Teenager scoliosis detection method and system based on three-dimensional imaging

Retrieval and processing of back point cloud data through three-dimensional structured light technology, the shortcomings of manual screening and X-ray detection are solved, and efficient and accurate adolescent scoliosis detection is achieved without contact and radiation, reducing the rate of misdiagnosis and protecting privacy .

CN118967585APending Publication Date: 2024-11-15SHAOXING BEYOND MEDICAL TECH CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410969457.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-18
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

In the prior art, artificial screening of scoliosis in adolescents has insufficient privacy protection, large labor and material resources, and high misdiagnosis rate. The method based on the body surface measuring instrument has limitations. X-ray detection has radiation hazards and is not suitable for preliminary screening. Check stage.

Method used

Three-dimensional structured light technology is used to obtain back point cloud data, and key point data are obtained through preprocessing and annotation, including waist point height difference, scapula height difference, spinal midline and lumbar symmetry data, and compared with preset thresholds to determine back morphology. Reduce the misdiagnosis rate, improve detection efficiency, and protect the privacy of the detection objects.

Benefits of technology

Three-dimensional imaging technology captures back morphology and structural information, reduces the misdiagnosis rate, improves detection accuracy and efficiency, and provides contactless and radiation-free detection methods, which are suitable for preliminary screening of scoliosis in adolescents.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118967585A_ABST
    Figure CN118967585A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of teenager scoliosis detection, and particularly provides a teenager scoliosis detection method and system based on three-dimensional imaging, and the method comprises the steps: obtaining back point cloud data through a three-dimensional structured light technology; preprocessing the back point cloud data to obtain target point cloud data; labeling the target point cloud data through a label tool to obtain human body back data; inputting the human body back data into a back segmentation model to obtain key point data; the key point data comprises waist fossa point height difference data, shoulder blade height difference data, spine center line data, shoulder blade symmetry value data and flank symmetry data; and comparing the key point data with a preset threshold value, and determining the back form. The seven initial key point data are processed through a three-dimensional geometric algorithm, a visual result is presented, the human body posture can be rapidly evaluated, and the detection accuracy and efficiency are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of adolescent scoliosis detection, and in particular to an adolescent scoliosis detection method and system based on three-dimensional imaging. Background Art

[0002] Adolescent scoliosis is a common structural abnormality of the spine that occurs primarily during adolescence between the ages of 10 and 18. It manifests as a lateral curvature and twisting of the spine in three dimensions, often accompanied by trunk asymmetry and imbalance between the shoulders and hips. Without prompt intervention and treatment, scoliosis may worsen, leading to physical dysfunction and psychological distress. Early detection, regular spinal screening, and appropriate treatment are key to helping adolescents reduce symptoms, improve posture, and enhance quality of life.

[0003] Manual screening for adolescent scoliosis is the most traditional screening method, which requires doctors to undergo professional training before conducting contact screening. However, there is insufficient privacy protection, and it consumes a lot of manpower, material resources and time, with a high misdiagnosis rate. Although some electronic ruler methods such as surface spine measuring instruments have reduced the false positive rate, the detection process still has many limitations. Although X-ray-based detection methods can intuitively determine whether there are abnormalities in the spine, this screening method involving radiation has certain hazards to the growth of adolescents and is not suitable for the initial screening stage of adolescent scoliosis.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] To this end, the technical problem to be solved by the present invention is that manual screening for adolescent scoliosis is the most traditional screening method, which requires doctors to undergo professional training and conduct contact screening, but there is insufficient privacy protection, and it consumes a lot of manpower, material resources and time, and has a high misdiagnosis rate; although some electronic ruler methods based on surface spine measuring instruments have reduced some false positive rates, the detection process still has many limitations; although the detection method based on X-rays can intuitively determine whether there is an abnormality in the spine, this screening method with the presence of radiation has certain hazards to the growth of adolescents and is not suitable for the initial screening stage of adolescent scoliosis.

