Temporomandibular joint surgery simulation method and system based on AI technology

Through AI-based methods, the problem of edge detection accuracy in temporomandibular joint surgery simulation was solved. By calculating the synchronization degree and distance of corner point movement, the marginal line of the temporomandibular joint was determined, and the anatomical structural model of the temporomandibular joint was constructed. It improves the accuracy and reliability of surgical simulation and enhances the treatment effect.

CN118762844BActive Publication Date: 2024-11-05SICHUAN UNIV
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202411253738.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-09
Publication Date
2024-11-05
Estimated Expiration
2044-09-09

AI Technical Summary

Technical Problem

In temporomandibular joint surgical simulation, artifacts appear during edge detection due to motion blur in MRI images, which cannot accurately identify the true edge of the mandible, reducing the accuracy of surgical simulation.

Method used

Using AI-based method, by constructing the coordinate system of each frame of temporomandibular joint image, calculating the synchronization degree and distance of corner position movement, determining the edge lines of the temporal and mandible bones, using optical flow method and Harris corner point detection matching angles Points, combined with the LOF algorithm, abnormal detection is performed, the real mandible marginal line is obtained, and the anatomical structural model of the temporomandibular joint is constructed.

Benefits of technology

It improves the accuracy and reliability of temporomandibular joint surgery simulation, enhances the success rate of the surgery, and increases the reliability of the patient's treatment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118762844B_ABST
    Figure CN118762844B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of medical care informatics, and in particular to a method and system for simulating temporomandibular joint surgery based on AI technology, including: obtaining temporal bone corner points and temporal bone edge lines in each frame of temporomandibular joint image according to the degree of synchronization and distance of movement of any two corner points of each frame of temporomandibular joint image; obtaining a number of mandibular corner points and mandibular movement directions according to the distance between the temporal bone corner points in each frame of temporomandibular joint image and the coordinate position of the corner points of the temporomandibular joint image on the coordinate system; obtaining the real mandibular edge line in each frame of temporomandibular joint image according to the edge line where the mandibular corner points are located and the mandibular movement direction in each frame of temporomandibular joint image; constructing an anatomical structure model of the temporomandibular joint according to the real mandibular edge lines and temporal bone edge lines of all frames of temporomandibular joint images. The present invention improves the accuracy of temporomandibular joint surgery simulation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of medical care informatics, and in particular to a temporomandibular joint surgery simulation method and system based on AI technology. Background Art

[0002] At present, the hospital has established an artificial intelligence-based temporomandibular joint surgery simulation system, which allows doctors to perform simulation exercises before real surgery and discover potential risks in surgery in advance. Before performing temporomandibular joint surgery, it is necessary to determine the anatomical structure model of the temporomandibular joint when the patient opens and closes his mouth. Generally, the positional relationship between the opening and closing of the temporal bone and the mandible can be determined through the patient's MRI video to establish an anatomical structure model. When determining the positional relationship between the opening and closing of the temporal bone and the mandible, the temporomandibular joint is affected by activities such as chewing and opening the mouth. In the MRI images, artifacts appear in some video frame images due to motion blur. When determining the distance between the temporomandibular joints by edge detection, there are multiple edges of the mandible, and the real edge of the mandible cannot be identified, which reduces the accuracy of the temporomandibular joint surgery simulation. Summary of the invention

[0003] The present invention provides a temporomandibular joint surgery simulation method and system based on AI technology to solve the problem that when the existing edge detection determines the distance between the temporomandibular joints, there are multiple edges of the mandible and the real edge of the mandible cannot be identified.

[0004] The AI-based temporomandibular joint surgery simulation method and system of the present invention adopts the following technical solutions:

[0005] An embodiment of the present invention provides a temporomandibular joint surgery simulation method based on AI technology, the method comprising the following steps:

[0006] Obtaining a plurality of frames of temporomandibular joint images and the corresponding corner points of each corner point in each frame of the temporomandibular joint image in other frames of the temporomandibular joint image;

[0007] Constructing a coordinate system of each frame of the temporomandibular joint image; on the coordinate system, obtaining the degree of synchronization of the position movement of any two corner points of each frame of the temporomandibular joint image according to the distance between the coordinates of the corner points in adjacent temporomandibular joint images; obtaining the temporal bone corner points and the temporal bone edge lines in each frame of the temporomandibular joint image according to the degree of synchronization and distance of the position movement of any two corner points of each frame of the temporomandibular joint image;

[0008] According to the distance between the temporal bone corner points in each frame of the temporomandibular joint image and the coordinate position of the corner points of the temporomandibular joint image on the coordinate system, a number of mandibular bone corner points and the mandibular bone movement direction are obtained;

[0009] According to the edge line of the mandibular corner point in each frame of the temporomandibular joint image and the mandibular movement direction, the real mandibular edge line in each frame of the temporomandibular joint image is obtained;

[0010] The anatomical structure model of the temporomandibular joint is constructed according to the real mandibular edge line and temporal bone edge line of all frames of temporomandibular joint images.

