Method for calibrating cortical electrode positions based on CT images and electrode positioning system

Abstract

The present invention belongs to the technical field of CT image processing, and particularly relates to a method for calibrating the position of cortical electrodes based on CT images and an electrode positioning system. By making full use of the electrode contact position information provided by CT images and combining the geometric properties of electrode patches, a non-parametric surface is fitted in the PCA space of each electrode, and the geometric constraints between electrode contacts are utilized to ensure the electrode positioning accuracy. On the premise of only relying on a small amount of manual annotation and information input, the calculation speed and calibration accuracy of the automatic calibration method are greatly improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of CT image processing, and in particular relates to a method for calibrating cortical electrode positions based on CT images and an electrode positioning system. Background Art

[0002] Clinically, for the diagnosis of neurological diseases such as epilepsy and glioma, it is necessary to implant intracranial cortical electrodes to identify epileptic foci and guide subsequent surgical resection. The accuracy of intracranial electrode position calibration determines the benefits of subsequent surgery. At present, the calibration of intracranial cortical electrodes mainly relies on manual annotation under CT images. The number of cortical electrode contacts is large, especially high-density cortical electrodes, which can reach 256 contacts. It takes a long time to annotate electrode contacts in three-dimensional CT images, and the spacing between cortical electrodes is small. The electrode boundaries are often difficult to distinguish in CT images, resulting in large manual annotation errors. In related technologies, automatic or semi-automatic cortical electrode calibration algorithms are generally used. For example, patent number CN116473565A discloses an image-based intracranial electrode positioning method, system, computer equipment and medium, which performs image registration of CT three-dimensional images and MRI images, obtains spatial transformation parameters generated during the registration process, and generates a transformation matrix through SPM8 software according to the spatial transformation parameters; the coordinates of the contact center point are converted into the electrode contact coordinates in the MRI image through the transformation matrix, and the electrode contact coordinates in the MRI image are visualized. This technical solution is to first determine the coordinate value of the electrode contact through the threshold method, and then use the spatial transformation parameters to match the CT three-dimensional image with the MRI image to obtain the electrode coordinates. Because the CT value of the electrode contact is often close to the skull, and the CT image contains various interference factors such as cables and bone screws, this technical solution cannot accurately locate the contact. Summary of the invention

[0003] The invention provides a method for calibrating the position of cortical electrodes based on CT images and an electrode positioning system to improve the positioning accuracy of intracranial electrodes.

[0004] In order to solve the above technical problems, the present invention provides a method for calibrating the position of intracranial electrodes based on CT images, comprising: preprocessing, separating a single electrode sheet in the CT image; establishing a PCA space coordinate system, establishing an original coordinate system based on the CT image and projecting each electrode sheet into the PCA space coordinate system; calibrating the position coordinates of the electrode contacts in the PCA space coordinate system; and inverting the output, inversely transforming the position coordinates of the electrode contacts to the original coordinate system and outputting them.

[0005] Further, calibrating the position coordinates of the electrode contacts in the PCA space coordinate system includes: fitting the non-parametric electrode sheet surface in the PCA space coordinate system; iteratively constructing a convex hull on the PCA two-dimensional plane to obtain the corner coordinates of the electrode sheet; obtaining the unstretched electrode grid points based on the corner coordinates of the electrode sheet; projecting the electrode grid points onto the non-parametric electrode sheet surface to obtain the deformed grid points; and correcting the positions of the grid points based on the neighborhood centroid correction method to obtain the position coordinates of the electrode contacts.

[0006] Further, the non-parametric electrode sheet surface is obtained by fitting using the locally linear embedding method in the PCA space coordinate system, including: finding the local neighborhood for each point in the PCA space; constructing the local linear basis function and fitting the local linear model by the least squares method; repeating the search for the local neighborhood and calculating the weighted regression coefficients on the new input points to obtain the predicted values, so as to achieve the smooth fitting of the non-parametric electrode sheet surface.

[0007] Further, the iterative construction of the convex hull includes: constructing a convex hull on the two-dimensional plane based on the input point set; adjusting the boundary points of the convex hull to maintain the preset shape; identifying and removing the points at the maximum curvature by angle calculation to gradually streamline the convex hull; repeating the above operations until only four key corner points remain in the convex hull, which are used as the corner coordinates of the electrode sheet.