[0006] In order to solve the above technical problems, the first aspect of the present invention provides a method for detecting adolescent scoliosis based on three-dimensional imaging, the method comprising:

[0007] Acquire back point cloud data through 3D structured light technology;

[0008] Preprocessing the back point cloud data to obtain target point cloud data;

[0009] The target point cloud data is labeled by using the labelme tool to obtain human back data;

[0010] Inputting the human back data into a back segmentation model to obtain key point data; the key point data includes waist dimple height difference data, shoulder blade height difference data, spine midline data, shoulder blade symmetry value data and waist dimple symmetry data;

[0011] The key point data is compared with a preset threshold to determine the back shape.

[0012] In one embodiment of the present invention, the step of acquiring back point cloud data by three-dimensional structured light technology includes:

[0013] The human back is modeled with three-dimensional imaging based on the sinusoidal stripes and Gray code projection technology of DLP projection to obtain the back point cloud data.

[0014] In one embodiment of the present invention, the step of preprocessing the back point cloud data to obtain target point cloud data includes:

[0015] Denoising the back point cloud data to obtain denoised data;

[0016] Projecting the denoised data onto a plane to obtain a triangular facet;

[0017] Calculate the normal of the triangular face to obtain a normal map;

[0018] The normal map is segmented to obtain target point cloud data.

[0019] In one embodiment of the present invention, the step of projecting the denoised data onto a plane to obtain a triangular facet comprises:

[0020] Obtaining pixels of the noise reduction data and counting them to obtain the number of rows and columns of a two-dimensional dot matrix;

[0021] Form a quaternion according to the number of rows and the number of columns, and find a corresponding three-dimensional point in the three-dimensional point cloud set;

[0022] Constructing a quadrilateral according to the four three-dimensional points;

[0023] The quadrilateral is divided according to diagonals of a preset three-dimensional distance to obtain triangular facets.

[0024] In one embodiment of the present invention, the step of calculating the normal of the triangular facet to obtain a normal map includes:

[0025] Calculating the triangular face to obtain a face normal;

[0026] Normalizing the surface normal to obtain normalized data;

[0027] The normalized data is linearly mapped to a preset range according to the minimum and maximum values ​​of the normal grayscale value to obtain a normal map.

[0028] In one embodiment of the present invention, the step of inputting the human back data into the back segmentation model to obtain key point data further includes:

[0029] Inputting the human back data into the back segmentation model to obtain initial key point data; the initial key point data includes vertebral data, left lumbar fossa data, right lumbar fossa data, left scapula upper edge data, left scapula lower edge data, right scapula upper edge data, and right scapula lower edge data;

[0030] Calculate the height difference data of the waist dimple points according to the left waist dimple data and the right waist dimple data;

[0031] Calculate the scapula height difference data according to the left scapula lower edge data and the right scapula lower edge data;

[0032] Dividing the back point cloud data into a plurality of slices;

[0033] Fitting the point cloud data obtained by the slicing a preset number of times to obtain a curve;

[0034] Derivative the curve to obtain extreme points;

[0035] Calculate the center of gravity of the extreme point to obtain the midpoint of the spine;

[0036] Fitting the midpoint of the spine to obtain the midline data of the spine;

[0037] Calculating according to the spinal midline to obtain the scapula symmetry value data;

[0038] The symmetry data of the lumbar dimples is obtained by performing calculation according to the left scapula lower edge data and the left lumbar dimple data.

[0039] In one embodiment of the present invention, the step of comparing the key point data with a preset threshold to determine the back shape includes:

[0040] If the key point data is greater than a preset threshold, the back morphology is abnormal;

[0041] If the key point data is less than or equal to the preset threshold, the back shape is normal.

