Scoliosis detection method based on piecewise polynomial fitting

Abstract

The invention relates to a scoliosis detection method based on piecewise polynomial fitting, and belongs to the technical field of image processing, and the method comprises the following steps: A, obtaining three-dimensional point cloud data of the back of a human body through a TOF camera, and carrying out the preprocessing of the three-dimensional point cloud data; b, performing voxelization and down-sampling processing on the point cloud data obtained in the step A; c, obtaining a spine point area according to the processed point cloud data, and converting the point cloud data into a depth image; d, after the spine point area is obtained, a certain threshold value is set, point clouds are grouped according to rows, the point with the minimum depth value in each row is the spine point of the row, segmented polynomial fitting processing is carried out on the spine according to a key point model trained by the depth image, and a spine curve is obtained; and E, finding out a point with the maximum positive curvature and the maximum negative curvature in the spine curve, and determining whether the spine is scoliosis or not by using an included angle between the two points as an estimated cobb angle.

Description

Technical Field

[0001] The invention belongs to the technical field of image processing and relates to a scoliosis detection method based on piecewise polynomial fitting. Background Art

[0002] Scoliosis is a relatively complex three-dimensional spinal deformity that can have a very serious impact on human health. At present, the commonly used scoliosis screening methods include the method of calculating the cobb angle through X-rays and the method of measuring the correlation between the human body surface morphology and asymmetry and the cobb angle for judgment and screening. However, because manual detection is inefficient and prone to errors, and X-rays can cause certain harm to the human body, non-invasive machine vision detection methods are constantly developing.

[0003] The extraction of spinal points and the fitting of spinal curves are crucial in this method. The traditional scoliosis examination method requires the examinee to take off his outer clothes for a visual physical examination by a doctor, and then use a moiré camera for comparison. Finally, those who are diagnosed with scoliosis in the above two stages undergo X-ray examination. This process and method are inefficient, prone to errors, and costly. X-rays are harmful to the human body. Existing non-invasive tests are expensive. Summary of the invention

[0004] In view of this, the object of the present invention is to provide a non-invasive automatic scoliosis detection method. The method collects three-dimensional point cloud data of the back of the human body through a TOF camera and extracts spine points, fits the spine points, and calculates the cobb angle through the spine curve to make a judgment.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] A scoliosis detection method based on piecewise polynomial fitting comprises the following steps:

[0007] Step A: Use a TOF camera to obtain the 3D point cloud data of the human back and perform preprocessing;

[0008] Step B: voxelize and downsample the point cloud data obtained in step A;

[0009] Step C: Obtain the spine point area from the processed point cloud data, and convert the point cloud data into a depth image;

[0010] Step D: After obtaining the spine point area, set a certain threshold to group the point cloud by row. The point with the smallest depth value in each row is the spine point of that row. Then, perform piecewise polynomial fitting on the spine according to the key point model trained with the depth image to obtain the spine curve.

[0011] Step E: Find the point where the positive and negative curvatures of the spine are the largest. The angle between the two points is the estimated Cobb angle, which can be used to determine whether it is scoliosis.

[0012] Furthermore, the preprocessing in step A is to filter out background parts that are not related to the human body.

[0013] Furthermore, the step C of converting the point cloud data into a depth image specifically includes: normalizing the depth values ​​of the point cloud data and converting them into a depth image of 640*576 pixels.

[0014] Furthermore, setting a certain threshold to group the point cloud by row in step D specifically includes: traversing all points, and if the difference between the vertical coordinates of two points is less than the threshold, they are identified as points in the same row.

[0015] Furthermore, the step D performs piecewise polynomial fitting processing on the spine according to the key point model trained with the depth image to obtain the spine curve, specifically comprising the following steps:

[0016] D1: Let the spine point set be P, x be the horizontal coordinate, y be the vertical coordinate, and the point set containing n points can be expressed as:

[0017]

[0018] D2: For these n data points, sort them first; for a conventional piecewise linear function, the segmentation point is x b ={p1,p2,p3,…,p n}, whose function set is:

[0019]

[0020] Its matrix form is:

[0021]

[0022] Among them, f is the marking function, which is used to mark whether the point is in the segment, expressed as:

[0023]

[0024] D3: The number of segmentation points is 3, the polynomial order is 3, and the linear segmentation fitting is extended to the segmentation 3rd order polynomial fitting, and the matrix form is:

[0025]

[0026] Let A be the regression matrix, β be the parameter vector, and y be the data vector, then the above formula can be expressed as:

[0027] Aβ=y

[0028] After solving this equation, a continuous piecewise fitting curve is obtained, and its solution is expressed as:

[0029] β=(A T A) -1 A T y.

[0030] Furthermore, in step E, the cobb angle is estimated, and when the cobb angle is greater than 5°, scoliosis is determined.

[0031] The beneficial effects of the present invention are as follows: based on conventional spinal curve polynomial fitting, the present invention performs segmented fitting and judgment of the spinal curve according to the spinal structure and characteristics, thereby improving detection accuracy, performing monitoring and detection through computer image processing methods, and reducing detection costs.

[0032] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below in conjunction with the accompanying drawings, wherein:

[0034] Figure 1 The flowchart of the spine measurement method based on piecewise polynomial fitting is shown in FIG. DETAILED DESCRIPTION

[0035] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0036] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0037] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present invention difficult to understand.

[0038] like Figure 1 As shown, the present invention provides a spinal column measurement and detection method based on piecewise polynomial fitting, comprising the following steps:

[0039] Step A: Use a TOF camera to obtain the 3D point cloud data of the human back and perform preprocessing.

[0040] The three-dimensional point cloud data of the human back is obtained, and the point cloud data is preprocessed to filter out the background parts irrelevant to the human body.

[0041] Step B: voxelize and downsample the point cloud data obtained in step A.

