Artificial cochlea implantation grinding path planning method, storage medium and equipment

Abstract

The invention discloses an artificial cochlea implantation grinding path planning method, a storage medium and equipment, and belongs to the technical field of path planning in artificial cochlea implantation. The problems that an existing grinding path planning method is low in efficiency, path deviation is likely to be introduced, and the individualized adaptive capacity of a planned path is poor are solved. According to the method, the preoperative CT image is preprocessed by fusing the anatomical structure model, and the key structure in the temporal bone can be accurately positioned in the preoperative CT image, so that the three-dimensional space geometric information of the key structures of the cochlea, the vestibule and the facial nerve is extracted. According to the method, the grinding path of the artificial cochlea operation area is automatically planned in the three-dimensional space in combination with prior medical knowledge, introduction of path deviation is avoided, the individualized adaptive capacity of path design is improved, the safety of the operation is ensured, and compared with an existing method, the path planning efficiency is improved. The method can be applied to medical robot assisted artificial cochlea implantation operations.

Description

Technical Field

[0001] This invention belongs to the field of path planning technology in cochlear implantation, specifically relating to a cochlear implantation grinding path planning method, storage medium, and device. Background Technology

[0002] In cochlear implantation surgery, the planning of the grinding path directly affects the safety of the surgery and the implantation effect. In order to avoid damage to key tissue structures such as the facial nerve and ossicles, doctors usually need to plan the grinding path based on preoperative CT images to expose key tissues for easy surgical manipulation, so that the implant can accurately reach the cochlear opening.

[0003] Currently, grinding path planning primarily relies on preoperative CT imaging. While CT images provide clear information about bony anatomy, their two-dimensional slice format lacks spatial intuitiveness. Surgeons must possess extensive clinical experience to analyze slices layer by layer, identify key structures, and mentally construct their three-dimensional spatial relationships for manual grinding path planning. This approach has significant limitations: it is not only labor-intensive and inefficient, but also highly dependent on the surgeon's subjective judgment, easily introducing path deviations and potentially causing intraoperative complications. Furthermore, the individualized adaptability of path design is poor. To overcome the limitations of two-dimensional images, three-dimensional reconstruction technology can be used to model CT images, visualizing anatomical structures. This assists surgeons in observing the spatial distribution of target structures and automatically planning grinding paths in three-dimensional space using prior medical knowledge. This significantly reduces reliance on the surgeon's subjective judgment, improves the individualized adaptability of path design, and lays the technological foundation for intelligent otological surgery systems. Summary of the Invention

[0004] The purpose of this invention is to solve the problems of low efficiency, easy introduction of path deviation, and poor individual adaptability of existing grinding path planning methods, and to propose a grinding path planning method, storage medium and device for cochlear implantation.

[0005] The technical solution adopted by the present invention to solve the above-mentioned technical problems is: a method for planning the grinding path of cochlear implantation, the method specifically including the following steps:

[0006] Step 1: Obtain a 3D CT image of the area where the human cochlea is located, and then perform semantic segmentation on the obtained 3D CT image to obtain the spatial location of the cochlea, vestibule and facial nerve tissue respectively.

[0007] Based on the spatial location of the cochlea, vestibule, and facial nerve tissues, point cloud data of the cochlea, vestibule, and facial nerve tissues were extracted from the three-dimensional CT images.

[0008] Step 2: Determine the boundary plane n between the cochlea and vestibule based on the cochlear point cloud data and the vestibule point cloud data, and then translate the boundary plane n towards the cochlea to obtain a new plane n1;

[0009] Step 3: Based on the new plane n1 and the point cloud data of the facial nerve tissue, obtain the conical region that needs to be ground, and plan the grinding path based on the conical region.