[0008] Further, obtaining the unstretched electrode grid points based on the corner coordinates of the electrode sheet includes: obtaining the unstretched electrode grid points based on the corner coordinates of the electrode sheet includes: using the three-dimensional intersection method to obtain the unstretched electrode grid points according to the number of rows and columns of the electrode sheet and the corner coordinates of the electrode sheet; the three-dimensional intersection method includes: sorting the four corner points of the electrode sheet first; calculating the reference vectors in the row and column directions according to the sorted corner points and correcting the number of rows and columns; generating the candidate segments of the grid in the row and column directions respectively according to the principle of equal distance splitting on both sides; performing three-dimensional geometric intersection operations on the corresponding candidate segments in the row and column directions to obtain the intersection coordinates of each grid unit; and combining all the intersection coordinates into an electrode grid point array.

[0009] Further, the neighborhood centroid correction method includes: obtaining the CT voxels in the local neighborhood of each grid point to calculate the centroid points of the CT voxels; correcting the grid points to the corresponding centroid points and obtaining the position coordinates of the electrode contacts in the current state.

[0010] Further, the preprocessing includes: separating the skull and the electrode sheet from the CT image by threshold calculation; setting the number of electrode sheets to separate each electrode sheet using the clustering method; generating a bounding box based on the clustering result and dividing out a single electrode sheet using the bounding box.

[0011] In a second aspect, the present invention provides an electrode positioning system, comprising: a processor that executes the computer program to implement the steps of the method; an image acquisition module that is configured to acquire a three-dimensional CT image after electrode implantation; and a contact point coordinate determination module that is configured to determine electrode contacts in the three-dimensional CT image and extract the position coordinates of the electrode contacts.

[0012] In a third aspect, the present invention provides a computer device, comprising a memory, a processor, and a computer program stored on the memory, wherein the processor executes the computer program to implement the steps of the method.

[0013] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method are implemented.

[0014] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, and when the computer program is executed by a processor, the steps of the method are implemented.

[0015] The beneficial effects of the present invention are as follows: The method for calibrating the position of cortical electrodes based on CT images of the present invention makes full use of the electrode contact position information provided by CT images, combines the geometric properties of electrode patches, fits a non-parametric surface in the PCA space of each electrode, and utilizes the geometric constraints between electrode contacts to ensure the electrode positioning accuracy. On the premise of only relying on a small amount of manual annotation and information input, the calculation speed and calibration accuracy of the automatic calibration method are greatly improved.

[0016] Other features and advantages of the present invention will be described in the following specification, and in part, will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention are achieved and obtained by the structures specifically pointed out in the specification, claims, and drawings.

[0017] To make the above objectives, features, and advantages of the present invention more obvious and understandable, the following specific preferred embodiments are given, and in conjunction with the accompanying drawings, the detailed description is as follows. Description of the Drawings

[0018] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 It is an operation flow chart for calibrating the position of intracranial electrodes provided by some embodiments.

[0020] Figure 2 Schematic diagrams of CT images after implanting intracranial electrodes provided by some embodiments.

[0021] Figure 3 Effect diagrams of manually adjusting the threshold to separate the skull and CT electrodes provided by some embodiments.

[0022] Figure 4 Schematic diagrams of obtaining a single electrode sheet in the original coordinate system based on CT images provided by some embodiments.

[0023] Figure 5 Schematic diagrams of a single electrode sheet in the PCA space coordinate system provided by some embodiments.

[0024] Figure 6 Schematic diagrams of fitting a non-parametric electrode sheet surface for a single electrode sheet in the PCA space coordinate system provided by some embodiments.

[0025] Figure 7 Schematic diagrams of iteratively constructing a convex hull for a single electrode sheet on the PCA two-dimensional plane provided by some embodiments.

[0026] Figure 8 Schematic diagrams of forming a non-stretched electrode lattice for a single electrode sheet provided by some embodiments.

