Vestibular dysfunction posture control evaluation method and evaluation system

Abstract

The invention discloses a vestibular dysfunction posture control evaluation method and evaluation system, and belongs to the technical field of medical treatment, and the method comprises the following steps: S1, a three-dimensional measurement camera collects an omnibearing human body image of an evaluator in a fixed state; s2, constructing a first human body three-dimensional model; s3, the three-dimensional measurement camera collects a reset all-around human body image after the assessor moves; s4, constructing a second human body three-dimensional model; s5, calculating the body deviation degree of the evaluator based on the first human body three-dimensional model and the second human body three-dimensional model; s6, evaluating the posture control advantage and disadvantage degree of the evaluator based on the calculated body deviation degree of the evaluator, namely the vestibular dysfunction posture control advantage and disadvantage degree; according to the method, any instrument does not need to be bound on the body of the evaluator, the evaluator cannot be influenced in the aspect of body posture control, meanwhile, the influence of swinging of the evaluator in the body posture control process is avoided, and the evaluation precision of the vestibular dysfunction posture control quality degree can be improved.

Description

Technical Field

[0001] The present invention belongs to the field of medical technology, and in particular relates to a vestibular dysfunction posture control assessment method and an assessment system. Background Art

[0002] The vestibular system is an important sensory system in the human body, which is mainly responsible for balance and spatial positioning. It is located in the inner ear and includes structures such as three semicircular canals, the utricle and the saccule.

[0003] When the vestibular system is affected by factors such as disease, trauma or aging, functional disorders may occur, manifested as symptoms such as dizziness, imbalance and spatial orientation disorder. Vertigo refers to the feeling that surrounding objects are rotating or that one is rotating, often accompanied by symptoms such as nausea and vomiting. Imbalance disorder refers to unstable standing, easy falling, and unsteady steps when walking. Spatial orientation disorder refers to the inability to control one's own posture and position relative to surrounding objects.

[0004] Objective and quantitative assessment of the degree of symptoms such as vertigo, imbalance and positioning disorders is important for determining the extent of vestibular dysfunction and evaluating the effectiveness of treatment.

[0005] The current assessment of spatial positioning disorders is generally to first bind angle sensors to the joints of the assessee, then fix the assessee's body position angle, and then have the assessee actively move for a period of time before returning to the body position angle. The degree of posture control is then assessed based on the difference displayed by the angle sensor.

[0006] The existing technology has the following problems: the existing method of evaluating spatial positioning disorders, that is, the evaluation method of controlling one's own posture, will affect the evaluator's movements to a certain extent because the angle sensor is bound to the evaluator, and the accuracy of the angle sensor will be affected by even the slightest swing of the body, which can easily affect the evaluation results and reduce the accuracy of the evaluation.

[0007] In view of this, a vestibular dysfunction posture control assessment method and assessment system were designed to solve the above problems. Summary of the invention

[0008] To solve the problems raised in the above background technology. The present invention provides a vestibular dysfunction posture control assessment method and assessment system, which has the characteristics of not needing to bind any instrument to the assessor's body, that is, not affecting the assessor's body posture control, and not being affected by the swing of the assessor's body posture control process, and can improve the assessment accuracy of the vestibular dysfunction posture control quality.

[0009] Another object of the present invention is to provide a vestibular dysfunction posture control assessment system.

[0010] To achieve the above object, the present invention provides the following technical solution: a method for evaluating posture control of vestibular dysfunction, comprising the following steps:

[0011] S1: The evaluator's body is actively fixed at a certain angle, and a 3D measurement camera is used to collect a full range of human body images of the evaluator;

[0012] S2: constructing a first human body three-dimensional model based on the collected omnidirectional human body images of the evaluator;

[0013] S3: The assessor actively moves, and after a period of time, the body returns to the fixed angle of step S1 through the sense of the assessor, and the 3D measurement camera collects the assessor's full range of body images;

[0014] S4: constructing a second human body three-dimensional model based on the collected omnidirectional human body images of the evaluator;

[0015] S5: calculating the degree of body deviation of the evaluator based on the first three-dimensional human body model and the second three-dimensional human body model;

[0016] S6: Evaluate the degree of the assessor's postural control based on the calculated degree of the assessor's body deviation, that is, the degree of vestibular dysfunction postural control.

