gap regions on MRI images at different angles for three-dimensional reconstruction. This invention obtains accurate three-dimensional reconstruction results by accurately marking tissue regions and gap regions between tissue regions.

[0088] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0089] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for three-dimensional brain reconstruction for surgical navigation in cases of low intracranial pressure headache, characterized in that, The method includes: Acquire multi-angle MRI images of the patient's brain, wherein the images contain multiple connected components; For any angle, based on the morphological characteristics of each connected region in the MRI image, the degree of narrowing and elongation of each connected region is obtained; based on the degree of narrowing and elongation of each connected region and the morphological change characteristics of adjacent connected regions, the local probability of interstitial gaps in each connected region is obtained. For adjacent MRI images, the positional morphological similarity between connected domains is obtained based on the morphological features and positional distribution features of different connected domains; based on the positional morphological similarity between connected domains in different adjacent MRI images, and the local probability distribution of interstitial gaps in connected domains, multiple sets of matching connected domains and the overall probability of interstitial gaps in each set of matching connected domains are obtained. Based on the overall probability of interstitial gaps in different groups of matched connected domains, interstitial gap regions on MRI images from different angles are selected and three-dimensional reconstruction is performed. The method for obtaining the local probability of interstitial gaps includes: The average number of pixels in all adjacent connected components of each connected component is used as the relative overall area. The local probability of interstitial gaps for each connected region is obtained based on the narrowness and elongation of each connected region, and the first difference between the relative overall area and the number of pixels in each connected region. Both the narrowness and elongation and the first difference are positively correlated with the local probability of interstitial gaps.

2. The method for three-dimensional cranial reconstruction for surgical navigation in cases of low intracranial pressure headache according to claim 1, characterized in that, The method for obtaining the degree of narrowness and elongation includes: The longest connected component is the one with the largest relative distance between two edge pixels in each connected component. Draw perpendicular lines connecting the corresponding edge pixels, and select the longest perpendicular line intercepted by the connected component as the maximum width of the connected component. The degree of narrowness and elongation is obtained by the ratio of the minimum circumcircle of each connected component to the number of pixels between connected components, as well as the difference between the longest length of the connected component and the maximum width of the connected component. Both the ratio of the number of pixels and the difference in the degree of narrowness and elongation are positively correlated with the degree of narrowness and elongation.

3. The method for three-dimensional brain reconstruction for surgical navigation in cases of low intracranial pressure headache according to claim 1, characterized in that, The method for obtaining the positional morphological similarity includes: For MRI images at adjacent angles, the ratio of the number of intersecting pixels between two connected regions to the relative distance between their centroids is obtained as the first similarity coefficient; Obtain the first relative distance between the centroid of two connected components and the center point of the image; obtain the second relative distance between any corner of the image and the center point. The positional morphological similarity between corresponding connected regions is obtained based on the angular difference between adjacent MRI images, the distance ratio between the first relative distance and the second relative distance between two connected regions, and the first similarity coefficient. The angular difference, the distance ratio, and the first similarity coefficient are all positively correlated with the positional morphological similarity.

4. The method for three-dimensional brain reconstruction for surgical navigation in cases of low intracranial pressure headache according to claim 3, characterized in that, The method for obtaining the positional morphological similarity includes: Obtain the sinusoidal value of the angular difference between adjacent MRI images, and weight the sinusoidal value with the ratio between the first relative distance and the second relative distance as the adjustment coefficient; The sum of the positive integer 1 and the adjustment coefficient is obtained as the adjustment weight; the product of the adjustment weight and the first similarity coefficient is obtained as the positional morphological similarity between connected components.

5. A method for three-dimensional brain reconstruction for surgical navigation in cases of low intracranial pressure headache, as described in claim 1, characterized in that... The method for obtaining the matched connected components includes: For MRI images at adjacent angles, the connected regions with the highest positional and morphological similarity values ​​are selected, and the corresponding connected regions are used as matching connected regions to obtain multiple sets of matching connected regions for all adjacent MRI images.

6. A method for three-dimensional cranial reconstruction for surgical navigation in cases of low intracranial pressure headache, as described in claim 1, characterized in that, The method for obtaining the overall probability of interstitial gaps includes: Based on the differences in positional and morphological similarity between MRI images at different adjacent angles for each set of matched connected domains, as well as the maximum and minimum values ​​of the local probability of interstitial gaps, the overall probability of interstitial gaps for each set of matched connected domains is obtained. The maximum and minimum values, as well as the differences in positional and morphological similarity, are all positively correlated with the overall probability of interstitial gaps.

7. A method for three-dimensional cranial reconstruction for surgical navigation in cases of low intracranial pressure headache, as described in claim 6, is characterized in that... The method for obtaining the overall probability of interstitial gaps includes: The cumulative difference in positional and morphological similarity between MRI images at different adjacent angles for each set of matched connected regions is obtained. The product of the cumulative difference, the maximum and minimum values ​​of the local probability of interstitial gaps in each set of matched connected regions is obtained and normalized to serve as the overall probability of interstitial gaps in each set of matched connected regions.

8. A method for three-dimensional brain reconstruction for surgical navigation in cases of low intracranial pressure headache, as described in claim 1, characterized in that... The method for obtaining the interstitial gap region includes: If the overall probability of interstitial gaps in each set of matched connected regions is greater than or equal to a preset gap threshold, the corresponding matched connected regions on MRI images at different angles are taken as interstitial gap regions.

9. A method for three-dimensional brain reconstruction for surgical navigation in cases of low intracranial pressure headache, as described in claim 8, characterized in that... The preset gap threshold value is 0.75.

Images