[0027] Figure 9 Schematic diagrams of the position coordinates of electrode contacts in a single electrode sheet provided by some embodiments. Detailed implementation manners

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0029] See Figures 1-9 , this embodiment provides a method for calibrating the position of intracranial electrodes based on CT images, including: Step S1, preprocessing, separating a single electrode sheet in the CT image; Step S2, establishing a PCA space coordinate system, establishing an original coordinate system based on the CT image and projecting each electrode sheet onto the PCA space coordinate system; Step S3, calibrating the position coordinates of electrode contacts in the PCA space coordinate system; Step S4, inverse transformation output, inverse-transforming the position coordinates of the electrode contacts to the original coordinate system and outputting. Specifically, the operations of each step can be implemented by means of a computer processor and its corresponding functional software, and the specific operation process is as follows.

[0030] Step S1, preprocessing.

[0031] See Figure 2 and Figure 3 , the preprocessing includes: separating the skull and electrode patches from the CT image through threshold calculation; setting the number of electrode patches to separate each electrode patch using a clustering method; generating a bounding box based on the clustering result, and using the bounding box to divide individual electrode patches. Specifically, the preprocessing part needs to separate individual electrode patches. First, it is necessary to separate the skull and CT electrodes from the CT image (as shown in Figure 2 ) through threshold calculation and achieve the best separation effect through manual adjustment of the threshold (as shown in Figure 3 ), retaining all electrode contacts as much as possible while suppressing most of the noise. Manually set the number of electrode patches, separate each electrode patch using a clustering method, generate a bounding box based on automatic clustering, and the bounding box can be manually adjusted and optimized, and individual electrode patches are divided using the bounding box. Then, an original coordinate system of individual electrode patches is established based on the CT image (as shown in Figure 4 ).

[0032] In this case, CT is Computed Tomography, and PCA is Principal Component Analysis. The CT image is, for example but not limited to, the CT image of cortical electrodes after implantation.

[0033] Step S2, establishing a PCA space coordinate system.

[0034] See Figure 5 , establishing a PCA space coordinate system includes: calculating the PCA principal components of the electrode points separated from the CT image and performing projections to obtain the corresponding PCA space coordinate system.

[0035] Optionally, for the PCA space coordinate system, principal component analysis PCA is a technique for simplifying data sets. It is a linear transformation. This transformation transforms the data into a new coordinate system such that the first largest variance of any data projection is on the first coordinate (referred to as the first principal component), the second largest variance is on the second coordinate (the second principal component), and so on. As shown in Figure 5 , the main axes PC1, PC2, and PC3 of PCA are the three projection directions, and their variances are arranged from large to small.

[0036] Step S3, calibrating the position coordinates of electrode contacts.

[0037] See Figures 6-9 , calibrating the position coordinates of electrode contacts in the PCA space coordinate system includes but is not limited to the following steps:

[0038] Step S31: Use the locally linear embedding method to fit the non-parametric electrode sheet surface in the PCA space coordinate system. See Figure 6 , specifically including:

[0039] (1) For each point in the PCA space, find its local neighborhood (as shown by the blue points in Figure 6 );

[0040] (2) Construct local linear basis functions and fit the local linear model by the least squares method;

[0041] (3) Repeat searching for local neighborhoods and calculating weighted regression coefficients at the new input points to obtain predicted values, thereby achieving smooth fitting of the overall surface (as shown by the yellow surface in Figure 6 ).

[0042] Step S32: Iteratively construct a convex hull on the PCA two-dimensional plane to obtain the corner coordinates of the electrode sheet. See Figure 7 , specifically including:

[0043] (1) Based on the input point set, construct a convex hull on the two-dimensional plane, that is, Figure 7 the view corresponding to iteration 1 in

[0044] (2) Adjust the boundary points of the convex hull to maintain a reasonable shape (as shown by the red wireframes in iteration 1, iteration 7, and iteration 11 in Figure 7 ), such as using local extreme values to correct the positions of the boundary points;

[0045] (3) Identify and remove the points at the maximum curvature by angle calculation, thereby successively refining the convex hull;

[0046] (4) Repeat the above operations, iteratively excluding the points with the largest angle among the angles formed by adjacent points in the convex hull (as shown by the views corresponding to iteration 7 and iteration 11 in Figure 7 ), until only four key corner points remain in the obtained convex hull, and finally output the corner positions of the target rectangle as the corner coordinates of the electrode sheet (as shown by the red points in the final result view in Figure 7 ).