[0017] Furthermore, the specific steps of step S2 include:

[0018] S201: preprocessing the collected multiple human body images, including denoising;

[0019] S202: extracting feature points of multiple human body images;

[0020] S203: Matching feature points of multiple human body images to find corresponding feature points between the multiple human body images;

[0021] S204: Calculating the three-dimensional coordinates of the feature points of the plurality of human body images based on the matching results of the feature points and the parameters of the three-dimensional measurement camera;

[0022] S205: converting the three-dimensional coordinates of the feature points of the plurality of human body images into point cloud data;

[0023] S206: Registering and fusing the point cloud data to obtain a first human body three-dimensional model.

[0024] Furthermore, in step S4, the steps of constructing the second human body three-dimensional model are the same as the steps of constructing the first human body three-dimensional model.

[0025] Furthermore, the specific steps of step S5 include:

[0026] S501: Converting the first human body 3D model and the second human body 3D model into point cloud data;

[0027] S502: extracting edge lines of two model point cloud data;

[0028] S503: Taking each edge point of the edge line as a feature point, a local coordinate system is established around the feature point;

[0029] S504: Calculate the rotation matrix between the local coordinate systems of the feature points of the two models. The expression of the rotation matrix is:

[0030]

[0031] Where: θ represents the rotation angle, I represents the 3×3 unit matrix, u represents the rotation axis unit vector, It is represented as an outer product calculation, and [u]x is represented as a 3×3 antisymmetric matrix;

[0032] S505: Calculate the degree of angular deviation of the feature points based on the angle between the rotation matrices of the feature points of the two models. The expression is:

[0033]

[0034] Wherein: R represents the product of two transformation matrices, that is, R=R1·R2, where R1 and R2 represent the rotation matrices of two model feature points.

[0035] Furthermore, the specific steps of step S6 include:

[0036] S601: Based on the guidance of professionals or big data analysis, a body deviation degree-posture control quality correlation table is preset;

[0037] S602: Based on the calculated body deviation degree of the evaluator and the body deviation degree-posture control quality association table, the postural control quality of the evaluator, that is, the vestibular dysfunction postural control quality, is evaluated.

[0038] The evaluation system of the vestibular dysfunction posture control evaluation method comprises:

[0039] The human body image acquisition module makes the assessor's body actively fixed at a certain angle, and the three-dimensional measurement camera collects the assessor's full-scale human body image. At the same time, the assessor actively moves, and after a period of time, the body returns to the above fixed angle through its own senses, and the three-dimensional measurement camera collects the assessor's full-scale human body image;

[0040] A human body three-dimensional model building module, which builds a first human body three-dimensional model and a second human body three-dimensional model based on the collected omnidirectional human body images of the evaluator;

[0041] A body deviation degree calculation module, which calculates the body deviation degree of the evaluator based on the first human body three-dimensional model and the second human body three-dimensional model;

[0042] The posture control degree assessment module assesses the posture control quality of the assessor based on the calculated body deviation degree of the assessor, that is, the posture control quality of vestibular dysfunction.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The present invention collects omnidirectional human body images of an evaluator before and after movement resetting based on a three-dimensional measuring camera, then constructs a three-dimensional human body model based on the omnidirectional human body images in the two states, then calculates the degree of body deviation of the evaluator in the two states based on the two three-dimensional human body models, and then evaluates the degree of body posture control of the evaluator based on the degree of body deviation. Compared with the prior art, there is no need to bind any instrument to the evaluator's body, that is, it will not affect the evaluator's body posture control, and will not be affected by the swing of the evaluator's body posture control process, which can improve the evaluation accuracy of the degree of posture control of vestibular dysfunction. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a flow chart of a method for evaluating posture control of vestibular dysfunction according to the present invention;