[0011] Preferably, the degree of synchronization of the position movement of any two corner points of each frame of the temporomandibular joint image is obtained according to the distance between the coordinates of the corner points in adjacent temporomandibular joint images, and the specific formula included is as follows:

[0012]

[0013] In the formula, Represents the first Corner points and The degree of synchronization of the movement of the corner points; Represents the number of all frames of the TMJ image; Indicates The first frame of the temporomandibular joint image The corner point and its The distance between the coordinates of the corner points corresponding to the frame temporomandibular joint image in the coordinate system; Indicates The first frame of the temporomandibular joint image The corner point and its The distance between the coordinates of the corner points corresponding to the frame temporomandibular joint image in the coordinate system; is the absolute value function.

[0014] Preferably, the step of obtaining the temporal bone corner points and the temporal bone edge lines in each frame of the temporomandibular joint image according to the synchronization degree and distance of the position movement of any two corner points in each frame of the temporomandibular joint image comprises the following specific steps:

[0015] According to the synchronization degree and distance of the position movement of any two corner points of each frame of the temporomandibular joint image, a movement synchronization index between any two corner points of each frame of the temporomandibular joint image is obtained;

[0016] According to the movement synchronization index between any two corner points of each frame of the temporomandibular joint image, the temporal bone corner points and the temporal bone edge lines in each frame of the temporomandibular joint image are obtained.

[0017] Preferably, the movement synchronization index between any two corner points of each frame of the temporomandibular joint image is obtained according to the synchronization degree and distance of the position movement of any two corner points of each frame of the temporomandibular joint image, and the specific formula included is as follows:

[0018]

[0019] In the formula, Represents the first Corner points and Movement synchronization indicators between corner points; Represents the first Corner points and The degree of synchronization of the movement of the corner points; Represents the number of all frames of the TMJ image; Indicates Frame TMJ image Corner points and The Euclidean distance between corner points; Represents the first frame of all temporomandibular joint images Corner points and The average of the Euclidean distances between the corner points; is the absolute value function.

[0020] Preferably, the step of obtaining the temporal bone corner points and the temporal bone edge lines in each frame of the temporomandibular joint image according to the movement synchronization index between any two corner points of each frame of the temporomandibular joint image comprises the following specific steps:

[0021] In each frame of the temporomandibular joint image, the two corner points corresponding to the smallest movement synchronization index are recorded as temporal bone corner points, and the edge line where the temporal bone corner points are located is recorded as the temporal bone edge line.

[0022] Preferably, the method of obtaining a plurality of mandibular corner points and the mandibular movement direction according to the distance between the temporal bone corner points in each frame of the temporomandibular joint image and the coordinate positions of the corner points of the temporomandibular joint image on the coordinate system comprises the following specific steps:

[0023] In each frame of the temporomandibular joint image, the center point of the straight line segment connecting the two temporal bone corners is recorded as the reference point, and the reference point is used as the center of the circle. The distance between two temporal bone corners is the radius, and a semicircle is constructed by rotating clockwise from the horizontal right direction to the horizontal left direction. The corners in the semicircle that are not temporal bone corners are recorded as suspected mandibular corners; among them, is the preset coefficient;

[0024] In the coordinate system, The first frame of the temporomandibular joint image The suspected mandibular angle is The distance and direction of the coordinates between the corner points corresponding to the frame temporomandibular joint image constitute the first The first frame of the temporomandibular joint image vectors of suspected mandibular corner points;

[0025] According to The direction of the vector of each suspected mandibular corner point in the frame temporomandibular joint image is used to detect abnormalities of all suspected mandibular corner points using the LOF algorithm to obtain abnormal mandibular corner points; among all suspected mandibular corner points, the suspected mandibular corner points except the abnormal mandibular corner points are recorded as mandibular corner points;

[0026] The first The direction of the sum of the vectors of all mandibular corner points in the frame temporomandibular joint image is denoted as Direction of mandibular motion in frame temporomandibular joint images.

[0027] Preferably, the method of obtaining the true mandibular edge line in each frame of the temporomandibular joint image according to the edge line where the mandibular corner point is located in each frame of the temporomandibular joint image and the mandibular movement direction comprises the following specific steps:

[0028] According to the edge line of the mandibular corner point in each frame of the temporomandibular joint image and the direction of mandibular movement, the similar mandibular edge index of each frame of the temporomandibular joint image is obtained;

[0029] According to the similar mandibular edge index of each frame of the temporomandibular joint image, the real mandibular edge line in each frame of the temporomandibular joint image is obtained.