[0047] Step S33: Obtain the non-stretched electrode grid points based on the corner coordinates of the electrode sheet. See Figure 8 , specifically including:

[0048] (1) First, sort the four corner points obtained in step S32 to ensure that they are arranged in a clockwise (or counterclockwise) order;

[0049] (2) Calculate the reference vectors in the row and column directions according to the sorted corner points, and perform adaptive correction on the corresponding number of rows and columns;

[0050] (3) Generate candidate segments of the grid in the row and column directions respectively according to the principle of equal-distance splitting on both sides.

[0051] (4) Obtain the intersection coordinates of each grid cell through three-dimensional geometric intersection operations on the corresponding candidate segments in the row and column directions.

[0052] (5) Combine all the intersection points to form a complete homogeneous electrode lattice array (as shown by the red points in Figure 8 ).

[0053] Step S34: Map the complete homogeneous electrode lattice array obtained in step S33 according to the non-parametric electrode sheet surface fitted in step S31 to obtain the deformed grid lattice points.

[0054] Step S35: Correct the positions of the grid lattice points based on the neighborhood centroid correction method to obtain the position coordinates of the electrode contacts. See Figure 9 , specifically including:

[0055] (1) Obtain the CT voxels within the neighborhood range of the set electrode spacing for each deformed grid lattice point in step S34 to calculate the centroid points of the CT voxels.

[0056] (2) Correct the grid lattice points to the corresponding centroid points and obtain the position coordinates of the electrode contacts in the current state (as shown by the red points in Figure 9 ).

[0057] By fitting the PCA space of the thresholded voxel points and performing projection operations in the PCA space, the position deviation brought by subsequent projection onto the cortical electrode surface can be minimized, improving the accuracy of electrode position; by using the non-parametric local linear fitting method, electrode surfaces of any complexity can be accurately fitted, ensuring the positioning accuracy of subsequent electrode points; by using the neighborhood centroid correction method, the electrode site information provided by the CT image can be fully utilized to improve the accuracy of electrode point calibration while ensuring geometric constraints. Combining the introduction of the PCA space of the electrode and the non-parametric surface fitting of the electrode sheet effectively reduces the error when the electrode is projected onto the distorted surface, and using the CT image information for correction effectively reduces the positioning error of the final electrode points. Compared with the prior art, the method for calibrating electrode contacts in this case makes full use of the CT image information and the geometric properties of the electrode grid itself, without increasing the additional burden of manual annotation, having a smaller computational burden, faster processing speed, and significantly improving the electrode annotation efficiency. At the same time, the calculation process is stable, the results are highly accurate and consistent, significantly improving the accuracy of electrode positioning.

[0058] In addition, step 31 is based on operations in the PCA space coordinate system, while step 32 is an operation in the PCA two-dimensional plane. Therefore, there is no sequential relationship between the two in principle. It is only necessary to combine the non-parametric electrode sheet surface fitted in step S31 for mapping after obtaining the complete homogeneous electrode lattice array in step S33 to complete the operation of step S34 and obtain the deformed grid lattice points.

[0059] Step S4, inverse transformation output, inverse-transforms the position coordinates of the electrode contacts to the original coordinate system and outputs them.

[0060] In some embodiments, the numerical values of the coordinate axes in each coordinate system are dimensions and do not represent the actual positions and actual distances of the electrode contacts.

[0061] In some embodiments, an electrode positioning system is provided, including: a processor that executes the computer program to implement the steps of the method; an image acquisition module for acquiring a three-dimensional CT image after electrode implantation; and a contact coordinate determination module for determining the electrode contacts in the three-dimensional CT image and extracting the position coordinates of the electrode contacts.

[0062] Of course, the electrode positioning system in this case also includes a data calculation module, a man-machine interaction machine, and its operation or display interface to facilitate operating parameters, inputting program instructions, etc. For example, manually setting the number of electrode sheets, manually adjusting and optimizing the bounding box, and achieving the best separation effect by manually adjusting the threshold.

[0063] In some embodiments, a computer device is provided, including a memory, a processor, and a computer program stored on the memory, and the processor executes the computer program to implement the steps of the method.

