A method and terminal for intelligently detecting scoliosis abnormalities

Abstract

The application discloses a method and a terminal for intelligently detecting scoliosis abnormality, and comprises the following steps: S1, receiving a first image of a measured person shot by a front camera, wherein the first image is an orthographic view of bilateral upper limbs of the measured person in a standing body forward bending posture; and receiving a second image of the measured person shot by a bottom camera, wherein the second image is an overhead view of the bilateral upper limbs of the measured person in the standing body forward bending posture; S2, acquiring first deviation degree information of the bilateral upper limbs in a coronal plane according to the first image; and acquiring second deviation degree information of the bilateral upper limbs in a horizontal plane according to the second image; and S3, analyzing the severity of scoliosis and spinal rotation according to the collected deviation degree information, so that the scoliosis abnormality is easy to measure, and the automation degree of detection is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of measuring instruments, in particular to a method and a terminal for intelligently detecting scoliosis abnormalities. BACKGROUND

[0002] The gold standard for existing scoliosis detection is X-ray or CT imaging of the spine, which analyzes and calculates the degree of scoliosis, but this method has the problem of radioactive radiation, which is not conducive to large-scale population screening. The existing method suitable for preliminary screening of scoliosis detection instruments is to use the concave groove of the instrument to hold the back spine, and then use the level attached to the instrument to measure the degree of scoliosis. The instrument is pushed from top to bottom or from bottom to top along the spine, which is prone to measurement errors caused by the operator, and the operation is cumbersome, which is not conducive to improving the degree of automation of detection. SUMMARY

[0003] The technical problem to be solved by the present application is to provide a method and a terminal for intelligently detecting scoliosis abnormalities, which makes it easy to measure scoliosis abnormalities and improves the degree of automation of detection using artificial intelligence computer vision technology.

[0004] To solve the above technical problems, one technical solution adopted by the present application is:

[0005] A method for intelligently detecting scoliosis abnormalities, comprising the following steps:

[0006] S1, receiving a first image of a subject taken by a front camera, the first image being a front view of the subject's bilateral upper limbs in a standing forward bending posture; receiving a second image of the subject taken by a bottom camera, the second image being a top view of the subject's bilateral upper limbs in a standing forward bending posture;

[0007] S2, according to the first image, obtaining first deviation degree information of the bilateral upper limbs in the coronal plane; according to the second image, obtaining second deviation degree information of the bilateral upper limbs in the horizontal plane;

[0008] S3, analyzing the severity of scoliosis and spinal rotation according to the collected deviation degree information.

[0009] To solve the above technical problems, another technical solution adopted by the present application is:

[0010] A terminal for intelligently detecting scoliosis abnormalities, 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 to implement the steps of the above method for intelligently detecting scoliosis.

[0011] The application has the beneficial effects that: a method and terminal for intelligently detecting scoliosis are provided, two cameras are arranged to obtain images of bilateral upper limbs of a measured person, radiation problems caused by using X-ray or CT to shoot spinal image photos when analyzing scoliosis are avoided, operation errors introduced when manually operating a detection instrument to measure scoliosis are overcome, thus it is easy to measure scoliosis abnormalities, and the automation degree of detection is improved in combination with artificial intelligence computer vision technology. BRIEF DESCRIPTION OF DRAWINGS

[0012] Figure 1 A detection schematic diagram of the method and terminal for intelligently detecting scoliosis abnormalities according to the embodiment of the application;

[0013] Figure 2 A flowchart of the method for intelligently detecting scoliosis abnormalities according to the embodiment of the application;

[0014] Figure 3 A structural diagram of the terminal for intelligently detecting scoliosis abnormalities according to the embodiment of the application.

[0015] LABEL EXPLANATION:

[0016] 1, front camera; 2, bottom camera. DETAILED DESCRIPTION

[0017] To explain the technical content, purposes and effects of the application in detail, the following will be explained in combination with the embodiments and the drawings.

[0018] Please refer to Figure 1 The embodiment of the application provides a method for intelligently detecting scoliosis abnormalities, which comprises the following steps:

[0019] S1, receiving a first image of a measured person shot by a front camera, the first image being an orthographic view of bilateral upper limbs of the measured person in a standing body forward bending posture; receiving a second image of the measured person shot by a bottom camera, the second image being a top view of the bilateral upper limbs of the measured person in the standing body forward bending posture;

[0020] S2, according to the first image, obtaining first deviation degree information of the bilateral upper limbs in a coronal plane; according to the second image, obtaining second deviation degree information of the bilateral upper limbs in a horizontal plane;

[0021] S3, according to the collected deviation degree information, analyzing the severity of scoliosis and spinal rotation.