[0010] Furthermore, the specific process of step three is as follows:

[0011] Step 3: 1. Identify the points in the cochlear point cloud data located on plane n1. Then, select the cochlear point from the identified points that is closest to the facial nerve tissue point cloud data. Use this selected cochlear point as the target point P. t ;

[0012] Step 3.2: Slice the 3D CT image along the Z-axis to obtain the target point P. t Draw a line through the target point P on the obtained two-dimensional image slice. t Horizontal line L1;

[0013] Calculate the target point P respectively t The distance from each facial nerve tissue point on the two-dimensional image slice to the horizontal line L1 is used to obtain the facial nerve tissue point P corresponding to the minimum distance. m ;

[0014] Step 33, with point P m With P as the center, at the target point P t Draw a planar circle O with radius D on the two-dimensional image slice;

[0015] Draw the target point P again t Find the tangent line L2 to the plane circle O, calculate the distance from each facial nerve tissue point to L2, and denote the facial nerve tissue point with the smallest distance as .

[0016] like If the distance to L2 is greater than or equal to the distance D, then proceed to step 30.

[0017] like If the distance to L2 is less than the distance to D, continue with steps three and four.

[0018] Steps 3 and 4: Determine the point Mapping point on tangent L2 With vectors As a direction vector, from point Start, Extend and The straight line continues until the point is reached. Make point With point The straight-line distance is D;

[0019] Step 3.5: Connect the target point P t and points Obtain a straight line Then proceed to step three six;

[0020] Step 36: Initialize the iteration count l = 0;

[0021] Step 37: Calculate the distance from each facial nerve tissue point to the target facial nerve tissue. The distance is calculated, and the facial nerve tissue point corresponding to the minimum distance is denoted as .

[0022] like arrive If the distance is greater than or equal to distance D, then proceed to step 39;

[0023] like arrive If the distance is less than the distance D, then continue with step three.

[0024] Step 38: Determine the point In a straight line Mapping points on With vectors As a direction vector, from point Start, Extend and The straight line continues until the point is reached. Make point With point The straight-line distance is D;

[0025] Connect target point P t and points Obtain a straight line Let l = l + 1 again, and return to step 37;

[0026] Step 39: Cut the line The straight line containing the generatrix of the cone, with Starting from, along Direction extended straight line Until point P5 is reached;

[0027] Among them, point P5 and point The straight-line distance is d;

[0028] Connect point P5 with point P t The intersection of the line connecting the two points and the outer surface of the skull on the CT scan is denoted as P6. Determine point P7 such that points P5 and P... tThe line connecting points P6 and P7 is perpendicular to the line connecting points P6 and P7.

[0029] Determine the base center with point P6, |P6P7| as the base radius, and P... t Let P6 and P be vertices. t The line connecting the central axis, point P7, and point P is the line connecting the two points. t The cone with the line connecting the points is the generatrix. The cone is used as the grinding target, and the grinding path is determined according to the grinding target.

[0030] Where |P6P7| is the straight-line distance from point P6 to point P7;

[0031] Step 30: Place P t As the vertex of the cone, take the tangent L2 as the line containing the generatrix of the cone, and... Projecting onto L2, we get Mapping point on tangent L2 Along Direction extended straight line Until point P5, such that point P5 is the same as point... Let the straight-line distance be d1, and connect point P5 and point P. t We obtain points P5 and P. t The intersection point P6 of the line and the outer surface of the skull in the CT image is taken as the center of the base of the cone;

[0032] Point P t The line connecting point P6 and point P7 is taken as the central axis of the cone. A point P7 is found on L2 such that the line connecting P7 and P6 intersects with point P7. t The line connecting point P6 and point P7 is perpendicular. Using |P6P7| as the radius of the base, points P7 and P... t The line connecting them is the generatrix of the cone;

[0033] The cone is used as the grinding target, and the grinding path is determined based on the grinding target.

[0034] Furthermore, the radius D is:

[0035] D = 1.5 × d safe

[0036] Where, d safe For a safe distance.

[0037] Furthermore, the distance d is calculated as follows:

[0038]

[0039] in, Let P be the point t Time The straight-line distance, where α is the vertex angle of the cone;

[0040] The straight-line distance d1 is:

[0041]

[0042] in, Let P be the point t Time The straight-line distance, "*" indicates multiplication.