[0064] In some embodiments, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method are implemented.

[0065] In some embodiments, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps of the method are implemented.

[0066] In the above embodiments, if the function or model is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0067] In the description of the embodiments of the present invention, unless otherwise clearly specified and limited, the terms "installation", "connection", and "coupling" shall be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0068] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention. In addition, the terms "first", "second", and "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0069] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical functional division, and there may be other division methods in actual implementation. Also, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0070] Based on the above-mentioned ideal embodiments of the present invention as inspiration, through the above description, relevant staff can completely make various changes and modifications without departing from the technical idea of this invention. The technical scope of this invention is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A method for calibrating intracranial electrode positions based on CT images, characterized in that: include: Preprocessing, separating individual electrode patches in CT images; Establishing a PCA space coordinate system, establishing an original coordinate system based on the CT image and projecting each electrode patch into the PCA space coordinate system; Calibrate the position coordinates of the electrode contacts in the PCA space coordinate system; Invert output, inversely transform the position coordinates of the electrode contact to the original coordinate system and output; The position coordinates of the electrode contacts are calibrated in the PCA space coordinate system: Fitting the non-parametric electrode patch surface in the PCA space coordinate system; Iteratively construct the convex hull on the PCA two-dimensional plane to obtain the coordinates of the corner points of the electrode sheet; The electrode grid without stretching is obtained based on the coordinates of the corner points of the electrode sheet; Projecting the electrode grid points onto the non-parametric electrode sheet surface to obtain the deformed grid grid points; Correcting the positions of the grid points based on the neighborhood centroid correction method to obtain the position coordinates of the electrode contacts; The non-parametric electrode sheet surface is a distorted surface, which is obtained by fitting using a local linear embedding method in a PCA space coordinate system, including: Find the local neighborhood for each point in the PCA space; Construct local linear basis functions and fit local linear models by least squares method; The local neighborhood is searched repeatedly on the new input point, and the weighted regression coefficient is calculated to obtain the predicted value, so as to achieve smooth fitting of the non-parametric electrode surface.

2. The method according to claim 1, characterized in that The iterative construction of the convex hull comprises: Based on the input point set, construct a convex hull on the two-dimensional plane; Adjust the convex hull boundary points to maintain the preset shape; The points with the maximum curvature are identified and eliminated through angle calculation to gradually simplify the convex hull; The above operation is repeated until only four key corner points remain in the convex hull, which serve as the corner point coordinates of the electrode sheet.

3. The method according to claim 1, characterized in that Obtaining the non-stretched electrode grid points based on the corner point coordinates of the electrode sheet includes: obtaining the non-stretched electrode grid points using a three-dimensional intersection method according to the number of rows and columns of the electrode sheet and the corner point coordinates of the electrode sheet; The three-dimensional intersection method comprises: Sort the four corner points of the electrode sheet first; Calculate the reference vectors in the row and column directions according to the sorted corner points, and correct the number of rows and columns; According to the principle of equidistant splitting on both sides, candidate segments of the grid in the row and column directions are generated respectively; By performing a three-dimensional geometric intersection operation on the candidate segments corresponding to the row and column directions, the intersection coordinates of each grid unit are obtained; The coordinates of all intersection points are combined into an electrode grid array.

4. The method according to claim 1, characterized in that: The neighborhood centroid correction method includes: Obtaining CT voxels in a local neighborhood of each grid point to calculate the centroid point of the CT voxel; Correct the grid points to the corresponding centroid points and obtain the position coordinates of the electrode contacts in the current state.

5. The method according to claim 1, characterized in that The pre-processing comprises: Separate the skull and electrodes from the CT images by threshold calculation; Setting the number of electrode sheets to separate the individual electrode sheets using a clustering method; A bounding box is generated based on the clustering results, and a single electrode patch is divided using the bounding box.

6. An electrode positioning system, characterized in that: include: A processor, executing the computer program to implement the steps of the method according to any one of claims 1 to 5; An image acquisition module, used for acquiring CT three-dimensional images after electrode implantation; The contact point coordinate determination module is used to determine the electrode contact points in the CT three-dimensional image and extract the position coordinates of the electrode contact points.

7. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images