[0046] Figure 2 It is a framework diagram of the vestibular dysfunction posture control evaluation system of the present invention;

[0047] In the figure: 1. Human body image acquisition module; 2. Human body three-dimensional model construction module; 3. Body deviation degree calculation module; 4. Posture control degree evaluation module. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] The present invention provides the following technical solutions: Figure 1 , a method for assessing postural control of vestibular dysfunction, comprising the following steps:

[0050] S1: The evaluator's body is actively fixed at a certain angle, and a 3D measurement camera is used to collect a full range of human body images of the evaluator;

[0051] S2: constructing a first human body three-dimensional model based on the collected omnidirectional human body images of the evaluator;

[0052] S3: The assessor actively moves, and after a period of time, the body returns to the fixed angle of step S1 through the sense of the assessor, and the 3D measurement camera collects the assessor's full range of body images;

[0053] S4: constructing a second human body three-dimensional model based on the collected omnidirectional human body images of the evaluator;

[0054] S5: calculating the degree of body deviation of the evaluator based on the first three-dimensional human body model and the second three-dimensional human body model;

[0055] S6: Evaluate the degree of the assessor's postural control based on the calculated degree of the assessor's body deviation, that is, the degree of vestibular dysfunction postural control.

[0056] Specifically, the specific steps of step S2 include:

[0057] S201: preprocessing the collected multiple human body images, including denoising;

[0058] The denoising preprocessing adopts the fast Fourier transform method to convert the human body image from the time domain to the frequency domain and denoise it in the frequency domain;

[0059] S202: extracting feature points of multiple human body images;

[0060] S203: Matching feature points of multiple human body images to find corresponding feature points between the multiple human body images;

[0061] The feature point extraction and matching of multiple human images adopts SIFT method, which extracts feature points and performs feature point matching by obtaining feature points in human images and their related scale and direction descriptors;

[0062] S204: Calculating the three-dimensional coordinates of the feature points of the plurality of human body images based on the matching results of the feature points and the parameters of the three-dimensional measurement camera;

[0063] The above can be refined as follows: calibrating the 3D measurement camera to obtain the internal parameters and external parameters of the 3D measurement camera;

[0064] The projection matrix is ​​constructed according to the internal and external parameters of the 3D measurement camera, which is a 3×4 matrix, and converts the 3D world coordinates into 2D image coordinates;

[0065] For each pair of matching feature points, the following equations are established:

[0066] sp=TP

[0067] Where: s represents the scale factor, p represents the pixel coordinates of the feature point in the human body image, T represents the projection matrix of the 3D measurement camera, and P represents the coordinates of the feature point in the 3D world coordinate system;

[0068] Solve P, which is the three-dimensional coordinates of the feature points of the human body image;

[0069] S205: converting the three-dimensional coordinates of the feature points of the plurality of human body images into point cloud data;

[0070] The three-dimensional coordinates of the feature points of the multiple human body images are calculated and organized according to the point cloud data structure definition, wherein the point cloud data structure definition is generally stored in the form of an array or a list, wherein each element represents the coordinate information of a point;

[0071] S206: registering and fusing the point cloud data to obtain a first human body three-dimensional model;

[0072] The registration and fusion of point cloud data adopts the ICP method, that is, iterative closest point registration. For each point in the point cloud to be registered, the KD tree is used to find the nearest point in the reference point cloud for matching. Based on the matching point pairs, the optimal registration transformation matrix is ​​calculated by the least squares method. The point cloud data set to be registered is transformed by the registration transformation matrix, the point cloud position is updated, and the iteration is continued until the iteration is terminated to complete the point cloud registration and fusion.

[0073] Specifically, in step S4, the steps of constructing the second human body three-dimensional model are the same as the steps of constructing the first human body three-dimensional model.