[0030] Preferably, the method of obtaining the similar mandibular edge index of each frame of the temporomandibular joint image according to the edge line of the mandibular corner point in each frame of the temporomandibular joint image and the mandibular movement direction includes the following specific steps:

[0031] The edge line where the mandibular corner point is located in each frame of the temporomandibular joint image is recorded as the mandibular edge line;

[0032] According to The first frame of the temporomandibular joint image A method for obtaining the vectors of the suspected mandibular corner points, and obtaining the vector of the center of the minimum circumscribed circle of each mandibular edge line in each frame of the temporomandibular joint image;

[0033] In each frame of the temporomandibular joint image, the vector of the center of the minimum circumscribed circle of each mandibular edge line is recorded as the motion vector of each mandibular edge line;

[0034] The first Frame to All the temporomandibular joint images between frames are recorded as The frame temporomandibular joint image corresponds to the reference temporomandibular joint image;

[0035] The direction of the sum of the motion vectors of all mandibular edge lines in all reference TMJ images is recorded as The movement direction of the mandibular edge line corresponding to the frame temporomandibular joint image;

[0036] No. The calculation method of the similar mandibular margin index of the frame temporomandibular joint image is as follows:

[0037]

[0038] In the formula, Indicates Similar mandibular margin indicators for framed TMJ images; Indicates The minimum angle between the mandibular movement direction corresponding to the frame temporomandibular joint image and the mandibular edge line movement direction; Indicates The mean of the curvatures of all mandibular edge lines in the frame-referenced TMJ image; Indicates The mean of the curvatures of all mandibular edge lines in the frame-referenced TMJ image; is the absolute value function; Indicates the number of reference TMJ images.

[0039] Preferably, the method of obtaining the true mandibular edge line in each frame of the temporomandibular joint image according to the similar mandibular edge index of each frame of the temporomandibular joint image comprises the following specific steps:

[0040] All frames of temporomandibular joint images are equally divided into several temporomandibular joint image sequence segments. In each temporomandibular joint image sequence segment, the mandibular edge line in the temporomandibular joint image corresponding to the minimum similar mandibular edge index is used as the true mandibular edge line in each frame of the temporomandibular joint image.

[0041] The present invention also proposes a temporomandibular joint surgery simulation system based on AI technology, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program stored in the memory to implement the steps of the aforementioned temporomandibular joint surgery simulation method based on AI technology.

[0042] The beneficial effects of the technical solution of the present invention are as follows: according to the distance between the coordinates of the corner points in adjacent temporomandibular joint images, the degree of synchronization of the position movement of any two corner points of each frame of the temporomandibular joint image is obtained; according to the degree of synchronization and the distance between the position movement of any two corner points of each frame of the temporomandibular joint image, the temporal bone corner points and the temporal bone edge lines in each frame of the temporomandibular joint image are obtained; according to the distance between the temporal bone corner points in each frame of the temporomandibular joint image and the coordinate positions of the corner points of the temporomandibular joint image on the coordinate system, a number of mandibular corner points and the mandibular movement direction are obtained; accurate The real mandibular edge line in each frame of the temporomandibular joint image is obtained to improve the reliability of the temporomandibular joint surgery simulation model; the real mandibular edge line in each frame of the temporomandibular joint image is obtained according to the edge line of the mandibular corner point in each frame of the temporomandibular joint image and the direction of mandibular movement; the anatomical structure model of the temporomandibular joint is constructed according to the real mandibular edge line and temporal bone edge line of all frames of the temporomandibular joint image; obtaining the temporal bone edge and the real mandibular edge can construct a reliable temporomandibular joint model, increase the reliability of patient treatment; and improve the success rate of temporomandibular joint surgery. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 A flowchart of the steps of the temporomandibular joint surgery simulation method based on AI technology of the present invention;

[0045] Figure 2 A schematic diagram of a human temporomandibular joint provided in this embodiment;

[0046] Figure 3 A schematic diagram of obtaining the movement direction of the mandible in each frame of the temporomandibular joint image provided in this embodiment. DETAILED DESCRIPTION

[0047] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the temporomandibular joint surgery simulation method and system based on AI technology proposed by the present invention, its specific implementation method, structure, characteristics and effects are described in detail as follows in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0048] Unless defined otherwise, 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 belongs.

[0049] The specific scheme of the temporomandibular joint surgery simulation method and system based on AI technology provided by the present invention is described in detail below with reference to the accompanying drawings.

[0050] See also Figure 1 , which shows a flowchart of a method for simulating temporomandibular joint surgery based on AI technology provided by an embodiment of the present invention, the method comprising the following steps:

[0051] Step S001: Obtain several frames of temporomandibular joint images.