[0043] Furthermore, the determination of the grinding path based on the grinding target is planned using a height-level stratification strategy, specifically as follows:

[0044] First, locate the center point P6 of the highest layer as the starting point of the path, move the tool to the outermost point P7 of the highest layer, set the vertical downward milling depth to Δt, and start the milling cutter to begin milling;

[0045] The outermost circumferential contour of the highest layer is used as the milling path. Milling one round ensures that the path is closed. After the outermost circumferential is ground, feed Δt towards the center point and enter the second outermost circumferential contour of the highest layer to continue grinding until the center point of the highest layer is finally ground, thus completing the grinding of the highest layer. Then, enter the second highest layer to continue grinding.

[0046] This process continues, starting from the highest layer and working downwards layer by layer until the grinding reaches the apex of the grinding target.

[0047] Furthermore, the semantic segmentation of the acquired 3D CT images is performed using a 3D network model, specifically a 3D Unet.

[0048] Furthermore, the determination of the boundary plane n between the cochlea and vestibule based on cochlear point cloud data and vestibular point cloud data uses a soft support vector machine.

[0049] Furthermore, the dividing plane n is translated towards the cochlea to obtain a new plane n1, with a translation distance of 0.5 mm.

[0050] A computer storage medium storing at least one instruction, which is loaded and executed by a processor to implement a cochlear implant grinding path planning method.

[0051] A cochlear implant grinding path planning device includes a processor and a memory, wherein the memory stores at least one instruction, which is loaded and executed by the processor to implement a cochlear implant grinding path planning method.

[0052] The beneficial effects of this invention are:

[0053] This invention proposes an automated grinding path planning method for cochlear implant electrode implantation surgery. By preprocessing preoperative CT images using an anatomical model, specifically through model segmentation technology, key structures in the temporal bone can be accurately located in the preoperative CT images to extract the three-dimensional spatial geometric information of key structures in the cochlea, vestibule, and facial nerve. Combined with prior medical knowledge, the grinding path of the cochlear implant surgical area is automatically planned in three-dimensional space. This significantly reduces reliance on the surgeon's subjective judgment, avoids introducing path deviations, and improves the individualized adaptability of path design, minimizing damage to key nerves and ensuring surgical safety. Furthermore, compared to existing methods, this method improves the efficiency of path planning and is suitable for robot-assisted cochlear implantation surgery. Attached Figure Description

[0054] Figure 1 This is a schematic diagram of the boundary plane between the cochlea and the vestibule;

[0055] Figure 2 This is a schematic diagram of the planning process within a two-dimensional image slice;

[0056] Figure 3 This is a diagram showing how to extend the lines containing P2 and P3 to obtain a new line;

[0057] Figure 4 A schematic diagram for determining the base radius of the grinding target;

[0058] Figure 5 A schematic diagram of the grinding target;

[0059] Figure 6 This is a diagram showing the planned grinding path.

[0060] (a) shows the grinding path between layers, and (b) shows the grinding path within a single layer. Detailed Implementation

[0061] Specific Implementation Method 1: The cochlear implantation grinding path planning method described in this implementation method specifically includes the following steps:

[0062] Step 1: Obtain a 3D CT image of the area where the human cochlea is located, and then perform semantic segmentation on the obtained 3D CT image to obtain the spatial location of the cochlea, vestibule and facial nerve tissue respectively.

[0063] Based on the spatial location of the cochlea, vestibule, and facial nerve tissues, point cloud data of the cochlea, vestibule, and facial nerve tissues were extracted from the three-dimensional CT images.

[0064] Step Two, as follows Figure 1As shown, the boundary plane n between the cochlea and the vestibular system is determined based on cochlear point cloud data and vestibular point cloud data. Then, the boundary plane n is translated towards the cochlea to obtain a new plane n1.

[0065] Step 3: Based on the new plane n1 and the point cloud data of the facial nerve tissue, obtain the conical region that needs to be ground, and plan the grinding path based on the conical region.