[0074] Specifically, the specific steps of step S5 include:

[0075] S501: Converting the first human body 3D model and the second human body 3D model into point cloud data;

[0076] S502: extracting edge lines of two model point cloud data;

[0077] The above can be refined as follows: search for neighboring points of each point cloud data based on the KD tree;

[0078] For each point in the point cloud data, the covariance matrix is ​​calculated based on the neighboring points. The expression is:

[0079]

[0080] Where: k represents the number of points, p i Represented as neighboring points, Represented as the centroid of the point set;

[0081] Perform eigenvalue decomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors. The eigenvalue decomposition formula is:

[0082] C=VΛV T

[0083] Where: V represents the eigenvector matrix, Λ represents the eigenvalue diagonal matrix;

[0084] Here, the eigenvector V corresponding to the minimum eigenvalue is the normal vector;

[0085] For each point in the point cloud data, the normal vector curvature between it and the neighboring points is calculated. The expression is:

[0086]

[0087] Where: n represents the normal vector, p1, p2 and p3 represent three adjacent points in the point cloud data;

[0088] If the normal vector curvature of a point cloud data point exceeds a preset threshold, the point is marked as a boundary point;

[0089] The detected boundary points are connected together by the least square method to form edge lines;

[0090] S503: Taking each edge point of the edge line as a feature point, a local coordinate system is established around the feature point;

[0091] Calculate the tangent vector of the feature point, and establish a local coordinate system centered on the feature point based on the tangent vector and normal vector of the feature point;

[0092] The tangent vector is obtained by taking the direction of the tangent vector perpendicular to the normal vector n as the first tangent vector t1, and the second tangent vector t2 perpendicular to the first tangent vector t1 is obtained by the cross product of the normal vector n and the first tangent vector t1, that is, t2 = n × t1;

[0093] S504: Calculate the rotation matrix between the local coordinate systems of the feature points of the two models. The expression of the rotation matrix is:

[0094]

[0095] Where: θ represents the rotation angle, I represents the 3×3 unit matrix, u represents the rotation axis unit vector, It is represented as an outer product calculation, and [u]x is represented as a 3×3 antisymmetric matrix;

[0096] S505: Calculate the degree of angular deviation of the feature points based on the angle between the rotation matrices of the feature points of the two models. The expression is:

[0097]

[0098] Wherein: R represents the product of two transformation matrices, that is, R=R1·R2, where R1 and R2 represent the rotation matrices of two model feature points.

[0099] Specifically, the specific steps of step S6 include:

[0100] S601: Based on the guidance of professionals or big data analysis, a body deviation degree-posture control quality correlation table is preset;

[0101] S602: Based on the calculated body deviation degree of the evaluator and the body deviation degree-posture control quality association table, the postural control quality of the evaluator, that is, the vestibular dysfunction postural control quality, is evaluated.

[0102] See attached Figure 2 , an assessment system for the vestibular dysfunction postural control assessment method, including:

[0103] The human body image acquisition module 1 makes the assessor's body actively fixed at a certain angle, and uses a three-dimensional measurement camera to collect the assessor's full-scale human body image. At the same time, the assessor actively moves, and after a period of time, the body is made to return to the above fixed angle through its own senses, and the assessor's full-scale human body image is collected through the three-dimensional measurement camera;

[0104] A human body three-dimensional model building module 2 is used to build a first human body three-dimensional model and a second human body three-dimensional model based on the collected omnidirectional human body images of the evaluator;

[0105] A body deviation degree calculation module 3, which calculates the body deviation degree of the evaluator based on the first human body three-dimensional model and the second human body three-dimensional model;

[0106] The posture control degree evaluation module 4 evaluates the posture control quality of the evaluator based on the calculated body deviation degree of the evaluator, that is, the posture control quality of vestibular dysfunction.