[0052] It should be noted that patients need to undergo MRI examinations before undergoing temporomandibular joint surgery. During the MRI examination, patients need to open and close their mouths to determine the current condition of the temporomandibular joint during the opening and closing process. The operator can record the video during the MRI scan and use the recorded video to assist in subsequent treatment; each frame of the image is grayed out to obtain a temporomandibular joint image.

[0053] Thus, several frames of continuous temporomandibular joint images are obtained.

[0054] Step S002: construct a coordinate system for each frame of the temporomandibular joint image; on the coordinate system, obtain the degree of synchronization between the position movement of any two corner points of each frame of the temporomandibular joint image based on the distance between the coordinates of the corner points in adjacent temporomandibular joint images; obtain the temporal bone corner points and temporal bone edge lines in each frame of the temporomandibular joint image based on the degree of synchronization and distance between the position movement of any two corner points of each frame of the temporomandibular joint image.

[0055] It should be noted that since the temporal bone and the mandible have protruding parts, they can be identified through corner points. First, the edge position of the temporal bone is determined by the positional relationship of the same corner points, and then the temporal bone is used as a reference to determine the movement direction of the mandibular joint according to the positional relationship of the corner points of the suspected mandible between multiple frames. Finally, the real mandibular edge is determined by the movement direction of the mandibular joint and the mandibular edge conditions between adjacent frames.

[0056] What needs to be explained is that the schematic diagram of the human temporomandibular joint is as follows Figure 2As shown in the figure, the temporomandibular joint is mainly composed of the mandible, the temporal bone and the articular disc. In its MRI image, the articular fossa of the temporal bone usually appears as a distinct concave area with protruding parts on both sides, namely the articular processes. The condyle of the mandible is the part of the mandible that extends upward. They interact with the articular fossa of the temporal bone to allow the mandible to open and close. In MRI images, the condyle usually appears as a relatively rounded protruding structure, and the articular disc is not shown in the image. When a person opens and closes his mouth, when closing the mouth, the condyle of the mandible slides into the articular fossa of the temporal bone, and the articular disc is located above the condyle to assist in buffering and reduce friction; when opening the mouth, the condyle slides along the articular fossa and gradually moves downward, and the articular disc moves forward and rotates with the condyle to adapt to the descent and extension of the mandible. During the movement of the temporomandibular joint, the articular fossa of the temporal bone will adapt to make small adjustments to accommodate the movement of the mandible.

[0057] It should be noted that during the whole process, the temporal bone moves slightly when the mouth is opened and closed, and the mandible slides along the articular fossa and moves downward. In the temporomandibular joint image, the articular process of the temporal bone has obvious characteristics of local grayscale changes.

[0058] Harris corner detection is performed on each frame of the temporomandibular joint image to obtain all corner points of each frame of the grayscale image in the temporomandibular joint video.

[0059] In each frame of the temporomandibular joint image, the optical flow method is used to match the corner points in each frame with the corner points in other frames to obtain the corresponding corner points of each corner point in each frame of the temporomandibular joint image in other frames of the temporomandibular joint image.

[0060] Among them, Harris corner detection and optical flow method are well-known technologies, and the specific methods are not introduced here.

[0061] Perform edge detection on each frame of the temporomandibular joint image to obtain the edge line.

[0062] It should be noted that, in this embodiment, only the corner points on the edge line are analyzed later.

[0063] The coordinate system is constructed with the vertex in the upper left corner of each frame of the temporomandibular joint image as the origin, the horizontal rightward as the horizontal axis, and the vertical downward as the vertical axis.

[0064] It should be noted that the change distance of the corner point at the current position is determined by matching the pixel points at the corner point of each frame of the image. Since there are corner points on both sides of the articular process, any two corner points on each frame of the grayscale image can be obtained. For all the corner points determined on each frame of the image, since there are corner points on both sides of the articular process, any two corner points on each frame of the image are obtained, and the articular process corner points of the temporal bone are determined according to the positional relationship of the corner points.

[0065] Each frame of the temporomandibular joint image Corner points and The degree of synchronization of the movement of the corner points is calculated as follows:

[0066]

[0067] In the formula, Represents the first Corner points and The degree of synchronization of the movement of the corner points; Represents the number of all frames of the TMJ image; Indicates The first frame of the temporomandibular joint image The corner point and its The distance between the coordinates of the corner points corresponding to the frame temporomandibular joint image in the coordinate system; Indicates The first frame of the temporomandibular joint image The corner point and its The distance between the coordinates of the corner points corresponding to the frame temporomandibular joint image in the coordinate system; is the absolute value function.

[0068] What needs to be explained is that when the denominator in the formula is 0, set the denominator to 1 and use this as an example to ensure that the formula is valid.