[0042] Step C: Obtain the spine point area from the processed point cloud data and convert the point cloud data into a depth image

[0043] Step C uses the point cloud data processed in step B to extract the area where the spine point is located, and normalizes the depth value of the point cloud data to convert it into a 640*576 pixel depth image.

[0044] Step D: After obtaining the spine point area, set a certain threshold and traverse all points. If the difference between the ordinates of two points is less than the threshold, they are identified as points in the same row. The lowest point found in each row is the spine point. Then, according to the key point model trained with the depth image, the spine is subjected to piecewise polynomial fitting to obtain the spine curve. The specific steps are as follows:

[0045] (1) Let the spine point set be P, x be the horizontal coordinate, y be the vertical coordinate, and the point set containing n points can be expressed as:

[0046]

[0047] (2) For these n data points, sort them first. For a conventional piecewise linear function, the segmentation points are , and its function set is:

[0048]

[0049] Its matrix form is:

[0050]

[0051] Among them, f is the marking function, which is used to mark whether the point is in the segment, which can be expressed as:

[0052]

[0053] (3) The number of segmentation points in the present invention is 3, and the polynomial order is 3. The linear segmentation fitting is extended to the segmentation 3rd order polynomial fitting, and the matrix form can be obtained as follows:

[0054]

[0055] (4) Let A be the regression matrix, β be the parameter vector, and y be the data vector. Then the original formula can be expressed as:

[0056] Aβ=y

[0057] (5) After solving this equation, a continuous piecewise fitting curve can be obtained, and its solution can be expressed as:

[0058] β=(A T A) -1 A T y

[0059] Step E: Find the point where the positive curvature and the negative curvature of the spine curve are the largest. The angle between the two points is the estimated Cobb angle. In this embodiment, when the Cobb angle is greater than 5°, it can be determined as scoliosis.

[0060] In the above embodiments, the description's reference to "this embodiment" indicates that a particular feature, structure, or characteristic described in conjunction with the embodiment is included in at least some embodiments, but not necessarily all embodiments. Multiple occurrences of "this embodiment" do not necessarily all refer to the same embodiment.

[0061] In the above-described embodiments, although the invention has been described in conjunction with specific embodiments of the invention, many substitutions, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art based on the foregoing description. For example, other storage structures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed. Embodiments of the invention are intended to encompass all such substitutions, modifications, and variations that fall within the broad scope of the appended claims.

[0062] This embodiment further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, any one of the methods in this embodiment is implemented.

[0063] This embodiment also provides an electronic terminal, including: a processor and a memory;

[0064] The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal executes any one of the methods in this embodiment.

[0065] The computer-readable storage medium in this embodiment can be understood by ordinary technicians in this field: all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to the computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other media that can store program codes.

[0066] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication with each other. The memory is used to store computer programs, the communication interface is used to communicate, and the processor and the transceiver are used to run computer programs so that the electronic terminal executes each step of the above method.

[0067] In this embodiment, the memory may include a random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0068] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0069] The present invention can be used in many general or special computing system environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like.

[0070] The present invention may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.

Claims

1. A scoliosis detection method based on piecewise polynomial fitting, characterized in that: The following steps are involved: Step A: Use a TOF camera to obtain the 3D point cloud data of the human back and perform preprocessing; Step B: voxelize and downsample the point cloud data obtained in step A; Step C: Obtain the spine point area from the processed point cloud data, and convert the point cloud data into a depth image; Step D: After obtaining the spine point area, set a certain threshold to group the point cloud by row. The point with the smallest depth value in each row is the spine point of that row. Then, perform piecewise polynomial fitting on the spine according to the key point model trained with the depth image to obtain the spine curve. Step E: Find the point in the spinal curve where the positive and negative curvatures are the largest. The angle between the two points is the estimated Cobb angle, which can be used to determine whether it is scoliosis.

2. The scoliosis detection method based on piecewise polynomial fitting according to claim 1, characterized in that: The preprocessing described in step A is to filter out background parts that are irrelevant to the human body.

3. The scoliosis detection method based on piecewise polynomial fitting according to claim 1, characterized in that: The step C of converting the point cloud data into a depth image specifically includes: normalizing the depth values ​​of the point cloud data and converting them into a depth image of 640*576 pixels.

4. The scoliosis detection method based on piecewise polynomial fitting according to claim 1, characterized in that: The step D of setting a certain threshold to group the point cloud by row specifically includes: traversing all points, and if the difference between the vertical coordinates of two points is less than the threshold, they are identified as points in the same row.

5. The scoliosis detection method based on piecewise polynomial fitting according to claim 1, characterized in that: Step D performs piecewise polynomial fitting processing on the spine according to the key point model trained with the depth image to obtain the spine curve, which specifically includes the following steps: D1: Let the spine point set be P, x be the horizontal coordinate, y be the vertical coordinate, and the point set containing n points can be expressed as: D2: For these n data points, sort them first; for a conventional piecewise linear function, the segmentation point is x b ={p1,p2,p3,…,p n }, whose function set is: Its matrix form is: Among them, f is the marking function, which is used to mark whether the point is in the segment, expressed as: D3: The number of segmentation points is 3, the polynomial order is 3, and the linear segmentation fitting is extended to the segmentation 3rd order polynomial fitting, and the matrix form is: Let A be the regression matrix, β be the parameter vector, and y be the data vector, then the above formula can be expressed as: Aβ=y After solving this equation, a continuous piecewise fitting curve is obtained, and its solution is expressed as: β=(A T A) -1 A T Yes.

6. The scoliosis detection method based on piecewise polynomial fitting according to claim 1, characterized in that: In step E, the cobb angle is estimated, and when the cobb angle is greater than 5°, scoliosis is determined.

Images