[0042] The second aspect of the present invention provides a juvenile scoliosis detection system based on three-dimensional imaging, the system comprising: a data acquisition module, a calculation module and a data processing module;

[0043] The data acquisition module is configured to: acquire back point cloud data through three-dimensional structured light technology;

[0044] The computing module is configured to: pre-process the back point cloud data to obtain target point cloud data; label the target point cloud data using the labelme tool to obtain human back data;

[0045] The data processing module is configured to: input the human back data into a back segmentation model to obtain key point data; the key point data includes lumbar dimple height difference data, scapula height difference data, spinal midline data, scapula symmetry value data and lumbar dimple symmetry data; compare the key point data with a preset threshold to determine the back shape.

[0046] The third aspect of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in the first aspect or any possible implementation of the first aspect is implemented.

[0047] A fourth aspect of the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect or any possible implementation of the first aspect.

[0048] The above technical solution of the present invention has the following advantages compared with the prior art:

[0049] The method and system for detecting adolescent scoliosis based on three-dimensional imaging described in the present invention, by applying three-dimensional structured light technology to human back detection, can capture the overall morphology and structural information of the back and spine compared to electronic rulers and ultrasonic methods, reduce the misdiagnosis rate, improve detection efficiency, and help doctors better grasp the back and spine information, providing assistance for subsequent diagnosis. The seven initial key point data are processed by a three-dimensional geometric algorithm to present a visual result, which can quickly evaluate the human body posture and improve the detection accuracy and efficiency. The point cloud image is converted into an image to be segmented, making the texture features of the back more obvious, facilitating AI processing, accelerating the calculation of normals, and also protecting the privacy of the detected object. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to make the contents of the present invention more clearly understood, the present invention is further described in detail below based on specific embodiments of the present invention in conjunction with the accompanying drawings.

[0051] Figure 1 It is a flow chart of a method and system for detecting adolescent scoliosis based on three-dimensional imaging provided by the present invention;

[0052] Figure 2 It is a schematic diagram of a method for detecting adolescent scoliosis based on three-dimensional imaging and point cloud slice fitting in the system provided by the present invention;

[0053] Figure 3 This is a system architecture diagram of a method and system for detecting adolescent scoliosis based on three-dimensional imaging provided by the present invention. DETAILED DESCRIPTION

[0054] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0055] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0056] Reference Figure 1 As shown, the present invention provides a method for detecting adolescent scoliosis based on three-dimensional imaging, the method comprising:

[0057] S100, obtains back point cloud data through 3D structured light technology;

[0058] In step S100, three-dimensional imaging modeling of the human back is performed based on the sinusoidal stripes and Gray code projection technology of DLP projection to obtain back point cloud data.

[0059] In actual application scenarios, 3D structured light technology is a non-contact, non-radiation measurement method that uses optical projection and image acquisition to obtain 3D depth information of the human back. This application uses DLP projection's sinusoidal stripes and Gray code projection technology for 3D imaging, models the human back, and obtains back point cloud data. Compared with electronic rulers and ultrasonic methods, the application of 3D structured light technology to human back detection can capture the overall morphology and structural information of the back and spine, reduce the misdiagnosis rate, improve detection efficiency, and help doctors better understand the back and spine information, providing assistance for subsequent diagnosis.

[0060] S200, preprocessing the back point cloud data to obtain target point cloud data;

[0061] In step S200, the steps of preprocessing the back point cloud data to obtain target point cloud data include: denoising the back point cloud data to obtain denoised data; projecting the denoised data onto a plane to obtain a triangular facet; calculating the normal of the triangular facet to obtain a normal map; and segmenting the normal map to obtain target point cloud data.

[0062] The step of projecting the denoised data onto a plane to obtain a triangular patch includes: obtaining pixels of the denoised data and counting them to obtain the number of rows and columns of a two-dimensional dot matrix; forming a quadrilateral based on the number of rows and the number of columns, and finding a corresponding three-dimensional point in a three-dimensional point cloud set; forming a quadrilateral based on the four three-dimensional points; and dividing the quadrilateral according to diagonals of a preset three-dimensional distance to obtain a triangular patch.

[0063] The steps of calculating the normal of the triangular face patch and obtaining a normal map include: calculating the triangular face patch to obtain the face normal; normalizing the face normal to obtain normalized data; linearly mapping the normalized data to a preset range according to the minimum and maximum values ​​of the normal grayscale value to obtain a normal map.