[0069] What needs to be explained is that Indicates The corner points in the temporomandibular joint image of the first frame and their The distances of the coordinates of the corner points corresponding to the frame temporomandibular joint images in the coordinate system are consistent, and the displacement is relatively small.

[0070] At this point, the synchronization degree of the position movement of any two corner points of each frame of the temporomandibular joint image is obtained.

[0071] Each frame of the temporomandibular joint image Corner points and The calculation method of the moving synchronization index between corner points is as follows:

[0072]

[0073] In the formula, Represents the first Corner points and Movement synchronization indicators between corner points; Represents the first Corner points and The degree of synchronization of the movement of the corner points; Represents the number of all frames of the TMJ image; Indicates Frame TMJ image Corner points and The Euclidean distance between corner points; Represents the first frame of all temporomandibular joint images Corner points and The average of the Euclidean distances between the corner points; is the absolute value function.

[0074] What needs to be explained is that It means that the relative distance between the two corner points in the temporomandibular joint image is unchanged. The smaller the value is, the more likely it is that the two corner points are on both sides of the articular process. When it is the smallest, it means that the current corner point is the corner point on both sides of the articular process. Since the temporal bone moves only slightly when the mouth is opened and closed, after determining the position of the articular process corner point, the position of the temporal bone in the joint can be determined.

[0075] In each frame of the temporomandibular joint image, the two corner points corresponding to the smallest movement synchronization index are recorded as temporal bone corner points, and the edge line where the temporal bone corner points are located is recorded as the temporal bone edge line.

[0076] It should be noted that: in each frame of the temporomandibular joint image, when there are multiple minimum movement synchronization indicators, any one of them is taken as an example for analysis.

[0077] At this point, two temporal bone corner points and a temporal bone edge line in each frame of the temporomandibular joint image are obtained.

[0078] Step S003: According to the distance between the temporal bone corner points in each frame of the temporomandibular joint image and the coordinate positions of the corner points of the temporomandibular joint image in the coordinate system, a number of mandibular bone corner points and the mandibular bone movement direction are obtained.

[0079] It should be noted that when the mouth is opened and closed, the mandible will slide along the articular fossa, and there will be artifacts of the mandible on the temporomandibular joint image. Therefore, when edge detection is performed on the temporomandibular joint image, multiple edges may appear, and the true mandibular edge cannot be determined. Since the mandible also has identifiable corner points, the movement direction of the mandible in the temporomandibular joint image of adjacent frames can be determined by the change of the corner points.

[0080] It should be noted that, since the corner point changes of the mandible are downward and the sliding amplitude may be relatively large, the corner point changes between adjacent frames of the temporomandibular joint images can be analyzed to determine the local sliding direction of the mandible in a relatively short period of time. When the model is subsequently established, the sliding directions in multiple short periods of time are integrated to determine the local sliding direction of the mandible during the entire opening and closing of the mouth.

[0081] It should be noted that, since the sliding directions of the temporomandibular joint images obtained in adjacent frames are too small and may have errors, the temporomandibular joint images of several adjacent frames are selected.

[0082] Preset quantity threshold is 4, taking each frame and its adjacent preset number threshold The analysis of the temporomandibular joint image of the frame is described as an example.

[0083] It should be noted that, since the mandible will slide, the temporal bone moves relatively little when the mouth is opened and closed, and since this embodiment is mainly used to determine the edge position relationship between the temporal bone and the mandible, the temporal bone is approximated as a standard point, and the sliding direction of the mandible in a preset threshold number of adjacent frames of the temporomandibular joint image is determined.

[0084] It should be noted that, since the condyle of the mandible is located below the articular process of the temporal bone, it can be determined that the distances between the two corner points of the edge of the condyle of the mandible and the articular process are relatively close.

[0085] It should be noted that the positions of the two corners of the temporal bone do not change. Since the condyle of the mandible is located below the articular process of the temporal bone, it can be determined that the distances between the edge of the condyle of the mandible and the two corners of the articular process are relatively close.

[0086] The vector between the relative positions of the same corner point in different temporomandibular joint video frames is the movement direction of the mandible. Since there may be corner points that do not belong to the mandible, the vector difference relative to other corner points is relatively large.

[0087] In each frame of the temporomandibular joint image, the center point of the straight line segment connecting the two temporal bone corners is recorded as the reference point, and the reference point is used as the center of the circle. The distance between two temporal bone corners is the radius, and a semicircle is constructed by rotating clockwise from the horizontal right direction to the horizontal left direction. The corners in the semicircle that are not temporal bone corners are recorded as suspected mandibular corners. is the preset coefficient.

[0088] In the coordinate system, The first frame of the temporomandibular joint image The suspected mandibular angle is The distance and direction of the coordinates between the corner points corresponding to the frame temporomandibular joint image constitute the first The first frame of the temporomandibular joint image vectors of suspected mandibular corner points;

[0089] So far, we have obtained The vector of each suspected mandibular corner point in the frame TMJ image.