[0022] From the above description, the beneficial effects of the present application are that a method and terminal for intelligently detecting scoliosis are provided, two cameras are arranged to obtain images of the upper limbs of the measured person on both sides, thereby avoiding the radiation problem caused by using X-ray or CT to take the image of the spine, and overcoming the operation error introduced when measuring the scoliosis by manually operating the instrument, so as to easily measure the scoliosis abnormality, and combining with the artificial intelligence computer vision technology, the automation degree of detection is improved.

[0023] Further, before step S1, it further includes:

[0024] Receiving the double-sole images of the measured person standing on the bottom scanner scanned by the bottom scanner, and analyzing the symmetry of the double-arches according to the double-sole images;

[0025] If the double-arches are not symmetrical, the abnormality of the spine is mainly investigated.

[0026] From the above description, in order to avoid the additional deviation introduced by the incorrect standing position of the measured person's feet, the foot position needs to be corrected, the symmetry related data of the arches is obtained through the bottom scanner, more comprehensive analysis data is provided for analyzing the scoliosis, the accuracy of analyzing the severity of the scoliosis and the rotation of the spine is improved, and for the case that the double-arches are not balanced, the abnormality of the spine is mainly investigated.

[0027] Further, the deviation degree information of the double upper limbs in step S2 includes the positions of the double fists, the double palms, the double wrists, the same joint positions of the double hands, the double elbow joint positions, and the double shoulder joint positions, and also includes the positions of the double upper limbs when the measured person grabs the heavy object in the standing forward bending posture.

[0028] From the above description, the positions of the multiple joints of the double upper limbs are detected, more accurate deviation degree measurement data is provided, and the severity of the scoliosis and the rotation of the spine is analyzed.

[0029] Further, the first deviation degree information of the double upper limbs in the coronal plane obtained according to the first image in step S2 is specifically:

[0030] Performing posture estimation or gesture estimation on the first image to estimate the first double-wrist position of the measured person;

[0031] Calculating the first inclination angle of the double wrists according to the connecting line of the first double-wrist position and the coronal axis, and obtaining the first deviation degree of the upper limbs through the first inclination angle;

[0032] The second deviation degree information of the double upper limbs in the horizontal plane obtained according to the second image is specifically:

[0033] performing pose estimation or gesture estimation on the second image to obtain a second double-wrist position of the subject;

[0034] performing target detection according to the second double-wrist position and extracting a second double-wrist image from the second image through image segmentation;

[0035] calculating a second tilt angle of the double wrist according to a connecting line of the second double-wrist image and a horizontal axis, and obtaining a second deviation degree of the upper limb through the second tilt angle.

[0036] It can be seen from the above description that the detection automation degree is improved by combining artificial intelligence computer vision technology.

[0037] Further, the analysis of the severity of scoliosis and spinal rotation according to the collected deviation degree information includes:

[0038] If the first tilt angle is greater than a first threshold or the second tilt angle is greater than a second threshold, it is determined that the subject has scoliosis or spinal rotation, otherwise, it is determined that the subject has normal spine.

[0039] It can be seen from the above description that the angle threshold corresponding to the disease is set, and the deviation degree information is compared, so that suspicious cases can be screened out, and X-ray or CT method is used for further diagnosis.

[0040] Please refer to Figure 3 Another embodiment of the present application provides a terminal for intelligently detecting scoliosis, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the following steps when executing the computer program:

[0041] S1, receiving a first image of a subject taken by a front camera, the first image being a front view of the subject's bilateral upper limbs in a standing forward bending posture; receiving a second image of the subject taken by a bottom camera, the second image being a top view of the subject's bilateral upper limbs in a standing forward bending posture;

[0042] S2, obtaining first deviation degree information of the bilateral upper limbs in the coronal plane according to the first image; and obtaining second deviation degree information of the bilateral upper limbs in the horizontal plane according to the second image;

[0043] S3, analyzing the severity of scoliosis and spinal rotation according to the collected deviation degree information.

[0044] From the above description, the beneficial effects of the present application are that a method and terminal for intelligently detecting scoliosis are provided, two cameras are arranged to obtain images of the upper limbs of the measured person on both sides, thereby avoiding the radiation problem caused by using X-ray or CT to take the image of the spine, and overcoming the operation error introduced when measuring the scoliosis by manually operating the instrument, so as to easily measure the scoliosis abnormality, and combining with the artificial intelligence computer vision technology, the automation degree of detection is improved.