[0107] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Method for evaluating posture control of vestibular dysfunction, characterized in that The following steps are involved: S1: Make the evaluator's body actively fix at a certain angle, and collect the evaluator's full-range human image through a three-dimensional measurement camera; S2: Construct the first three-dimensional model of human body based on the all-round human body images of the acquired evaluator; S3: The evaluator actively moves, and after a period of time, the body returns to the fixed angle of step S1 through its own sensation, and collects the evaluator's omni-dimensional human image through a three-dimensional measurement camera; S4: Construct a second human body three-dimensional model based on the all-round human body images of the acquired evaluator; S5: Calculate the degree of body deviation of the evaluator based on the first human body three-dimensional model and the second human body three-dimensional model; S6: Assess the evaluator's pros and cons, i.e. the vastibule dysfunction posture control is based on the calculated evaluator's degree of physical deviation.

2. The method of evaluating posture control of vestibular dysfunction according to claim 1, characterized in that: The specific steps of step S2 include: S201: Preprocess multiple human images collected, including denoising; S202: extracting feature points of multiple human body images; S203: Match the feature points of multiple human body images and find the corresponding feature points between multiple human body images; S204: Calculate the three-dimensional coordinates of feature points of multiple human body images based on the matching results of feature points and the parameters of the three-dimensional measurement camera; S205: Convert the three-dimensional coordinates of feature points of multiple human body images to point cloud data; S206: Register and fusion of point cloud data to obtain the first human body three-dimensional model.

3. The method of evaluating posture control of vestibular dysfunction according to claim 2, characterized in that: In the step S4, the construction step of the second human body three-dimensional model is the same as the construction step of the first human body three-dimensional model.

4. The method of evaluating posture control of vestibular dysfunction according to claim 1, characterized in that: The specific steps of step S5 include: S501: Convert the first human body three-dimensional model and the second human body three-dimensional model into point cloud data; S502: Extract edge lines of point cloud data of two models; S503: Use each edge point of the edge line as the feature point to establish a local coordinate system around the feature point; S504: Calculate the rotation matrix between the local coordinate system of the feature points of the two models. The expression of the rotation matrix is: Where: θ represents the rotation angle, I represents the unit matrix of 3×3, u represents the unit vector of rotation axis, Denoted as an outer product calculation, [u]x is represented as an antisymmetric matrix of 3×3; S505: Calculate the degree of angular deviation of the feature points based on the angle between the rotation matrices of the feature points of the two models. The expression is: In the formula: R is represented as the product of two transformation matrices, that is, R=R1·R2, where R1 and R2 are represented as rotation matrixes of two model feature points.

5. The method of evaluating posture control of vestibular dysfunction according to claim 1, characterized in that: The specific steps of step S6 include: S601: Based on professional guidance or big data analysis, preset the degree of physical deviation-level correlation table of posture control; S602: Based on the calculated body deviation degree of the evaluator and the body deviation degree-posture control quality association table, the postural control quality of the evaluator, that is, the vestibular dysfunction postural control quality, is evaluated.

6. The evaluation system of the vestibular dysfunction posture control evaluation method according to any one of claims 1-5, characterized in that, include: The human body image acquisition module (1) makes the assessor's body actively fixed at a certain angle, and uses a three-dimensional measurement camera to collect a full range of human body images of the assessor. At the same time, the assessor actively moves, and after a period of time, the body is made to return to the above fixed angle through its own senses, and the three-dimensional measurement camera collects a full range of human body images of the assessor; The human body three-dimensional model construction module (2), constructing the first human body three-dimensional model and the second human body three-dimensional model based on the all-round human body images of the collected evaluator; The body deviation degree calculation module (3), calculates the body deviation degree of the evaluator based on the first human body three-dimensional model and the second human body three-dimensional model; The postural control degree evaluation module (4) evaluates the evaluator's postural control excellence and disadvantages based on the calculated evaluator's physical deviation, that is, the vestibule dysfunction posture control excellence and disadvantages.

Images