[0090] According to The direction of the vector of each suspected mandibular corner point in the frame temporomandibular joint image is used to detect abnormalities of all suspected mandibular corner points using the LOF algorithm to obtain abnormal mandibular corner points; among all suspected mandibular corner points, the suspected mandibular corner points except the abnormal mandibular corner points are recorded as mandibular corner points.

[0091] Among them, the LOF algorithm is a well-known technology, and the specific method will not be introduced here.

[0092] The first The direction of the sum of the vectors of all mandibular corner points in the frame temporomandibular joint image is denoted as The movement direction of the mandible in each frame of the temporomandibular joint image is shown in FIG. Figure 3 shown. Figure 3 The center between the two temporal bone corners is taken as the reference point. There is a mandibular corner below the two temporal bone corners. The motion direction vector is obtained according to the movement of the mandibular corner between different frames.

[0093] At this point, the mandibular movement direction in each frame of the temporomandibular joint image is obtained.

[0094] Step S004: According to the edge line of the mandibular corner points in each frame of the temporomandibular joint image and the mandibular movement direction, the real mandibular edge line in each frame of the temporomandibular joint image is obtained.

[0095] It should be noted that after determining the direction of mandibular movement in each frame of the temporomandibular joint image, it can be determined that the edge of the mandible also has the same direction of movement. At the same time, since the edge of the mandible will not deform, its edge will produce artifacts in the temporomandibular joint image, resulting in multiple edges. The edge may be shortened due to the artifact, but the overall shape of the edge remains fixed. Since the condyle of the mandible usually appears as a relatively rounded protruding structure, the curvature can be used to represent the shape of the condyle of the current mandible. Therefore, the real mandibular edge in multiple adjacent frames of temporomandibular joint images can be determined by combining the direction of mandibular movement and the similarity of mandibular edges between adjacent frames.

[0096] It should be noted that, since the edge cannot directly determine the direction of movement and the edge is round, the circumscribed circle of the edge is obtained to determine the change direction of the center of the circle as the change direction of the current edge.

[0097] The edge line where the mandibular corner point is located in each frame of the temporomandibular joint image is recorded as the mandibular edge line;

[0098] According to The first frame of the temporomandibular joint image The method for obtaining the vectors of the suspected mandibular corner points is to obtain the vector of the center of the minimum circumscribed circle of each mandibular edge line in each frame of the temporomandibular joint image.

[0099] In each frame of the temporomandibular joint image, the vector of the center of the minimum circumscribed circle of each mandibular edge line is recorded as the motion vector of each mandibular edge line.

[0100] The first Frame to All the temporomandibular joint images between frames are recorded as The frame temporomandibular joint image corresponds to the reference temporomandibular joint image;

[0101] The direction of the sum of the motion vectors of all mandibular edge lines in all reference TMJ images is recorded as The movement direction of the mandibular edge line corresponding to the frame temporomandibular joint image;

[0102] No. The calculation method of the similar mandibular margin index of the frame temporomandibular joint image is as follows:

[0103]

[0104] In the formula, Indicates Similar mandibular margin indicators for framed TMJ images; Indicates The minimum angle between the mandibular movement direction corresponding to the frame temporomandibular joint image and the mandibular edge line movement direction; Indicates The mean of the curvatures of all mandibular edge lines in the frame-referenced TMJ image; Indicates The mean of the curvatures of all mandibular edge lines in the frame-referenced TMJ image; is the absolute value function; Indicates the number of reference TMJ images.

[0105] What needs to be explained is that Indicates The minimum angle between the mandibular movement direction corresponding to the frame temporomandibular joint image and the mandibular edge line movement direction indicates that the movement directions are consistent; It represents the difference in the mean curvature of all mandibular edge lines in the reference TMJ images of adjacent frames. The larger the value, the more inconsistent it is.

[0106] According to the above method, the similar mandibular edge index of each frame of the temporomandibular joint image is obtained.

[0107] All frames of temporomandibular joint images are equally divided into several temporomandibular joint image sequence segments. In each temporomandibular joint image sequence segment, the mandibular edge line in the temporomandibular joint image corresponding to the minimum similar mandibular edge index is used as the true mandibular edge line in each frame of the temporomandibular joint image.

[0108] It should be noted that the preset value is 5, and the number of temporomandibular joint images in the equally divided temporomandibular joint image sequence segments is the preset value, which is used as an example for description.

[0109] At this point, all the true mandibular edge lines in the temporomandibular joint image are obtained.

[0110] Step S005: constructing an anatomical structure model of the temporomandibular joint according to the real mandibular edge line and temporal bone edge line of all frames of the temporomandibular joint image.