[0064] In actual application scenarios, the back point cloud data is denoised to obtain denoised data. In the denoising process, bilateral filtering or Gaussian filtering can be used to remove noise. The pixels of the denoised data are obtained and counted to obtain the number of rows and columns of the two-dimensional dot matrix. That is, the number of rows and columns of the two-dimensional dot matrix is ​​determined by counting the pixels in the measured back image, and then a corresponding quadruple is formed according to the coordinates of each point, and the corresponding three-dimensional point is found in the three-dimensional point cloud set. The quadrilateral composed of four three-dimensional points is divided along the diagonal with a preset three-dimensional distance, wherein the preset three-dimensional distance is a shorter three-dimensional distance, and a triangular face can be obtained. Finally, the face normals of these triangular face are calculated. Each obtained face normal is normalized to obtain normalized data. According to the minimum and maximum values ​​of the normal grayscale value, the normalized data is linearly mapped to a preset range to obtain a normal map, wherein the preset range is 0-255, so as to retain the relative brightness and contrast of the normal, making it easier to visualize and analyze. The normal map is segmented to obtain the target point cloud data.

[0065] S300, labeling the target point cloud data by using the labelme tool to obtain human back data;

[0066] In step S300, the labelme tool is usually used to label, annotate or classify data. By using this tool to operate and annotate in the target point cloud data, the part belonging to the back of the human body is identified and separated from the numerous data points, thereby obtaining point cloud data specifically about the back of the human body, that is, obtaining the back of the human body data.

[0067] S400, inputting the human back data into a back segmentation model to obtain key point data; the key point data includes waist dimple height difference data, shoulder blade height difference data, spine midline data, shoulder blade symmetry value data and waist dimple symmetry data;

[0068] In step S400, the human back data is input into the back segmentation model to obtain initial key point data; the initial key point data includes vertebral prominence data, left lumbar fossa data, right lumbar fossa data, left scapula upper edge data, left scapula lower edge data, right scapula upper edge data, and right scapula lower edge data; the lumbar fossa point height difference data is obtained by calculation based on the left lumbar fossa data and the right lumbar fossa data; the scapula height difference data is obtained by calculation based on the left scapula lower edge data and the right scapula lower edge data; the back point cloud data is divided into multiple slices; the point cloud data obtained from the slices is fitted a preset number of times to obtain a curve; the curve is derived to obtain extreme points; the center of gravity of the extreme points is calculated to obtain the midpoint of the spine; the midpoint of the spine is fitted to obtain the spinal midline data; the scapula symmetry value data is obtained by calculation based on the spinal midline; the lumbar fossa symmetry data is obtained by calculation based on the left scapula lower edge data and the left lumbar fossa data.

[0069] In the actual application scenario, the back segmentation model is trained by the YOLOv8 key point detection model to obtain 7 initial key point data; the initial key point data include vertebral prominence data (A), left lumbar dimple data (CL), right lumbar dimple data (CR), left scapula upper edge data (DLU), left scapula lower edge data (DLD), right scapula upper edge data (DRU), and right scapula lower edge data (DRD).

[0070] Subtract the ordinate of the left lumbar dimple data (CL) from the ordinate of the right lumbar dimple data (CR) to calculate the height difference data of the lumbar dimple points. The specific calculation formula is shown in formula (1).

[0071] C d =abs(Y CL -Y CR ) (1);

[0072] Among them, Y CL Y is the ordinate of the left lumbar fossa data (CL), CRIt is the vertical coordinate of the right lumbar data (CR).

[0073] The ordinate of the left scapula lower edge data (DLD) is subtracted from the ordinate of the right scapula lower edge data (DRD) to obtain the scapula height difference data. The specific calculation formula is shown in formula (2).

[0074] D d =abs(Y DLD -Y DRD ) (2);

[0075] Among them, Y DLD Y is the ordinate of the lower edge of the left scapula (DLD), DRD It is the ordinate of the lower edge data of the right scapula (DRD).