[0066] Specific Implementation Method Two: This implementation method is a further limitation of Specific Implementation Method One. The specific process of step three is as follows:

[0067] Step 3: 1. Identify the points on plane n1 in the cochlear point cloud data. Then, select the cochlear point from the identified points that is closest to the facial nerve tissue point cloud data (i.e., for any identified point, calculate the distance between that point and each point in the facial nerve tissue to obtain the minimum distance corresponding to that point; after calculating the minimum distance for each identified point, compare the minimum values ​​to obtain the cochlear point corresponding to the minimum value). Use the selected cochlear point as the target point P. t ;

[0068] Step 3.2: Slice the 3D CT image along the Z-axis to obtain the target point P. t The two-dimensional image slice in which it is located, such as Figure 2 As shown, a line is drawn through the target point P on the obtained two-dimensional image slice. t Horizontal line L1;

[0069] Calculate the target point P respectively t The distance from each facial nerve tissue point on the two-dimensional image slice to the horizontal line L1 is used to obtain the facial nerve tissue point P corresponding to the minimum distance. m ;

[0070] Step 33, with point P m With P as the center, at the target point P t Draw a planar circle O with radius D on the two-dimensional image slice;

[0071] Draw the target point P again t Find the tangent line L2 to the plane circle O, calculate the distance from each facial nerve tissue point to L2, and denote the facial nerve tissue point with the smallest distance as .

[0072] like If the distance to L2 is greater than or equal to the distance D, then proceed to step 30.

[0073] like If the distance to L2 is less than the distance to D, continue with steps three and four.

[0074] Steps 3 and 4: Determine the point Mapping point on tangent L2 (i.e., from point) tangent Draw a perpendicular line, and the foot of the perpendicular is... ),like Figure 3 As shown, in vector As a direction vector, from point Start, Extend and The straight line continues until the point is reached. Make point With point The straight-line distance is D;

[0075] Step 3.5: Connect the target point P t and points Obtain a straight line Then proceed to step three six;

[0076] Step 36: Initialize the iteration count l = 0;

[0077] Step 37: Calculate the distance from each facial nerve tissue point to the target facial nerve tissue. The distance is calculated, and the facial nerve tissue point corresponding to the minimum distance is denoted as .

[0078] like arrive If the distance is greater than or equal to distance D, then proceed to step 39;

[0079] like arrive If the distance is less than the distance D, then continue with step three.

[0080] Step 38: Determine the point In a straight line Mapping points on With vectors As a direction vector, from point Start, Extend and The straight line continues until the point is reached. Make point With point The straight-line distance is D;

[0081] Connect target point P t and points Obtain a straight line Let l = l + 1 again, and return to step 37;

[0082] Step 39: Cut the line The straight line containing the generatrix of the cone, with Starting from, along Direction extended straight line Until point P5 is reached;

[0083] Among them, point P5 and point The straight-line distance is d;

[0084] like Figure 4 As shown, connect point P5 with point P t The intersection of the line connecting the two points and the outer surface of the skull on the CT scan is denoted as P6 (or it can be the intersection of point P5 and point P). t The point on the line connecting the two points is located on the outer side of the CT skull surface (at which point there is a small distance between this point and the intersection point), on the straight line. Determine point P7 such that points P5 and P... t The line connecting points P6 and P7 is perpendicular to the line connecting points P6 and P7.

[0085] like Figure 5 As shown, the base circle is defined with point P6 as its center, |P6P7| as its radius, and P... t Let P6 and P be vertices. t The line connecting the central axis, point P7, and point P is the line connecting the two points. t The cone with the line connecting the points is the generatrix. The cone is used as the grinding target, and the grinding path is determined according to the grinding target.

[0086] Where |P6P7| is the straight-line distance from point P6 to point P7;

[0087] Step 30: Place P t As the vertex of the cone, take the tangent L2 as the line containing the generatrix of the cone, and... Projecting onto L2, we get Mapping point on tangent L2 Along Direction extended straight line Until point P5, such that point P5 is the same as point... Let the straight-line distance be d1, and connect point P5 and point P. t We obtain points P5 and P. t The intersection point P6 (or points P5 and P6) of the line and the outer surface of the skull in the CT image. t The point on the line connecting the two points is located on the outer side of the CT skull (at which point there is a small distance between this point and the intersection point). Point P6 is taken as the center of the base of the cone.