[0111] According to the real mandibular edge line and temporal bone edge line of all frames of temporomandibular joint images, three-dimensional reconstruction is performed through AI algorithm to generate the anatomical structure model of the temporomandibular joint.

[0112] Among them, AI algorithm is a well-known technology and the specific method will not be introduced here.

[0113] What needs to be explained is that the real mandibular edge of the temporomandibular joint image is determined, and the position changes of the mandibular and temporal bone edges are tracked and recorded by analyzing the video sequence during the opening and closing process; the tracked edge position change data is visualized to establish a schematic diagram of the dynamic changes of the temporomandibular joint during the opening and closing process; and a three-dimensional anatomical structure model of the temporomandibular joint is constructed through the information extracted from the image data.

[0114] It should be noted that based on the anatomical structure model and the dynamic change diagram, doctors can evaluate the functional status of the joint and identify any abnormalities or lesions, such as disc displacement, osteoarthritis, etc. For patients who need surgery, the temporomandibular joint model can be used for surgical simulation, predict the surgical effect, and plan key surgical steps such as incision location and prosthesis placement.

[0115] An embodiment of the present invention also provides a temporomandibular joint surgery simulation system based on AI technology, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above steps S001 to S005 when executing the computer program.

[0116] At this point, this embodiment is completed.