[0076] The back point cloud data is divided into several slices. According to the characteristic that the human spine is located in the concave area in the middle of the back, the point cloud data obtained from each slice is fitted, such as Figure 2 As shown, after a preset number of fittings, a curve is obtained, wherein the preset number is five. The curve is differentiated to obtain the extreme points, and then the center of gravity is obtained for all points. The extreme point closest to the center of gravity is determined to be the midpoint of the spine. The obtained midpoint of the spine is fitted to obtain the spinal midline data.

[0077] Calculate on the fitted spine midline respectively, that is, find the points P and Q with the same ordinates of the left scapula lower edge data DLD and the right scapula lower edge data DRD, and calculate the distance D from each point P and D Q , compare the sizes of the two, divide the smaller value by the larger value, and get the scapula symmetry value data.

[0078] Draw a horizontal straight line EF through the left shoulder blade lower edge data DLD and intersect the back contour at points E and F. Draw a horizontal straight line GH through the left lumbar dimple data (CL) and intersect the back contour at points G and H. Draw a vertical line at the midpoint I of EF and intersect GH at J. Calculate the areas S1 and S2 enclosed by EIJG and IFHJ respectively, and then calculate the overlapping area S3 of the two. Calculate the symmetry data of the dimples.

[0079] S500, comparing the key point data with a preset threshold to determine the back shape.

[0080] In step S500, the key point data is compared with a preset threshold, and the step of determining the back shape includes: if the key point data is greater than the preset threshold, the back shape is abnormal; if the key point data is less than or equal to the preset threshold, the back shape is normal.

[0081] In actual application scenarios, a preset threshold is set for each key point data according to the actual situation. When the value of the key point data exceeds the preset threshold, it is judged that the back morphology is abnormal and further diagnosis is required. When the key point data is less than or equal to the preset threshold, the back morphology is judged to be normal and no further diagnosis is required.

[0082] Reference Figure 3 As shown, the second aspect of the present invention provides a system for detecting adolescent scoliosis based on three-dimensional imaging, the system comprising: a data acquisition module 100, a calculation module 200 and a data processing module 300;

[0083] The data acquisition module 100 is configured to: acquire back point cloud data by three-dimensional structured light technology;

[0084] The computing module 200 is configured to: pre-process the back point cloud data to obtain target point cloud data; label the target point cloud data using the labelme tool to obtain human back data;

[0085] The data processing module 300 is configured to: input the human back data into the back segmentation model to obtain key point data; the key point data includes lumbar dimple height difference data, scapula height difference data, spinal midline data, scapula symmetry value data and lumbar dimple symmetry data; compare the key point data with a preset threshold to determine the back shape.

[0086] The effects of the above-mentioned system when applying the above-mentioned method can be found in the description of the above-mentioned method embodiment, which will not be repeated here.

[0087] The third aspect of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in the first aspect or any possible implementation of the first aspect is implemented.

[0088] A fourth aspect of the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect or any possible implementation of the first aspect.

[0089] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0090] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0091] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0092] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0093] Obviously, the above embodiments are merely examples for clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from them are still within the protection scope of the invention.

Claims

1. A method for detecting adolescent scoliosis based on three-dimensional imaging, characterized in that: The method comprises: Acquire back point cloud data through 3D structured light technology; Preprocessing the back point cloud data to obtain target point cloud data; The target point cloud data is labeled by using the labelme tool to obtain human back data; Inputting the human back data into a back segmentation model to obtain key point data; the key point data includes waist dimple height difference data, shoulder blade height difference data, spine midline data, shoulder blade symmetry value data and waist dimple symmetry data; The key point data is compared with a preset threshold to determine the back shape.

2. The method for detecting adolescent scoliosis based on three-dimensional imaging according to claim 1, characterized in that: The steps of obtaining back point cloud data through 3D structured light technology include: The human back is modeled with three-dimensional imaging based on the sinusoidal stripes and Gray code projection technology of DLP projection to obtain the back point cloud data.