[0088] Point P t The line connecting point P6 and point P7 is taken as the central axis of the cone. A point P7 is found on L2 such that the line connecting P7 and P6 intersects with point P7. t The line connecting point P6 and point P7 is perpendicular. Using |P6P7| as the radius of the base, points P7 and P... tThe line connecting them is the generatrix of the cone;

[0089] The cone is used as the grinding target, and the grinding path is determined based on the grinding target.

[0090] The overall programming algorithm is shown in Table 1:

[0091] Table 1

[0092]

[0093]

[0094] The other steps and parameters are the same as in Specific Implementation Method 1.

[0095] Specific Implementation Method Three: This implementation method is a further limitation of Specific Implementation Method Two, wherein the radius D is:

[0096] D = 1.5 × d safe (1)

[0097] Where, d safe For a safe distance.

[0098] The other steps and parameters are the same as in Specific Implementation Method Two.

[0099] The grinding boundary should maintain a safe distance of at least 0.5 mm from the facial nerve to prevent damage to the facial nerve during surgery. In this invention, the safe distance d is defined as follows: safe Set to 1.0 mm.

[0100] Specific Implementation Method Four: This implementation method further defines Specific Implementation Method Three. The method for calculating the distance d is as follows:

[0101]

[0102] in, Let P be the point t Time The straight-line distance, where α is the apex angle of the cone (determined by the doctor based on clinical surgical experience and the diameter of the grinding drill bit);

[0103] The straight-line distance d1 is:

[0104]

[0105] in, Let P be the point t Time The straight-line distance.

[0106] The other steps and parameters are the same as in Specific Implementation Method 3.

[0107] Specific Implementation Method Five: This implementation method further defines Specific Implementation Method Four. The determination of the grinding path based on the grinding target is planned using a contour layering strategy, specifically as follows:

[0108] First, locate the center point P6 of the highest layer as the starting point of the path. Move the tool to the outermost point P7 of the highest layer and set the vertical downward milling depth to Δt, that is, the feed depth to Δt. This depth must be less than the tool diameter to ensure sufficient milling. Start the milling cutter to begin milling.

[0109] The outermost circumferential contour of the highest layer is used as the milling path. Milling one round ensures that the path is closed. After the outermost circumferential is ground, feed Δt towards the center point and enter the second outermost circumferential contour of the highest layer to continue grinding until the center point of the highest layer is finally ground, thus completing the grinding of the highest layer. Then, enter the second highest layer to continue grinding.

[0110] This process continues, starting from the highest layer and working downwards layer by layer until the grinding reaches the apex of the grinding target.

[0111] The other steps and parameters are the same as in Specific Implementation Method Four.

[0112] like Figure 6 As shown, (a) represents the grinding path between each layer, and (b) represents the grinding path planning for a single layer. The dashed arrows indicate that the tool movement path does not involve grinding, and the solid lines represent the grinding path.

[0113] Specific Implementation Method Six: This implementation method is a further limitation of Specific Implementation Method Five. The semantic segmentation of the acquired three-dimensional CT images is performed using a 3D network model, which is 3D Unet.

[0114] The other steps and parameters are the same as in Specific Implementation Method 5.

[0115] The 3D network models that can be used in this invention include, but are not limited to, 3D Unet.

[0116] Specific Implementation Method Seven: This implementation method is a further limitation of Specific Implementation Method Six. The method for determining the boundary plane n between the cochlea and vestibule based on cochlear point cloud data and vestibular point cloud data is a soft support vector machine (SoftSVM).

[0117] The other steps and parameters are the same as in Specific Implementation Method Six.

[0118] Specific Implementation Method Eight: This implementation method is a further limitation of Specific Implementation Method Seven. The boundary plane n is translated towards the cochlea to obtain a new plane n1, and the translation distance is 0.5 mm.

[0119] The other steps and parameters are the same as in Specific Implementation Method Seven.

[0120] Based on prior knowledge, the circular window is located approximately 0.5 mm below the junction of the cochlea and vestibule. Therefore, the present invention sets the translation distance to 0.5 mm. Specific implementation method nine:

[0122] This embodiment is a computer storage medium that stores at least one instruction, which is loaded and executed by a processor to implement the cochlear implant grinding path planning method.