[0117] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A temporomandibular joint surgery simulation method based on AI technology, characterized in that: The method comprises the following steps: Obtaining a plurality of frames of temporomandibular joint images and the corresponding corner points of each corner point in each frame of the temporomandibular joint image in other frames of the temporomandibular joint image; Constructing a coordinate system of each frame of the temporomandibular joint image; on the coordinate system, obtaining the degree of synchronization of the position movement of any two corner points of each frame of the temporomandibular joint image according to the distance between the coordinates of the corner points in adjacent temporomandibular joint images; obtaining the temporal bone corner points and the temporal bone edge lines in each frame of the temporomandibular joint image according to the degree of synchronization and distance of the position movement of any two corner points of each frame of the temporomandibular joint image; According to the distance between the temporal bone corner points in each frame of the temporomandibular joint image and the coordinate position of the corner points of the temporomandibular joint image on the coordinate system, a number of mandibular bone corner points and the mandibular bone movement direction are obtained; According to the edge line of the mandibular corner point in each frame of the temporomandibular joint image and the mandibular movement direction, the real mandibular edge line in each frame of the temporomandibular joint image is obtained; According to the real mandibular edge line and temporal bone edge line of all frames of temporomandibular joint images, an anatomical structure model of the temporomandibular joint is constructed; The specific formula for obtaining the degree of synchronization of the position movement of any two corner points of each frame of the temporomandibular joint image according to the distance between the coordinates of the corner points in adjacent temporomandibular joint images is as follows: In the formula, Represents the first Corner points and The degree of synchronization of the movement of the corner points; Represents the number of all frames of the TMJ image; Indicates The first frame of the temporomandibular joint image The corner point and its The distance between the coordinates of the corner points corresponding to the frame temporomandibular joint image in the coordinate system; Indicates The first frame of the temporomandibular joint image The corner point and its The distance between the coordinates of the corner points corresponding to the frame temporomandibular joint image in the coordinate system; is the absolute value function; The method of obtaining the temporal bone corner points and the temporal bone edge lines in each frame of the temporomandibular joint image according to the synchronization degree and distance of the positions of any two corner points of each frame of the temporomandibular joint image comprises the following specific steps: According to the synchronization degree and distance of the position movement of any two corner points of each frame of the temporomandibular joint image, a movement synchronization index between any two corner points of each frame of the temporomandibular joint image is obtained; According to the movement synchronization index between any two corner points of each frame of the temporomandibular joint image, the temporal bone corner point and the temporal bone edge line in each frame of the temporomandibular joint image are obtained; The movement synchronization index between any two corner points of each frame of the temporomandibular joint image is obtained according to the synchronization degree and distance of the position movement of any two corner points of each frame of the temporomandibular joint image, and the specific formula included is as follows: In the formula, Represents the first Corner points and Movement synchronization indicators between corner points; Represents the first Corner points and The degree of synchronization of the movement of the corner points; Represents the number of all frames of the TMJ image; Indicates Frame TMJ image Corner points and The Euclidean distance between corner points; Represents the first frame of all temporomandibular joint images Corner points and The average of the Euclidean distances between the corner points; is the absolute value function; The method of obtaining the temporal bone corner points and the temporal bone edge lines in each frame of the temporomandibular joint image according to the movement synchronization index between any two corner points of each frame of the temporomandibular joint image comprises the following specific steps: In each frame of the temporomandibular joint image, the two corner points corresponding to the smallest motion synchronization index are recorded as the temporal bone corner points, and the edge line where the temporal bone corner points are located is recorded as the temporal bone edge line; The method of obtaining a plurality of mandibular bone corner points and the mandibular bone movement direction according to the distance between the temporal bone corner points in each frame of the temporomandibular joint image and the coordinate position of the corner points of the temporomandibular joint image on the coordinate system includes the following specific steps: In each frame of the temporomandibular joint image, the center point of the straight line segment connecting the two temporal bone corners is recorded as the reference point, and the reference point is used as the center of the circle. The distance between two temporal bone corners is the radius, and a semicircle is constructed by rotating clockwise from the horizontal right direction to the horizontal left direction. The corners in the semicircle that are not temporal bone corners are recorded as suspected mandibular corners; among them, is the preset coefficient; In the coordinate system, The first frame of the temporomandibular joint image The suspected mandibular angle is The distance and direction of the coordinates between the corner points corresponding to the frame temporomandibular joint image constitute the first The first frame of the temporomandibular joint image vectors of suspected mandibular corner points; According to The direction of the vector of each suspected mandibular corner point in the frame temporomandibular joint image is used to detect abnormalities of all suspected mandibular corner points using the LOF algorithm to obtain abnormal mandibular corner points; among all suspected mandibular corner points, the suspected mandibular corner points except the abnormal mandibular corner points are recorded as mandibular corner points; The first The direction of the sum of the vectors of all mandibular corner points in the frame temporomandibular joint image is denoted as The direction of mandibular motion in the frame TMJ images; The method of obtaining the true mandibular edge line in each frame of the temporomandibular joint image according to the edge line where the mandibular corner point is located and the mandibular movement direction in each frame of the temporomandibular joint image comprises the following specific steps: According to the edge line of the mandibular corner point in each frame of the temporomandibular joint image and the direction of mandibular movement, the similar mandibular edge index of each frame of the temporomandibular joint image is obtained; According to the similar mandibular edge index of each frame of the temporomandibular joint image, the real mandibular edge line in each frame of the temporomandibular joint image is obtained; The method of obtaining a similar mandibular edge index of each frame of the temporomandibular joint image according to the edge line of the mandibular corner point in each frame of the temporomandibular joint image and the mandibular movement direction includes the following specific steps: The edge line where the mandibular corner point is located in each frame of the temporomandibular joint image is recorded as the mandibular edge line; According to The first frame of the temporomandibular joint image A method for obtaining the vectors of the suspected mandibular corner points, and obtaining the vector of the center of the minimum circumscribed circle of each mandibular edge line in each frame of the temporomandibular joint image; In each frame of the temporomandibular joint image, the vector of the center of the minimum circumscribed circle of each mandibular edge line is recorded as the motion vector of each mandibular edge line; The first Frame to All the temporomandibular joint images between frames are recorded as The frame temporomandibular joint image corresponds to the reference temporomandibular joint image; The direction of the sum of the motion vectors of all mandibular edge lines in all reference TMJ images is recorded as The movement direction of the mandibular edge line corresponding to the frame temporomandibular joint image; No. The calculation method of the similar mandibular margin index of the frame temporomandibular joint image is as follows: In the formula, Indicates Similar mandibular margin indicators for framed TMJ images; Indicates The minimum angle between the mandibular movement direction corresponding to the frame temporomandibular joint image and the mandibular edge line movement direction; Indicates The mean of the curvatures of all mandibular edge lines in the frame-referenced TMJ image; Indicates The mean of the curvatures of all mandibular edge lines in the frame-referenced TMJ image; is the absolute value function; represents the number of reference TMJ images; The method of obtaining the true mandibular edge line in each frame of the temporomandibular joint image according to the similar mandibular edge index of each frame of the temporomandibular joint image comprises the following specific steps: All frames of temporomandibular joint images are equally divided into several temporomandibular joint image sequence segments. In each temporomandibular joint image sequence segment, the mandibular edge line in the temporomandibular joint image corresponding to the minimum similar mandibular edge index is used as the true mandibular edge line in each frame of the temporomandibular joint image.

2. A temporomandibular joint surgery simulation system based on AI technology, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by a processor, the steps of the temporomandibular joint surgery simulation method based on AI technology as claimed in claim 1 are implemented.

Citation Information

Patent Citations

  • Temporomandibular joint movement reconstruction method and system thereof

    CN106875432A

  • Method for detecting the movement of a temporomandibular joint

    CN110022792A

  • Temporomandibular joint motion reconstruction method based on multi-modal information fusion

    CN114708312A

  • Digital three-dimensional measurement and evaluation method for temporomandibular joint

    CN114998349A

  • Temporomandibular joint motion analog simulation method

    CN116227235A