[0045] Further, before step S1, it further includes:

[0046] Receiving the double-sole images of the measured person standing on the bottom scanner scanned by the bottom scanner, and analyzing the symmetry of the double-arches according to the double-sole images;

[0047] If the double-arches are not symmetrical, the abnormality of the spine is mainly investigated.

[0048] From the above description, in order to avoid the additional deviation introduced by the incorrect standing position of the measured person's feet, the foot position needs to be corrected, the symmetry related data of the arches is obtained through the bottom scanner, more comprehensive analysis data is provided for analyzing the scoliosis, the accuracy of analyzing the severity of the scoliosis and the rotation of the spine is improved, and for the case that the double-arches are not balanced, the abnormality of the spine is mainly investigated.

[0049] Further, the deviation degree information of the double upper limbs in step S2 includes the positions of the double fists, the double palms, the double wrists, the same joint positions of the double hands, the double elbow joint positions, and the double shoulder joint positions, and also includes the positions of the double upper limbs when the measured person grabs the heavy object in the standing forward bending posture.

[0050] From the above description, the positions of the multiple joints of the double upper limbs are detected, more accurate deviation degree measurement data is provided, and the severity of the scoliosis and the rotation of the spine is analyzed.

[0051] Further, the first deviation degree information of the double upper limbs in the coronal plane obtained according to the first image in step S2 is specifically:

[0052] The first double-wrist position of the measured person is estimated by performing posture estimation or gesture estimation on the first image;

[0053] The first inclination angle of the double wrists is calculated according to the connecting line of the first double-wrist position and the coronal axis, and the first deviation degree of the upper limbs is obtained through the first inclination angle;

[0054] The second deviation degree information of the double upper limbs in the horizontal plane obtained according to the second image is specifically:

[0055] performing pose estimation or gesture estimation on the second image to obtain a second double-wrist position of the subject;

[0056] performing target detection according to the second double-wrist position and extracting a detected second double-wrist image through image segmentation;

[0057] calculating a second tilt angle of the double wrist according to a connecting line of the second double-wrist image and a horizontal axis, and obtaining a second deviation degree of the upper limb through the second tilt angle.

[0058] It can be known from the above description that the detection automation degree is improved by combining artificial intelligence computer vision technology.

[0059] Further, the analysis of the severity of scoliosis and spinal rotation according to the collected deviation degree information includes:

[0060] If the first tilt angle is greater than a first threshold or the second tilt angle is greater than a second threshold, it is determined that the subject has scoliosis or spinal rotation, otherwise, it is determined that the subject has normal spine.

[0061] It can be known from the above description that the angle threshold corresponding to the disease is set, and the deviation degree information is compared, so that suspicious cases can be screened out, and X-ray or CT method is used for further diagnosis.

[0062] The above-mentioned method and terminal for intelligently detecting scoliosis can easily measure scoliosis, and improve the detection automation degree. The following will be described through specific embodiments:

[0063] Embodiment one

[0064] Please refer to Figure 1 and Figure 2 A method for intelligently detecting scoliosis includes the following steps:

[0065] S1, receiving a first image of a subject taken by a front camera 1, the first image being a front view of the subject's bilateral upper limbs in a standing body forward bending posture; receiving a second image of the subject taken by a bottom camera 2, the second image being a top view of the subject's bilateral upper limbs in a standing body forward bending posture; preferably, the subject is measured not only in the relaxed state of the upper limbs, but also in the standing body forward bending posture, and the bilateral hands are measured by grabbing an equal weight object.

[0066] Preferably, to avoid the introduction of additional deviation caused by the incorrect standing position of the subject, the standing position of the feet needs to be corrected. The image of the feet of the subject standing on the bottom scanner is received by the bottom scanner, the symmetry of the arches of the feet is analyzed according to the image of the feet, and the possibility of the abnormality of the spine is judged according to the symmetry of the arches of the feet. For the case of imbalance of the arches of the feet, the abnormality of the spine needs to be focused on. And the subject is required to keep the sides of the feet and the heels close to the fixed bars, the arms are naturally drooping, the hands are clenched, the dumbbells are held tightly, and the fists are opposite to each other.

[0067] In this embodiment, the shooting direction of the front camera 1 is horizontal, the shooting direction of the bottom camera 2 is vertically upward, and the shooting ranges of the two cameras intersect; wherein the bottom camera 2 is a backlit camera, which can reduce the influence of light or sunlight during shooting.