3. The method for detecting adolescent scoliosis based on three-dimensional imaging according to claim 1, characterized in that: The step of preprocessing the back point cloud data to obtain target point cloud data includes: Denoising the back point cloud data to obtain denoised data; Projecting the denoised data onto a plane to obtain a triangular facet; Calculate the normal of the triangular face to obtain a normal map; The normal map is segmented to obtain target point cloud data.

4. The method for detecting adolescent scoliosis based on three-dimensional imaging according to claim 3, characterized in that: The step of projecting the denoised data onto a plane to obtain a triangular facet comprises: Obtaining pixels of the noise reduction data and counting them to obtain the number of rows and columns of a two-dimensional dot matrix; Form a quaternion according to the number of rows and the number of columns, and find a corresponding three-dimensional point in the three-dimensional point cloud set; Constructing a quadrilateral according to the four three-dimensional points; The quadrilateral is divided according to diagonals of a preset three-dimensional distance to obtain triangular facets.

5. The method for detecting adolescent scoliosis based on three-dimensional imaging according to claim 3, characterized in that: The step of calculating the normal of the triangular patch and obtaining a normal map comprises: Calculating the triangular face to obtain a face normal; Normalizing the surface normal to obtain normalized data; The normalized data is linearly mapped to a preset range according to the minimum and maximum values ​​of the normal grayscale value to obtain a normal map.

6. The method for detecting adolescent scoliosis based on three-dimensional imaging according to claim 1, characterized in that: The step of inputting the human back data into the back segmentation model to obtain key point data also includes: Inputting the human back data into the back segmentation model to obtain initial key point data; the initial key point data includes vertebral data, left lumbar fossa data, right lumbar fossa data, left scapula upper edge data, left scapula lower edge data, right scapula upper edge data, and right scapula lower edge data; Calculate the height difference data of the waist dimple points according to the left waist dimple data and the right waist dimple data; Calculate the scapula height difference data according to the left scapula lower edge data and the right scapula lower edge data; Dividing the back point cloud data into a plurality of slices; Fitting the point cloud data obtained by the slicing a preset number of times to obtain a curve; Derivative the curve to obtain extreme points; Calculate the center of gravity of the extreme point to obtain the midpoint of the spine; Fitting the midpoint of the spine to obtain the midline data of the spine; Calculating according to the spinal midline to obtain the scapula symmetry value data; The symmetry data of the lumbar dimples is obtained by performing calculation according to the left scapula lower edge data and the left lumbar dimple data.

7. The method for detecting adolescent scoliosis based on three-dimensional imaging according to claim 1, characterized in that: The step of comparing the key point data with a preset threshold value to determine the back shape comprises: If the key point data is greater than a preset threshold, the back morphology is abnormal; If the key point data is less than or equal to the preset threshold, the back shape is normal.

8. A 3D imaging-based adolescent scoliosis detection system, characterized in that: The system comprises: a data acquisition module, a calculation module and a data processing module; The data acquisition module is configured to: acquire back point cloud data through three-dimensional structured light technology; The computing module is configured to: pre-process the back point cloud data to obtain target point cloud data; label the target point cloud data using the labelme tool to obtain human back data; The data processing module is configured to: input the human back data into a back segmentation model to obtain key point data; the key point data includes lumbar dimple height difference data, scapula height difference data, spinal midline data, scapula symmetry value data and lumbar dimple symmetry data; compare the key point data with a preset threshold to determine the back shape.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of a method for detecting adolescent scoliosis based on three-dimensional imaging as described in any one of claims 1 to 7 are implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for detecting adolescent scoliosis based on three-dimensional imaging as described in any one of claims 1 to 7 are implemented.

Citation Information

Cited By

  • Spinal line generation method, spinal line generation device and electronic equipment

    CN120130998A

  • Teenager scoliosis detection method based on three-dimensional imaging

    CN121059145A