[0123] It should be understood that the instructions include computer program products, software, or computerized methods corresponding to any method described in this invention; the instructions can be used to program computer systems or other electronic devices. Computer storage media may include readable media on which instructions are stored, and may include, but are not limited to, magnetic storage media, optical storage media; magneto-optical storage media include read-only memory (ROM), random access memory (RAM), erasable programmable memory (e.g., EPROM and EEPROM), and flash memory layers, or other types of media suitable for storing electronic instructions. Specific Implementation Method Ten:

[0125] This embodiment is a cochlear implantation grinding path planning device. The device includes a processor and a memory. It should be understood that it includes any device including a processor and a memory described in this invention. The device may also include other units or modules that perform display, interaction, processing, control and other functions through signals or instructions.

[0126] The memory stores at least one instruction, which is loaded and executed by the processor to implement the cochlear implant grinding path planning method.

[0127] Those skilled in the art will understand that at least one stored instruction constitutes a computer program product corresponding to a method or system. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0128] This application is described with reference to flowchart illustrations and / or block diagrams of methods, systems, and computer program products according to embodiments of this application, and can also be used with corresponding devices. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0129] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0130] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0131] The above examples of the present invention are merely illustrative of the computational model and process of the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is impossible to exhaustively list all possible implementations here. Any obvious variations or modifications derived from the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for planning the grinding path for cochlear implantation, characterized in that, The method specifically includes the following steps: Step 1: Obtain a 3D CT image of the area where the human cochlea is located, and then perform semantic segmentation on the obtained 3D CT image to obtain the spatial location of the cochlea, vestibule and facial nerve tissue respectively. Based on the spatial location of the cochlea, vestibule, and facial nerve tissues, point cloud data of the cochlea, vestibule, and facial nerve tissues were extracted from the three-dimensional CT images. Step 2: Determine the boundary plane n between the cochlea and vestibule based on the cochlear point cloud data and the vestibule point cloud data, and then translate the boundary plane n towards the cochlea to obtain a new plane n1; Step 3: Based on the new plane n1 and the point cloud data of the facial nerve tissue, obtain the conical region that needs to be ground, and plan the grinding path based on the conical region.

2. The method for planning the grinding path for cochlear implantation according to claim 1, characterized in that, The specific process of step three is as follows: Step 3:

1. Identify the points in the cochlear point cloud data located on plane n1. Then, select the cochlear point from the identified points that is closest to the facial nerve tissue point cloud data. Use this selected cochlear point as the target point P. t ; Step 3.2: Slice the 3D CT image along the Z-axis to obtain the target point P. t Draw a line through the target point P on the obtained two-dimensional image slice. t Horizontal line L1; Calculate the target point P respectively t The distance from each facial nerve tissue point on the two-dimensional image slice to the horizontal line L1 is used to obtain the facial nerve tissue point P corresponding to the minimum distance. m ; Step 33, with point P m With P as the center, at the target point P t Draw a planar circle O with radius D on the two-dimensional image slice; Draw the target point P again t Find the tangent line L2 to the plane circle O, and calculate the distance from each facial nerve tissue point to L2. Record the facial nerve tissue point with the smallest distance as... like If the distance to L2 is greater than or equal to the distance D, then proceed to step 30. like If the distance to L2 is less than the distance to D, continue with steps three and four. Steps 3 and 4: Determine the point Mapping point on tangent L2 With vectors As a direction vector, from point Start, Extend and The straight line continues until the point is reached. Make point With point The straight-line distance is D; Step 35: Connect the target point P t and points Obtain a straight line Then proceed to step three six; Step 36: Initialize the iteration count l = 0; Step 37: Calculate the distance from each facial nerve tissue point to the target facial nerve tissue. The distance is calculated, and the facial nerve tissue point corresponding to the minimum distance is denoted as . like arrive If the distance is greater than or equal to distance D, then proceed to step 39; like arrive If the distance is less than the distance D, then continue with step three. Step 38: Determine the point In a straight line Mapping points on With vectors As a direction vector, from point Start, Extend and The straight line continues until the point is reached. Make point With point The straight-line distance is D; Connect target point P t and points Obtain a straight line Let l = l + 1 again, and return to step 37; Step 39: Cut the line The straight line containing the generatrix of the cone, with Starting from, along Direction extended straight line Until point P5 is reached; Among them, point P5 and point The straight-line distance is d; Connect point P5 with point P t The intersection of the line connecting the two points and the outer surface of the skull on the CT scan is denoted as P6. Determine point P7 such that points P5 and P... t The line connecting points P6 and P7 is perpendicular to the line connecting points P6 and P7. Determine the base center with point P6, |P6P7| as the base radius, and P... t Let P6 and P be vertices. t The line connecting the central axis, point P7, and point P is the line connecting the two points. t The cone with the line connecting the points is the generatrix. The cone is used as the grinding target, and the grinding path is determined according to the grinding target. Where |P6P7| is the straight-line distance from point P6 to point P7; Step 30: Place P t As the vertex of the cone, take the tangent L2 as the line containing the generatrix of the cone, and... Projecting onto L2, we get Mapping point on tangent L2 Along Direction extended straight line Until point P5, such that point P5 is the same as point... Let the straight-line distance be d1, and connect point P5 and point P. t We obtain points P5 and P. t The intersection point P6 of the line and the outer surface of the skull in the CT image is taken as the center of the base of the cone; Point P t The line connecting point P6 and point P7 is taken as the central axis of the cone. A point P7 is found on L2 such that the line connecting P7 and P6 intersects with point P7. t The line connecting point P6 and point P7 is perpendicular. Using |P6P7| as the radius of the base, points P7 and P... t The line connecting them is the generatrix of the cone; The cone is used as the grinding target, and the grinding path is determined based on the grinding target.

3. The method for planning the grinding path for cochlear implantation according to claim 2, characterized in that, The radius D is: D=1.5×d safe Where, d safe For a safe distance.

4. The method for planning the grinding path for cochlear implantation according to claim 3, characterized in that, The distance d is calculated as follows: in, Let P be the point t Time The straight-line distance, where α is the vertex angle of the cone; The straight-line distance d1 is: in, Let P be the point t Time The straight-line distance.

5. The method for planning the grinding path for cochlear implantation according to claim 4, characterized in that, The determination of the grinding path based on the grinding target is planned using a height-level stratification strategy, specifically as follows: First, locate the center point P6 of the highest layer as the starting point of the path, move the tool to the outermost point P7 of the highest layer, set the vertical downward milling depth to Δt, and start the milling cutter to begin milling; The outermost circumferential contour of the highest layer is used as the milling path. Milling one round ensures that the path is closed. After the outermost circumferential is ground, feed Δt towards the center point and enter the second outermost circumferential contour of the highest layer to continue grinding until the center point of the highest layer is finally ground, thus completing the grinding of the highest layer. Then, enter the second highest layer to continue grinding. This process continues, starting from the highest layer and working downwards layer by layer until the grinding reaches the apex of the grinding target.

6. The method for planning the grinding path for cochlear implantation according to claim 5, characterized in that, The semantic segmentation of the acquired 3D CT images is performed using a 3D network model, which is 3D Unet.

7. The method for planning the grinding path for cochlear implantation according to claim 6, characterized in that, The method for determining the boundary plane n between the cochlea and vestibule based on cochlear point cloud data and vestibular point cloud data uses a soft support vector machine.

8. The method for planning the grinding path for cochlear implantation according to claim 7, characterized in that, The dividing plane n is translated towards the cochlea to obtain a new plane n1, and the translation distance is 0.5 mm.

9. A computer storage medium, characterized in that, The storage medium stores at least one instruction, which is loaded and executed by a processor to implement the cochlear implant grinding path planning method according to any one of claims 1 to 8.

10. A cochlear implantation grinding path planning device, characterized in that, The device includes a processor and a memory, the memory storing at least one instruction, which is loaded and executed by the processor to implement the cochlear implant grinding path planning method according to any one of claims 1 to 8.

Images