[0068] S2, according to the first image shot by the front camera 1, the first deviation degree information of the bilateral upper limbs in the coronal plane is obtained; according to the second image shot by the bottom camera 2, the second deviation degree information of the bilateral upper limbs in the horizontal plane is obtained;

[0069] In step S2, the first deviation degree information of the bilateral upper limbs in the coronal plane is obtained according to the first image, which is specifically: posture estimation or gesture estimation is performed on the first image, and the first wrist position of the subject is estimated; the first inclination angle of the wrists is calculated according to the connecting line of the first wrist position and the coronal axis, and the first deviation degree of the upper limbs is obtained through the first inclination angle;

[0070] According to the second image, the second deviation degree information of the bilateral upper limbs in the horizontal plane is obtained, which is specifically: posture estimation or gesture estimation is performed on the second image, and the second wrist position of the subject is estimated; target detection is performed according to the second wrist position, and the detected second wrist image is extracted through image segmentation; the second inclination angle of the wrists is calculated according to the connecting line of the second wrist image and the horizontal axis, and the second deviation degree of the upper limbs is obtained through the second inclination angle.

[0071] In which, the deviation degree information includes the positions of the fists, the palms, the wrists, the same joint positions of the hands, the elbow joint positions, the shoulder joint positions, also includes the positions of the objects grasped by the hands, and also includes the positions of the bilateral upper limbs when the subject grasps the heavy objects in the standing forward bending posture.

[0072] S3, according to the collected deviation degree information, the severity of the scoliosis and the rotation of the spine is analyzed.

[0073] That is, according to the deviation degree information of the bilateral upper limbs, the positions of the bilateral upper limbs are obtained, the deviation angle of the line connecting the positions of the bilateral upper limbs is calculated by referring to the coronal axis, and whether the deviation angle reaches an angle threshold is calculated according to the deviation angle obtained in the first image and the deviation angle obtained in the second image.

[0074] Wherein, according to the collected deviation degree information, the severity of scoliosis and spinal rotation is analyzed, including: if the first inclination angle is greater than the first threshold or the second inclination angle is greater than the second threshold, it is determined that the subject has scoliosis or spinal rotation, otherwise, it is determined that the subject has normal spine.

[0075] Specifically, for the video frame images obtained by the two orientation cameras, a pose estimation algorithm or a gesture estimation algorithm of artificial intelligence computer vision is applied respectively, for example, a pose estimation or gesture estimation algorithm of OpenPose, AlphaPose or Mediapipe is used to obtain the positions of the wrists of the two hands (or the positions of the index joints of the two hands), and the inclination angle of the line connecting the positions of the wrists of the two hands is automatically calculated, which represents the deviation degree of the bilateral upper limbs, and the severity of scoliosis and spinal rotation is further analyzed. For example, if the angle of the line connecting the index joints of the two hands in the front image exceeds the threshold of 20°, or the deviation degree of the angle of the line connecting the index joints of the two hands in the image taken from below exceeds the threshold of 20°, it is considered that a suspicious case is found, and X-ray or CT is needed for further diagnosis.

[0076] Embodiment two

[0077] Please refer to Figure 3 An intelligent terminal for detecting scoliosis, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps in embodiment one when executing the computer program.

[0078] In summary, the present application provides an intelligent method and terminal for detecting scoliosis, which measures the deviation degree of the bilateral upper limbs of the subject in a standing position by using double cameras combined with artificial intelligence computer vision technology, analyzes the severity of scoliosis and spinal rotation according to the deviation degree, and thus makes it easy to measure scoliosis abnormalities and improves the degree of automation of detection.

[0079] It is to be understood that the above described arrangements are merely illustrative of the many possible embodiments of the present application. Numerous modifications and alterations thereto will become apparent to those skilled in the art without departing from the scope of the present application and it is intended that the scope of the present application should be determined by reference to the claims.

[0080] The embodiments described above are merely illustrative of one possible embodiment of the present application and are not intended to limit the scope of the patent. The illustrative embodiments described in the detailed description, drawings and claims are not intended to limit. Other embodiments can be employed without departing from the spirit or scope of the application. Various aspects of the content of the present application are configured in a variety of different configurations, substitutions, combinations, designs, and the like, which are still part of the content of the application.

[0081] The above is only an embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent method or equivalent process transformation using the content of the specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for intelligently detecting scoliosis, characterized in that, Includes the following steps: S1. Receive a first image of the subject captured by the front camera, the first image being a frontal view of the subject's bilateral upper limbs in a standing forward flexion posture; receive a second image of the subject captured by the bottom camera, the second image being a back view of the subject's bilateral upper limbs in a standing forward flexion posture. S2. Based on the first image, obtain the first deviation information of both upper limbs in the coronal plane; Based on the second image, information on the degree of deviation of both upper limbs in the horizontal plane is obtained; S3. Based on the collected deviation information, analyze the severity of scoliosis and spinal rotation.

2. The method for intelligent detection of scoliosis abnormalities according to claim 1, characterized in that, Step S1 is preceded by: Receive images of the bottom of the subject's feet when standing on the bottom scanner, and analyze the symmetry of the arches of the feet based on the bottom images. If the arches of the feet are asymmetrical, then spinal abnormalities should be the primary focus of investigation.

3. The method for intelligent detection of scoliosis abnormalities according to claim 1, characterized in that, The deviation information of the bilateral upper limbs mentioned in step S2 includes the position of the fists, the position of the palms, the position of the wrists, the position of the same joints of both hands, the position of the elbows, the position of the shoulders, and also the position of the bilateral upper limbs when the subject grasps an object of equal weight with both hands in a standing forward bending posture.

4. The method for intelligent detection of scoliosis abnormalities according to claim 1, characterized in that, Step S2, which involves obtaining the first deviation information of both upper limbs in the coronal plane based on the first image, specifically includes: Pose estimation or gesture estimation is performed on the first image to estimate the position of the subject's first two wrists; The first tilt angle of the two wrists is calculated based on the connecting line of the first two wrist positions and the coronal axis, and the first deviation of the upper limb is obtained through the first tilt angle. Based on the second image, the information on the degree of deviation of both upper limbs in the horizontal plane is obtained as follows: Pose estimation or gesture estimation is performed on the second image to estimate the position of the subject's second wrists; The second tilt angle of the wrists is calculated based on the connecting line of the second wrist position and the horizontal axis, and the second deviation of the upper limb is obtained through the second tilt angle.

5. The method for intelligent detection of scoliosis abnormalities according to claim 4, characterized in that, The analysis of the severity of scoliosis and spinal rotation based on the collected deviation information includes: If the first tilt angle is greater than the first threshold or the second tilt angle is greater than the second threshold, the subject is determined to have scoliosis or spinal rotation; otherwise, the subject's spine is determined to be normal.

6. A terminal for intelligent detection of scoliosis, 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 computer program, it performs the following steps: S1. Receive a first image of the subject captured by the front camera, the first image being a frontal view of the subject's bilateral upper limbs in a standing forward flexion posture; receive a second image of the subject captured by the bottom camera, the second image being a back view of the subject's bilateral upper limbs in a standing forward flexion posture. S2. Based on the first image, obtain the first deviation information of both upper limbs in the coronal plane; Based on the second image, information on the degree of deviation of both upper limbs in the horizontal plane is obtained; S3. Based on the collected deviation information, analyze the severity of scoliosis and spinal rotation.

7. A terminal for intelligent detection of scoliosis abnormalities according to claim 6, characterized in that, The procedure preceding step S1 also includes: Receive images of the bottom of the subject's feet when standing on the bottom scanner, and analyze the symmetry of the arches of the feet based on the bottom images. If the arches of the feet are asymmetrical, then spinal abnormalities should be the primary focus of investigation.

8. A terminal for intelligent detection of scoliosis abnormalities according to claim 6, characterized in that, The deviation information of the bilateral upper limbs mentioned in step S2 includes the position of the fists, the position of the palms, the position of the wrists, the position of the same joints of both hands, the position of the elbows, the position of the shoulders, and also the position of the bilateral upper limbs when the subject grasps an object of equal weight with both hands in a standing forward bending posture.

9. A terminal for intelligent detection of scoliosis abnormalities according to claim 6, characterized in that, Step S2, which involves obtaining the first deviation information of both upper limbs in the coronal plane based on the first image, specifically includes: Pose estimation or gesture estimation is performed on the first image to estimate the position of the subject's first two wrists; The first tilt angle of the two wrists is calculated based on the connecting line of the first two wrist positions and the coronal axis, and the first deviation of the upper limb is obtained through the first tilt angle. Based on the second image, the information on the degree of deviation of both upper limbs in the horizontal plane is obtained as follows: Pose estimation or gesture estimation is performed on the second image to estimate the position of the subject's second wrists; The second tilt angle of the wrists is calculated based on the connecting line of the second wrist position and the horizontal axis, and the second deviation of the upper limb is obtained through the second tilt angle.

10. A terminal for intelligent detection of scoliosis abnormalities according to claim 9, characterized in that, The analysis of the severity of scoliosis and spinal rotation based on the collected deviation information includes: If the first tilt angle is greater than the first threshold or the second tilt angle is greater than the second threshold, the subject is determined to have scoliosis or spinal rotation; otherwise, the subject's spine is determined to be normal.

Images